WHEN SEARCH MEETS

WEB USABILITY

SHARI THUROW AND NICK MUSICA

When Search Meets Web Usability
Shari Thurow and Nick Musica

New Riders
1249 Eighth Street
Berkeley, CA 94710
510/524-2178
510/524-2221 (fax)

Find us on the web at www.newriders.com

To report errors, please send a note to [email protected]

New Riders is an imprint of Peachpit Press, a division of Pearson Education

Copyright © 2009 by Shari Thurow and Nick Musica

Project Editor: Nancy Peterson

Development Editor: Robyn G. Thomas
Technical Editors: Anne Kennedy and John Sorflaten
Production Coordinator: Myrna Vladic
Copy Editor: Doug Adrianson
Compositor: David Van Ness
Marketing Manager: Glenn Bisignani
Indexer: Valerie Perry
Cover and interior design: Mimi Heft
Cover photo: Francois Roux
Cover production: Mike Tanamachi

Notice of Rights
All rights reserved. No part of this book may be reproduced or transmitted in any form by
any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior
written permission of the publisher. For information on getting permission for reprints and
excerpts, contact [email protected]. Illustrations courtesy of ToonClipart.com © 2009.
All rights reserved.

Notice of Liability
The information in this book is distributed on an “As Is” basis without warranty. While every
precaution has been taken in the preparation of the book, neither the authors nor Peachpit
shall have any liability to any person or entity with respect to any loss or damage caused or
alleged to be caused directly or indirectly by the instructions contained in this book or by the
computer software and hardware products described in it.

Trademarks
Many of the designations used by manufacturers and sellers to distinguish their products are
claimed as trademarks. Where those designations appear in this book, and Peachpit was aware
of a trademark claim, the designations appear as requested by the owner of the trademark.
All other product names and services identified throughout this book are used in editorial
fashion only and for the benefit of such companies with no intention of infringement of the
trademark. No such use, or the use of any trade name, is intended to convey endorsement or
other affiliation with this book.

ISBN 13: 978-0-321-60589-4

ISBN 10: 0-321-60589-6

9 8 7 6 5 4 3 2 1

Printed and bound in the United States of America

Dedications
To our parents:
Dale and Edwin Thurow
Susan and Nicholas Musica

And in loving memory:

Big Jim Schultz
Donald Londrie
Jack Wendt
Tierney Smutny
This page intentionally left blank
 V

! ABOUT THE AUTHORS

Shari Thurow is the founder and Search Engine Optimization (SEO)
Director for Omni Marketing Interactive, a full-service search engine
optimization, web design, and website usability firm. As the leading
expert on search-engine-friendly web design, Shari lectures on this topic
and works for many firms worldwide.
She is also the author of Search Engine Visibility (New Riders, 2007),
which teaches web developers how to build a site that pleases both site
visitors and the commercial web search engines. Search Engine Visibility
has been translated into French, Japanese, Korean, and Polish.
Clients include Microsoft, AOL, Yahoo, National Cancer Institute,
HSBC, Expedia, WebMD and MedicineNet.com, ABC News, Wharton
School of the University of Pennsylvania, and Dow Corning.
Additionally, Shari has designed and successfully marketed websites
for businesses in fields such as medicine, science, biotechnology, account-
ing and finance, computers and software, manufacturing, real estate, law,
e-commerce and online stores, art and design, marketing, insurance,
employment, education, and travel.
Shari has been featured in many publications, including the New
York Times, USA Today, Wired Magazine, Entrepreneur, Internet Retailer,
and Crain’s Chicago Business.
Shari received her B.A. in Genetics and Developmental Biology and
an M.A. in Asian Studies/Japanese from University of Illinois. When she
is not busy being a search geek, Shari is a die-hard Metallica fan, loves
traveling to Iceland, and is not convinced that surfboards make great
wall art. She resides in Elgin, Illinois.

Nick Musica has over 10 years experience designing, marketing, and

managing websites. He’s the founder of Think Green Media, an Internet
marketing firm that offers search engine optimization, search marketing,
web design, and web usability services.
He holds a B.A. in Communication with a concentration in Public
Relations and Journalism from Monmouth University and an M.A. in
Graphic Communications Management and Technology from New York
University (NYU), where he lecturers on web design, search engine market-
ing, and web usability. He’s also a Certified Usability Analyst (CUA).
Nick has designed and marketed websites for businesses in the
fields of alternative energy, cosmetics, education, healthcare, finance,
the nonprofit sector, retail, technology, and telecom. He’s also a speaker
to marketers and business owners on the topics of web design, search
marketing, web usability, writing for the web, and search usability.
VI

When he isn’t working on search usability, you can find Nick surfing
in the Atlantic Ocean or playing mediocre renditions of ’80s and ’90s
songs on his ukulele. For those wondering: Yes, there is surf in New
Jersey, yes, surfing in mid-February can sometimes get just a little cold,
and yes, surfboards make great wall art.

! CONTRIBUTORS

Technical Editors
Anne F. Kennedy is a managing partner and founder of Beyond Ink.
With nearly 40 years’ experience in marketing and public relations,
Anne founded Beyond Ink in 1997 to bring the fundamental principles of
marketing communications to online media. A search engine marketer
for more than 10 years, she is an industry thought leader, sought-after
speaker, and writer worldwide.
Currently, Anne represents professional SEO firms on the Search
Engine Strategies (SES) Domestic Advisory Board, an influential group
of business professionals from various industries and fields of expertise
selected to propel the search industry forward.
In 2001 she launched SEOnews.net to foster better understanding of
the emerging search marketing platform and opportunities to a growing
market of potential customers.
Beyond Ink uses ethical principles of effective marketing commu-
nications to achieve critical third-party endorsements in popular, highly
visible portals such as Google, MSN, Yahoo, and other top search media
for clients, which include large household name brands as well as many
small businesses, nonprofits, and online start-ups.
Anne resides in Maine and Oregon, but can always be found at
Beyondink.com.

John Sorflaten founded Optimum Performance Training in 1985 after

filing several patent applications on a Japanese word processing input
system. His design recommendations dramatically reduced the keyboard
learning process through user-centered design. He then moved into
designing usable interfaces as a response to his question “Why train,
when you can design it to be usable right away?”
John accumulated 20 years of consulting and teaching experience
with the largest worldwide user-interface consulting firm, Human
Factors International (HFI). He taught hundreds of courses supporting
 VII

certification as a Usability Analyst since the inception of the program at

HFI (www.humanfactors.com).
Since 1992, John has been a Certified Professional Ergonomist, the
top professional certification for usability. He fills demands as a public
speaker, writer, and innovator in social-web and rich interface applications.
His redesign of www.ninds.nih.gov won a Best of the Fed award.
With a degree in Cinema from the University of Southern California
and a Ph.D. in Instructional Design, John provides clients with seasoned
advice and user research. This specialization ensures effective integration
of video, text, images, and brand moxie for websites.

Cover Photography
Francois Roux focuses his photography on nature, particularly in Central
Park in New York City. Living in New York has provided him the unique
opportunity to photograph natural settings and then to step immedi-
ately into an urban landscape of unique and historic architecture. His
pictures of France have often focused on black and white images of the
mountains of Hautes Alpes, the region of southeastern France where
he was raised.
When Shari and Nick asked Francois to contribute a cover photo-
graph for this book, his inclination was to use an architectural image to
suggest the convergence of search and usability. What better place to
find such an image than at the Rose Center for Earth and Space at New
York’s American Museum of Natural History? The building’s contained
sphere, itself containing the planetarium, is surfaced in a mesh-like
pattern of intersecting rays and concentric circles. This visual pattern
lends itself to two-dimensional representation while still suggesting
three-dimensional structure. You can see more of his photography at:
francois-roux-photography.com.

Cartoons
Ron Leishman is an illustrator/cartoonist and owner of Toonaday.com
and ToonClipart.com.
Ron has worked as an illustrator for over 20 years. His work has
appeared in greeting cards, magazines, and newspapers, where he worked
as an editorial cartoonist for a number of years. Along the way he has
created a number of logos for new companies.
Earlier in his career, Ron also was co-creator of Captain Canuck,
who was published in comic book form for a number of years in the
’70s and ’80s.
VIII

With the advent of the Internet, Ron moved to the web in 1996 with
his site Toonaday.com, for which he has drawn a daily cartoon for the
past 13 years and to which he continues to contribute on a daily basis.
Recently, he has expanded his offerings with another website called
ToonClipart.com.
Ron is a resident of Calgary, Alberta, and can be contacted at
[email protected].

! ACKNOWLEDGMENTS
This book took years of testing and experience to create. It also took the
knowledge and support of some very important people.
Thank you to those directly involved in the creation of this book.
To the staff at Peachpit Press—Nancy Peterson and Robyn Thomas for
managing this project, for being great sounding boards, and for giving
gentle nudges to keep things going. A special thanks goes to Michael Nolan
and Doug Adrianson for continuing to work with us on multiple book
projects. Thank you to the production and marketing staff at Peachpit
for helping to shape this book and getting it out into the world.
We extend our heart-felt gratitude to Anne Kennedy and John Sorflaten,
our technical editors, for their much valued feedback and suggestions.
Your unique perspectives and encouraging comments on the chapters
helped to focus the writing during late-night jam sessions.
To Francois Roux, thank you for the many photographs you took
during the cover conception phase and for the final photograph that
appears on the cover. To Ron Leishman, thank you for bringing the
stories in this book to life with your illustrations and for injecting the
book with your distinctive brand of visual humor.
Thank you to our colleagues whose work with web usability, search
usability, search engine optimization, web design and development, copy-
writing, information architecture, and more have helped lay the ground-
work for books like this. It’s your work, some of which is referenced in
this book, we recommend to clients, colleagues, and students.

Shari Thurow would like to personally thank her co-author, Nick Musica.
Years ago, I knew I needed a co-author to address this complex but fasci-
nating topic. I remember constantly saying to you, “Keep writing things
down. I know you will use these experiences some time in the future.”
Thank you for taking my advice and for being one of my most valued
friends and colleagues.
 IX

I would also like to thank the folks at Incisive Media and Search
Marketing Expo—Karen DeWeese, Chris Elwell, Stewart Quealy, and
Marilyn Crafts—for providing me with a venue to present this topic. You
have been wonderful, supportive colleagues and friends.
My success in this industry clearly lies with the existence of the
commercial web search engines and their vast knowledge of information
retrieval and search usability. I extend my heartfelt thanks to my search
engine colleagues—Matt Cutts, Jon Glick, Amit Kumar, Tim Mayer, Priyank
Garg, Rajat Mukherjee, Amit Singhal, Bryan White, and Michael Yang.
And finally, my sincerest gratitude goes to the four people who inspire
me to continue my lifelong foray into search usability: Danny Sullivan,
Chris Sherman, Susan Weinschenk, and Peter Morville. A chance online
meeting with Danny in the mid 1990s inspired me to pursue search-
engine-friendly web design. And both Susan’s and Peter’s work inspired
me to pursue website usability and information architecture as important
elements of search-engine-friendly web design.

Nick Musica would like to personally thank his friend, colleague, and
co-author Shari Thurow. You are a true inspiration and a pleasure to
work with. Thank you for your generosity and for inviting me to take
this ride with you.
Thank you to those indirectly involved in the creation of this book.
To my clients and employers along the way who have given me their
trust. It’s because of your faith in my abilities that I have been able to
grow and learn professionally from the experiences.
I’ve been fortunate to work with many people whose influence
has helped shape my professional life. The list is long, but at the top are
Andy Fischler, Tracy Wehringer, Nick Valente, Marc Buro, Cle Scouten,
Bonnie Blake, and Shari Thurow.
Thank you to my family and friends whose phone calls, emails, and
texts were returned later than I would have liked. Thank you for your
understanding, support, and the occasional healthy distraction during the
writing of this book. Thank you to those who shared in the life experi-
ences used to relate some of the concepts in this book. Thank you to those
across the border who helped me get away, even if it was just for the day.
And thank you to those on two and four legs who dropped by to see how
things were going and let me know when the Atlantic was calling.
And last but not least, to users who aren’t afraid to “think out loud,”
thank you for showing me what only you can show.
X

! CONTENTS AT A GLANCE

Foreword ................................................................................................... xv

Introduction ............................................................................................ xvii

CHAPTER 1: Understanding Search Usability 1

CHAPTER 2: The Scent of Information and Web Search Engines 17

CHAPTER 3: Navigational Searches—Where Can I Go? 33

CHAPTER 4: Informational Searches—What Can I Learn? 47

CHAPTER 5: Transactional Searches—What Can I Do? 61

CHAPTER 6: The Scent of Information and Landing Pages 71

CHAPTER 7: Search Usability and Your Site’s Success 89

CHAPTER 8: Search Usability Is Everyone’s Job 107

CHAPTER 9: How To Improve Your Website’s Search Usability 141

Index ........................................................................................................ 171

 CONTENTS XI

! CONTENTS

Foreword ................................................................................................... xv

Introduction ............................................................................................ xvii

CHAPTER 1: Understanding Search Usability 1

Differing Perspectives on Search .................................................... 2
Search Engine Optimization (SEO) ................................................ 5
Website Usability ......................................................................... 6
The Scent of Information ............................................................... 8
Why Search Usability Is Important .............................................. 13
Conclusion .................................................................................... 14

CHAPTER 2: The Scent of Information and Web Search Engines 17

Different Perceptions of Querying Behavior ............................... 19
Missing Information Scent .......................................................... 20
Irrelevant or Poor Search Results ................................................. 20
How To Provide Information Scent in Search Listings ................ 22
Title Tags ................................................................................... 23
Meta-Tag Description or Page Snippet ......................................... 24
URL Structure (Web Address) ..................................................... 26
3 Types of Web Search Engine Queries ....................................... 29
Navigational Queries .................................................................. 30
Transactional Queries ................................................................. 30
Informational Queries ................................................................. 30
Conclusion .................................................................................... 31

CHAPTER 3: Navigational Searches—Where Can I Go? 33

Navigational Intent ....................................................................... 34
Example 1: Home Pages .............................................................. 35
Example 2: Login Pages ............................................................... 37
Example 3: Category and Channel Pages ...................................... 39
Search Listings and Navigational Queries .................................... 42
Navigational Query Indicators ..................................................... 45
Optimizing for Navigational Queries ........................................... 45
Conclusion .................................................................................... 46
XII

CHAPTER 4: Informational Searches—What Can I Learn? 47

Informational Intent ..................................................................... 48
Example 1: Quick Facts ............................................................... 49
Example 2: Questions and Answers .............................................. 51
Example 3: Lists ......................................................................... 53
Search Listings and Informational Queries .................................. 55
Informational Query Indicators ................................................... 57
Optimizing for Informational Queries ......................................... 58
Conclusion .................................................................................... 59

CHAPTER 5: Transactional Searches—What Can I Do? 61

Transactional Intent ...................................................................... 62
Example 1: Buying Products ........................................................ 63
Example 2: Entertainment .......................................................... 64
Search Listings and Transactional Queries ................................... 67
Transactional Query Indicators .................................................... 67
Optimizing for Transactional Queries .......................................... 68
Conclusion .................................................................................... 70

CHAPTER 6: The Scent of Information and Landing Pages 71

Scanning Behavior ........................................................................ 72
Anticipate Searcher Needs with Usability Tests ............................. 75
Avoid the Iceberg Effect ................................................................ 77
Avoid Misleading Links ............................................................... 79
Foster Orienting Behavior ............................................................ 81
Pogo-sticking: Negative Search Behavior ..................................... 86
Conclusion .................................................................................... 87

CHAPTER 7: Search Usability and Your Site’s Success 89

Search Usability Metrics by Site Type .......................................... 90
Selling Products and Services with Search Usability ...................... 90
Selling Ads with Search Usability ................................................ 94
Supporting Existing Products and Services ................................... 97
Search Usability for Employee Intranets .................................... 101
Common Search Usability Goals ................................................ 102
Top SERP Rankings ................................................................. 102
 CONTENTS XIII

Bounce Rates ............................................................................ 103

Time-on-Site and Page Views .................................................... 104
Conclusion .................................................................................. 105

CHAPTER 8: Search Usability Is Everyone’s Job 107

The Hole in the Fence: A Story About Perspective .................... 108
It Takes a Village To Build for Search Usability .......................... 109
Graphic Designers and Search Usability ..................................... 109
Web Copywriters and Search Usability ....................................... 112
Marketing Communications and Search Usability ...................... 117
Usability Professionals and Search Usability ............................... 119
Management and Search Usability ............................................. 125
Market Researchers and Search Usability ................................... 130
IT and Search Usability ............................................................ 132
Information Architects and Search Usability ............................... 136
Conclusion .................................................................................. 139

CHAPTER 9: How To Improve Your Website’s Search Usability 141

Ideal Test Participants ................................................................. 142
Reverse Card Sort Test ................................................................ 143
One-on-One Field Interviews ..................................................... 147
Functional Salience Test ............................................................. 148
Home Page Expectancy Test ...................................................... 149
Step 1: Click Affordance Test ..................................................... 150
Step 2: Expected Content Test .................................................... 152
Eight-Second Usability Test on Content Pages .......................... 153
Observe Navigational Web Searches .......................................... 154
Review Site Search Data ............................................................. 155
Brand Perception Test ................................................................. 157
Check Your Error Messages ........................................................ 159
Free Exploration Test .................................................................. 160
Deconstructing Common Myths and Misconceptions .............. 161
Myth #1: There Is a Magic Bullet .............................................. 161
Myth #2: There’s No Such Thing as Perceived vs.
Actual Website Speed ........................................................ 162
Myth #3: The Competition Does It Right ................................... 163
XIV

Myth #4: You Can Design a “Good” Site Without

Feedback from Users .......................................................... 163
Myth #5: People Will Use Your Site Since It’s the
Best of Its Kind ................................................................. 164
Myth #6: You Lose Users After Three Clicks ............................... 165
Myth #7: The User Is the Problem ............................................. 166
Myth #8: The User Is Always Right .......................................... 166
Myth #9: Usability Tests Are 100% Accurate ............................. 167
Myth #10: Users Will Figure Out What Is a Link
by Clicking It .................................................................... 167
Myth #11: Usability Guidelines Will Make Your
Site Successful ................................................................... 168
Myth #12: The Web Is Yours ..................................................... 168
Conclusion .................................................................................. 169

Index ........................................................................................................ 171

 FOREWORD XV

! FOREWORD
In the heart of the Yorkshire Dales in Northern England, a pleasant wan-
der leads through the woods from Bolton Abbey to the Strid, a notorious
stretch of water where the River Wharfe is forced into a deep and narrow
channel. At its narrowest point, the Strid is about two metres (or a lengthy
stride) wide, just right to tempt a daring child or a reckless adult. But,
jumping the gap is risky, for the rocks are slippery, and nobody survives
a fall into the thundering waters and deep, dark caverns of the Strid.
I’m reminded of this chasm not crossed when I ponder the state of
search, a vast sociosemantic territory that’s riddled with potholes, gaps,
and schisms. Many of these fissures are carved by the relentless process
of specialization. Designers, developers, information architects, search
engine optimization specialists, and web analyticists are valued for their
unique strengths, but their singular skills often come with a narrow focus
and an idiosyncratic vocabulary that makes teamwork difficult at best.
And, of course, this collaboration chasm creates an experience gap
that frustrates users to no end. Our expectations, raised high by Google,
are all too often shattered by the sad state of site search. And, as the deep
web grows ever more vast and unruly, even mighty Google cannot keep
pace. The gulf between people and the answers they seek threatens to
widen, and it’s for this reason that this book that you hold in your hands
is so vital to the future of findability.
Shari Thurow is among the few specialists brave enough to bridge
the gap between search engine optimization and web usability. As a
result, she has learned how and where to place stepping stones and build
bridges. She can speak the language of link analysis and relevance ranking
algorithms, while also understanding user psychology and information-
seeking behavior. In this book, Shari and her co-author Nick Musica
explain how you can design websites that are friendly to spiders and
humans, so that even novice users can find the site and find their way
around the site without plunging headfirst into the abyss.
So, if you’re crazy enough to leap the Strid alone, it’s ready and
waiting. But, if you’d prefer a safer path across the chasm, then read this
book. See you on the other side!
—Peter Morville
—Author of Ambient Findability: What We Find Changes Who We Become
(O’Reilly Publishing, 2005) and Information Architecture for the World
Wide Web, 3rd Edition (O’Reilly Publishing, 2006).
This page intentionally left blank
 INTRODUCTION
Why is this book needed?

In the United States alone, users submit more than 10 billion searches a
month on commercial search engines such as Google, Yahoo, and MSN.
Many of these searches leave users frustrated because what they find
on search engine results pages and on resulting websites doesn’t meet
the expectations of their search queries. This means a poor experience
for users and lost revenue for website owners.

So, what’s the problem? The problem isn’t the users or their queries.
It’s not the commercial search engines either. We are the problem—
search engine optimizers, web usability professionals, web designers
and developers, information architects and copywriters and other web
professionals whose work either helps users find what they are looking
for or doesn’t. We are also the solution.
XVIII INTRODUCTION

This book covers how users look for information and how search
engines anticipate the intent behind users’ queries so that we can work
collectively to bridge the gap between what users search for on search
engines and what they find on our websites. In short, it’s your work that
can improve the user experience and increase revenue by getting the right
user to the right content if you understand the intent behind a user’s
query and the many ways users search for information on the web.
This book also provides a common language for web professionals
from various backgrounds to use so we can better communicate with
each other. For example, when you hear someone talking about a navi-
gational query you’ll know that they are talking about a user trying to
locate a specific website using a commercial search engine. When you
hear someone say they saw someone pogo-sticking on a website, you
know they were referring to a user’s search behavior and not to someone
actually using a pogo stick.
It takes many web professionals, from multiple disciplines, to get the
right user to the right content. Users don’t know this, nor do they care.
They just want to find their content. This book will help you get users
to their content and meet business goals with the same user click.

! WHO SHOULD READ THIS BOOK?

This book is for search engine optimization professionals, web usability
professionals, web designers, web copywriters, web developers, online
marketers, brand managers, website owners, management, and anyone
who wants to improve their users’ experience and increase revenue.
The main intended audience for When Search Meets Web Usability is
search engine optimization (SEO) professionals who want to go beyond
achieving page one rankings to creating a better experience for their
users that also contributes to the bottom line of the business. For these
SEO pros, this book will be valuable for self-education and as a refer-
ence book.
This book will assist SEO professionals in accomplishing the how of
search engine optimization by explaining why search engines serve up
the results they do and understanding the various behaviors users exhibit
as they look for information on the web.
 INTRODUCTION XIX

We have sprinkled resources throughout these pages for those who

wish to educate themselves on web usability and search usability topics
that go beyond the focus of When Search Meets Web Usability.
Web usability professionals are the second target audience. Web
usability professionals will most likely be familiar with many of the
concepts discussed in When Search Meets Web Usability. What you will
find here that you haven’t seen in other usability books is how to apply
your skills and background beyond the websites to where many times the
user experience really begins—the search engines. This book is meant
to supplement your current usability methodologies and practices by
showing how you can apply your knowledge of human factors to web
users’ querying behaviors on the commercial search engines.
Web usability professionals will also see topics they aren’t familiar
with that are of a search engine optimization nature and are critical to
understanding how users behave when they query on the commercial
search engines.
The third target audience includes, but is not limited to, web design-
ers and developers, information architects, copywriters, and other web
professionals. If you perform one or more of these roles, this book is
meant to complement your background and supplement your skills by
providing another perspective of the web and how your work affects
users and the bottom line.

! INGREDIENTS TO MAKE YOUR OWN

SECRET SAUCE
There is no off-the-shelf secret sauce that this or any other book can offer
you that will magically make your website number one in the commer-
cial search engines, make users love your site, and make you millions
of dollars.
What we can offer you instead are the ingredients to make your own
secret sauce. All of the ingredients are in the pages of this book. They
include understanding how search engines anticipate the intent of users,
what influences users to click on your listing in the search engine results
pages (SERPs), and understanding the many behaviors users exhibit
when they search for information in the SERPs and on your website.
It also includes something that strings everything together, called the
scent of information.
XX INTRODUCTION

Figure i.1 When it comes

to a secret sauce, some like
it hot, some like it not. You
need to season your website
to your users’ taste.

By using these ingredients, you’ll be able to create a unique recipe

that will serve your users and your clients or employers better than any
generic secret sauce ever could. We’ll show you how.

! WHAT YOU’LL LEARN

" The how and why of search engine optimization return on
investment (ROI)
" Why Google doesn’t want your site to be number one and
what to do about it
" What makes a user click a SERP listing and how you can use
that knowledge for profit
" The behaviors users exhibit when they look for information on
the web
" How one user’s click can both satisfy user goals and increase
your revenue

! WHERE DID THE RESEARCH IN THIS BOOK

COME FROM?
The three query types (navigational, informational, and transactional
queries) covered in this book originated from research by engineers at
commercial search engines such as Google, Yahoo, and MSN. Since com-
mercial search engines started using these query types they have been
 INTRODUCTION XXI

widely written about in academic circles but have never been presented
for consumption to search optimization professionals, web usability pro-
fessionals, and other web professionals in the context of search usability
and in the detail in which they are covered in this book.
Additional research was adapted from the web usability profession.
We’ve taken concepts and practices of web usability that historically
only have been applied to websites and have applied them to how users
use commercial search engines. We’ve also applied these concepts and
practices to the websites that users find via a search engine query to show
how we can improve the entire user experience starting with the query
and finishing on your website.
Finally, you will see that some of the content in this book comes
from our personal experience. We’d like to note that when we write from
our personal experience we are not writing from our personal prefer-
ence, but from our professional experience observing users and their
behavior. We all have opinions, but this book is not about opinions; it’s
about understanding user behavior by distinguishing fact from opinions.
Even our opinions.

! HOW TO USE THIS BOOK

We suggest you read this book from beginning to end regardless of
your experience as an SEO professional, usability professional, or other
web professional; each chapter builds on concepts covered in previous
chapters.
Chapters 1 through 6 define search usability terminology including
information scent, the three types of queries that search engines use to
anticipate the intent of a user’s search, and the behaviors users exhibit
on the web. The concepts and vocabulary defined in Chapters 1 through
6 are necessary to understand Chapters 7 through 9, which illustrate the
return on investment of search usability, how different web profession-
als help or hurt search usability, and how you can get one-on-one direct
feedback from your users by conducting web usability tests.
If you are a web usability professional, a glance at the this book’s
table of contents might tempt you to skip certain chapters. Don’t skip
chapters that look familiar to you, as they contain something new. We’ve
taken concepts that many of you are familiar with and applied them to
how users query on the web. So, while concepts such as eye tracking,
functional salience tests, and expectancy tests might be familiar, how they
are applied in the context of search usability most likely won’t be.
XXII INTRODUCTION

! WWW.SEARCH-MEETS-USABILITY.COM
We’ve supplemented this book with a companion site located at
www.search-meets-usability.com. Here you will find:
" A list of resources for you to continue your search usability
education
" Microsoft Excel spreadsheets from Chapter 7
" Content that didn’t make it into the book due to publishing
limitations
" Additional content, resulting from feedback we get from read-
ers like yourself
" A growing glossary of search usability terminology for you to
reference, recommend to colleagues, and use to communicate
with other web professionals
CHAPTER 1
UNDERSTANDING SEARCH
USABILITY
A user-friendly, search-friendly website benefits both site visitors and
businesses. Site visitors easily find what they need, accomplish their goals,
and leave satisfied. Businesses reap the benefits of a happy customer.
However, search engine optimizers and website usability professionals
focus on different aspects of search usability. By applying a holistic
approach and merging the skills and widening the focus of both search
engine optimizers and usability professionals, you can increase traffic,
leads, sales, and happy customers. But first, we need to understand
what search actually means.

If you ask people to envision the word search, you will get a wide variety
of responses. On the web, the word search has become synonymous
with Google, as in, “I Googled his name to get some background
information before we went on a date.” Others envision a scene from
a Sherlock Holmes story, perhaps an image of a bloodhound sniffing
a crime scene for clues. To others, the word search is associated with
2 CHAPTER 1: UNDERSTANDING SEARCH USABILITY

misplaced keys and a scramble to locate them before leaving for work
(Figure 1.1). Or even a scavenger hunt at a family picnic. Clearly, the
meaning of search is multifaceted. Its meaning depends on its context—
both online and offline.

Figure 1.1 Thinking that Google is the beginning and end of how people look for con-
tent on the web makes just as much sense as trying to use Google to find your keys.

! DIFFERING PERSPECTIVES ON SEARCH

On the web, it is easy to see why the word search is associated with
search engines only. The two main visual cues that communicate online
searching are: a button labeled “Search” and a text box where searchers
can describe their desired information, destination, or activity. Billions
of searches are performed on Google, Yahoo, and Microsoft Live every
month. Millions of websites have a site search engine. Therefore, con-
sidering the tremendous use of web and site searches, millions of people
associate online searching with search engines. Figure 1.2 and Figure 1.3
are examples of two types of search engines. The interfaces for both
websites contain a Search button and a text box.

Note A web search engine,

such as Google, is a tool
designed to discover,
retrieve, analyze, and
display information from
the World Wide Web.
On the other hand, a site
search engine is a tool that
retrieves, analyzes, and
displays snippets of infor- Figure 1.2 Web search engine.
mation about pages from a
specific website.

Figure 1.3 Site search engine.

 DIFFERING PERSPECTIVES ON SEARCH 3

However, people do not use only the commercial web search engines
to look for content on the web. People might go to a specific web page
after they remember a reference from newspaper, billboard, television
show, radio program, or even word of mouth. They type the website
address (URL) in an address bar (Figure 1.4 and Figure 1.5).
Figure 1.4 The web address
q bar on the current Firefox
browser (for laptop and
desktop computers).

Figure 1.5 The web address

bar on an Apple iPhone.

In addition, people might look for web content by clicking a link

from an email, text message, or an online advertisement. They also locate
web content by clicking links from one site to another, commonly known
as surfing or browsing the web. When people look for content via these
methods, they are still searching. When searching these other ways, we
still “search” even though we don’t see the familiar text entry box and
Search button.
On the web, search usability refers to how easily users can locate
and discover content on a site via retrieval (searching/querying) and
navigation (browsing). Search usability also refers to users’ level of sat-
isfaction as they discover or locate their desired content. Therefore, to
say a website is “usable” means that users find no obstacles in finding
what they want. In fact, they actually get satisfaction from the process
of finding what they want.

On the web, search usability refers to how easily users

can locate and discover content on a site via retrieval
(searching/querying) and navigation (browsing).
4 CHAPTER 1: UNDERSTANDING SEARCH USABILITY

Because users locate and discover web content using multiple tech-
niques, search usability addresses both browsing behavior—clicking a
link from one web page to another—and querying behavior. People find
web content in a wide variety of ways, not only via the commercial web
search engines.
Search engine optimizers and usability professionals focus on dif-
ferent aspects of search usability (Figure 1.6). Let’s look at what each
practitioner focuses on.
Figure 1.6 Both search
engine optimizers and web-
site usability professionals
only address part of search
usability. Users exhibit all
four behaviors.
 DIFFERING PERSPECTIVES ON SEARCH 5

Search Engine Optimization (SEO)

Search engine optimization has traditionally been defined as the process
of designing, writing, coding (in HTML), scripting, and programming
an entire website so that there is a good chance that web-page listings
will appear in top positions in web search results for selected keywords.
Most people view SEO as an online marketing strategy. Many search
engine optimizers focus on web search engine optimization (Figure
1.7) because they see SEO in the context of designing and writing web
content for search engines, not for human beings.
Figure 1.7 Many search
engine optimizers focus
on web search engine
optimization.

Search engine optimization is not optimizing a website for search

engines. Rather, SEO is optimizing a website for people who use search
engines. There is a human element to SEO that usability professionals
might not realize.
Note An example of an
Understanding searchers and their goals is a large part of the optimi- objective third-party link
zation process. Search engine optimizers gather information about users from one website to another
with keyword research tools and web analytics data. Link popularity, or is a link on Wikipedia that
leads to a web page that is
the more appropriate term link development, is one of the key building
not within the Wikipedia
blocks of search engine optimization. Link development is the number and site, such as a link from
quality of external, third-party links pointing to a website. All the major http://en.wikipedia.org/wiki/
web search engines use link popularity to determine positioning in search National_cancer_institute to
the official National Cancer
results pages. Therefore, the goal for most search engine optimizers is to Institute site at
assist website owners in obtaining high quality links (Figure 1.8). http://www.cancer.gov.

Figure 1.8 Search engine

optimizers also focus on
link development.

Search engine optimizers specialize in areas of search usability that

usability professionals do not. Unlike usability professionals, for example,
most search engine optimizers do not have one-to-one face time with
actual users.
Now let’s look at what usability professionals focus on.
6 CHAPTER 1: UNDERSTANDING SEARCH USABILITY

Website Usability
Just as the word search elicits a wide variety of images and interpretations,
so does the word usability. In fact, many people have never even heard
of the word usability. If you ask the average person on the street or in
an elevator what they believe usability means, you might get a confused
expression in response. One common interpretation of the word is to
break it down into two components, use and ability, resulting in the
perceived definition of usability as a person’s ability to use something.
When referring to a website, people often think, “I can use my website.
Therefore, it must be user friendly.”
Unfortunately, this common interpretation of the word usability is
incorrect, because, when it is applied to a website, usability has nothing
to do with personal opinion. Website usability is task-oriented. Through
Note For a more detailed
usability testing, usability professionals measure how well users can
explanation of website
usability, see http://www. accomplish specific tasks on a website with efficiency and a high level
w3.org/WAI/redesign/ucd. of satisfaction.
During usability testing on a website, usability professionals
measure:
" Effectiveness. Can site visitors achieve what they need from
using your website? Can they easily find their desired informa-
tion, destination, or activity?
" Efficiency. How quickly can site visitors complete their goals
on your website? How many steps are required to complete
basic tasks? What elements on a site discouraged or prevented
site visitors from achieving their goals?
" Learnability. How quickly and easily can new site visitors
learn to use your website? What elements on a site can help
site visitors form an accurate mental model of your site?
" Memorability. How quickly and easily can repeat site visitors
remember how to use your website effectively to accomplish
their goals?
" Error prevention and recovery. No website is 100 percent
error-free. Does the website implement a design that allows for
easy recovery from errors?
" Satisfaction. User satisfaction is heavily influenced by the
ability to achieve goals. If site visitors are able to achieve their
goals on a website with ease and efficiency, they report high
satisfaction. If site visitors encounter too many roadblocks,
they report low satisfaction, even if they were able to accom-
plish their goals. Would a test participant recommend the
website to others?
 DIFFERING PERSPECTIVES ON SEARCH 7

Good website usability is not the same as people saying they like
or use your site. Focus group opinions bear little resemblance to actual
behaviors exhibited during usability tests. Because they do not wish
to appear stupid in front of a group, people downplay problems and
issues with a site. They will say that they like a site even though they
are unable to find their desired content. People are not always objective
when evaluating their own behavior.
Usability professionals recognize the value of one-on-one interac-
tion with actual users through usability testing and field studies. They
see firsthand that a positive user experience does not come from a brand
manager’s perspective, a marketing manager’s perspective, management’s
perspective, or even an SEO professional’s perspective. Positive user
experiences come from observing users’ successful completion of a task,
and not solely from users saying that they felt good about it. Usability
professionals focus on search usability when users are already on a website,
usually beginning at a home page (Figure 1.9). Usability professionals
do not always consider how users actually arrive on a site.
Figure 1.9 Usability tests
are conducted on a web-
site, typically beginning
with a home page.

Finally, in the website usability industry, the term search usability is

typically used to describe the user-friendliness of a site search engine
(Figure 1.10). What both search engine optimizers and usability profes-
sionals might not realize is that the techniques search engine optimizers
use often make site search results more accurate. In other words, many
web SEO techniques make site search results more accurate because
page content becomes more keyword focused.
Figure 1.10 Usability tests
are conducted on a web-
site, including tests on a
site search engine.

Problems arise when search engine optimizers and usability profes-

sionals work against each other, sometimes without even realizing it. For
example, many search engine optimizers put so many keywords on a
web page that the page becomes unusable. On the flip side, in the name
of reducing the amount of text on a page, many usability professionals
8 CHAPTER 1: UNDERSTANDING SEARCH USABILITY

remove important keywords from web pages, making them appear

unfocused to both site visitors and search engines. The result? The user
experience and business goals are compromised.
Note See http://www. Findability expert Peter Morville eloquently stated that people
findability.org/archives/ cannot use what they cannot find. Search engine optimization profes-
cat_findability.php for the sionals help make content easy to find via the commercial web search
link to Peter Morville’s
statement.
engines. Website usability professionals help make content easy to find
on a website. The string that connects these two groups is the scent of
information. People need the scent of information in order to locate
and discover desired content. Let’s explore what we mean by the scent
Note The concept of the of information next.
scent of information comes
from information foraging
theory by Peter Pirolli, ! THE SCENT OF INFORMATION
Stuart Card, and colleagues
at the Palo Alto Research A web page consists of many graphical and textual cues that facilitate
Center. See Information navigation (where can I go?), orientation (where am I?), and assessment
Foraging Theory: Adaptive
Interaction with Information of content value (should I click this link?). These words and images are
(New York, NY: Oxford the scent of information. If the scent of information is strong, people click.
University Press, 2007). If the scent of information is too weak or disappears, people do not click.
To make web content findable, information scents should be clearly
established and consistently maintained throughout a website.

If the scent of information is strong, people click.

Note On a website, orienta- We react to various information scents in our everyday life without
tion is a behavior whereby even realizing it. For example, when you ride in an elevator, the elevator
users determine their posi- buttons and lights provide a scent of information. If you press the button
tion with reference to
another point, establishing a
labeled “3,” the elevator will take you to the third floor. If you press the
“sense of place.” For exam- button labeled “5,” the elevator will take you to the fifth floor. If you press
ple, locational breadcrumb the button labeled “Lobby,” the elevator will take you to the lobby.
links help people understand
Sometimes, though, the scent of information on elevators is not clear.
where they are in the hierar-
chy of a website, such as The following photographs and illustrations show two sets of elevator
Home > Teas > Green Teas. buttons from the same building. In Elevator 1, shown in Figure 1.11, if
Orientation will be discussed you wanted to go to the lobby, which button would you push?
in detail in Chapter 6, “The
Scent of Information and When we observed people using the set of buttons in Elevator 1, they
Landing Pages.” had confused expressions on their faces and hesitated to push a button.
They were unsure which button would lead them to the lobby.
Most of them assumed that the buttons labeled 1R or 1F were the
two best options, but they were unsure what the abbreviations “R” and
“F” represented.
 THE SCENT OF INFORMATION 9

Figures 1.11 In Elevator 1, the

scent of information to the
lobby is weak and confusing.

Some people mentioned that the lobby is not always on the first
floor and were ready to press either the button labeled B or D. However,
they assumed that the letter “B” was the abbreviation for the basement,
which confused them even more. No one knew what the letter “D”
represented.
Ultimately, many of them said aloud, “Which one goes to the
lobby?”
Interestingly, in the same building, a different situation was available
in Elevator 2, shown in Figures 1.12. If you want to go to the lobby and
are riding in Elevator 2, the correct button to push is more obvious.
Figures 1.12 In Elevator 2,
the scent of information to
the lobby is very clear.
10 CHAPTER 1: UNDERSTANDING SEARCH USABILITY

Note According to Gestalt When we observed people using the set of buttons in Elevator 2,
Principles of Visual there was no hesitation. There were no questions. People pressed the
Organization, items placed button labeled 1F because they correctly determined that pressing that
close to each other are
associated with each other.
button would get them to the lobby due to the strong scent informa-
This is called the Principle tion. Two items, in particular, established the scent of information: the
of Proximity. See http:// word “Lobby” and its placement near the elevator button that will take
www.interaction-design.
them to the lobby.
org/encyclopedia/gestalt_
principles_of_form_ Another item that communicates a strong scent of information is
perception.html. a map you might find in a shopping mall, or a mall directory, shown in
Figure 1.13:
Figure 1.13 The scent of
information on this mall
directory contains both
graphical and textual cues
to help mall visitors estab- Bloomingdale’s
lish where they are and
where they can go.

1 2 3 4 5 6 7
8 9 10 11 12 13

14 15 16 17 18 19

20 21 22 23 24 25

26 27 28 29 30 31
32 33 34 35 36 37 38
You are here

First Floor

This diagram helps people understand where they are located within
the mall. There is a graphical cue (a star) with a corresponding text label,
“You are here.” Another text label, “First Floor,” lets them know what
floor they are on.
Additionally, this mall directory shows people where they can go in
multiple ways. People do not enter and exit a shopping mall through a
 THE SCENT OF INFORMATION 11

single entry point. Shopping malls have multiple entrances. Many shop-
ping malls have multiple wings, and there might be a main entrance for
each wing. People can enter and exit a shopping mall via individual stores
as well (such as the Bloomingdale’s store shown in Figure 1.13).
Websites are similar to shopping malls in that they have multiple entry
and exit points. When people arrive at a website, they do not always land
on a home page, especially if they are coming from a web search engine.
They might land on a page in the middle of a site. Site visitors need to
see the textual and graphical cues that comprise the scent of information
on a web page. If they don’t, then they abandon a site. Let’s look at a
before-and-after example from the National Library of Medicine.
Back in 2002, the National Library of Medicine’s home page looked
like Figure 1.14:
Figure 1.14 Back in 2002,
the scent of information
on the National Library of
Medicine’s home page was
not very effective.

The scent of information provided on this home page might be

effective for medical librarians but not for other site visitors. For example,
what are MEDLINE and MEDLINEplus? These items appear multiple
times on this home page as text and graphic images, but why are these
items important to site visitors? Why would a person who wants health
information click either of these links? Since this page provides little
information about the value of the link, people are not as likely to con-
tinue to navigate the site for health information. Furthermore, the text
links under the Health Information do not look clickable.
12 CHAPTER 1: UNDERSTANDING SEARCH USABILITY

Now let’s look at the revised version of the National Library of

Medicine’s home page in Figure 1.15.
Figure 1.15 The new home
page provides more effective
information scents for both
users and search engines.
 WHY SEARCH USABILITY IS IMPORTANT 13

The revised home page provides more effective information scents Note The photograph is a
for both site visitors and search engines. Now people who are unfamiliar good choice for the scent of
with MEDLINEplus understand what they can learn by clicking the link. information because women
typically make healthcare
The MEDLINEplus graphic includes explanatory text, and people can get decisions in a family. If a sec-
health and drug information for patients, family, and friends. This portion tion of this website were tar-
of the website is clearly written for consumers, not medical professionals. geting medical professionals,
a photo of a physician, nurse,
The photograph of people who are not medical professionals and the
a person wearing a lab coat,
additional words in the annotation provide a stronger scent of informa- or a group of these people
tion for all site visitors: laypersons, librarians, and medical professionals would be a good choice.
than the word MEDLINEplus. The additional words, in particular, also
help make web pages more search-engine-friendly.

! WHY SEARCH USABILITY IS IMPORTANT

As we mentioned in the introduction, a user-friendly, search-friendly web-
site benefits both site visitors and businesses. The primary benefits to site
visitors are that they are able to achieve their goals easily, efficiently, and
with a high degree of satisfaction. The primary benefits for businesses
are increased traffic, qualified leads, closed sales, and happy customers.
Ignoring or minimizing search usability can result in substantial
costs to a business not only in terms of lost sales and prospects, but also
in terms of customer satisfaction, increased advertising expenses, and
staff time. Websites with content that is difficult to find can result in the
following business costs:
" If people find it difficult to locate or discover their desired
content, they will not use your site.
Whenever people make a choice, they tend to pick the
easiest option. This is especially common on the web, where a
more user-friendly website is only one or two clicks away.
If people believe a website does not contain their desired
content, they will quickly abandon the site. They will not Add
to cart. They will not write a product review. They will not join
your online community. Moreover, they are unlikely to return
to the site, remembering it as the site that delivered a poor user
experience because:
" They could not locate their desired content at all.
" The website made it very difficult to locate their desired
content.
" The content they found was not what they expected or
wanted.
14 CHAPTER 1: UNDERSTANDING SEARCH USABILITY

" If people feel they must use a website where desired content
is difficult to find, they will use the site as little as possible.
If a website is not user-friendly, site visitors will go to great
lengths to avoid using the site, even if they believe the desired
content is outstanding. Site visitors often create elaborate
workarounds to avoid using the website, and will share these
workarounds with others.
For example, it is not uncommon to see people use Google
to go directly to a specific page within a website, because the
site’s navigation scheme is too confusing or the URL (web
address) is too complex.
" If people have a difficult time finding desired content on
your site, they will waste time.
A website that is not user-friendly and search-friendly will
make site visitors hunt for their desired information unneces-
sarily. Site visitors will take longer to learn how to use your
website, and any tasks that they do not perform regularly will
take longer to complete, resulting in more errors.
" The website will need more changes and enhancements.
If searchers’ needs and abilities were not considered when
determining the requirements, design, and programming
of a website, then the site is likely to require more changes
and enhancements. Result? Businesses must allocate more
staff and/or more staff time to a website to fix problems that
should have been addressed before the site was launched.

! CONCLUSION
Key points in this chapter:
" On the web, the word search is associated with search engines
because of a button labeled “Search” and a text box where
people can describe their desired information, destination,
or activity.
" Search usability refers to how well users can locate and discover
content on a website via retrieval (searching/querying) and nav-
igation (browsing) in order to achieve their goals on the web.
" Search engine optimizers and website usability professionals
focus on different aspects of search usability.
" Search engine optimization (SEO) is not designing and writing
a site for search engines. Best practice SEO is optimizing a
website for people who use search engines.
 CONCLUSION 15

" Search engine optimizers focus on both querying and brows-

ing behaviors before people arrive on a website. Usability
professionals focus on querying and browsing behaviors after
people arrive on a website.
" Website usability is task oriented. Through usability testing,
usability professionals measure how well users can accomplish
specific tasks with efficiency and a high level of satisfaction.
" Website usability is about effectiveness, efficiency, learnability,
memorability, error prevention, and satisfaction. It is not a set
of focus group opinions.
" People cannot use what they cannot find, in the words of Peter
Morville.
" The string that connects website usability professionals and
search engine optimizers is the scent of information.
" The scent of information consists of graphical and textual
cues that facilitate navigation (where can I go?), orientation
(where am I?), and assessment of content value (should I click
this link?).
" To make web content findable, make your information irresist-
ibly meaningful and consistent in its allure throughout your
website.
This page intentionally left blank
CHAPTER 2
THE SCENT OF
INFORMATION AND
WEB SEARCH ENGINES
People use commercial web search engines for a wide variety of reasons.
They might want to look up a phone number or get directions to a res-
taurant. They might want to read consumer reviews before purchasing
an item. Or they might want to play a quick game of Sudoku online.

So, to help searchers succeed in getting what they want, savvy marketers
must understand: how people are searching, and why people are searching.
Once marketers know the answers to these questions, they can adjust
their search marketing strategies. Not only will their websites receive
increased traffic, they will also receive more qualified traffic that results
in higher conversions, sales, and lifetime customers (Figure 2.1).
18 CHAPTER 2: THE SCENT OF INFORMATION AND WEB SEARCH ENGINES

Figure 2.1 Both search

engine marketers and usabil-
ity professionals can get a
higher return on investment
HOW + WHY = ROI
people search people search (return on investment)
if they understand how and
why people search. They can
use that knowledge to build
more user-friendly websites. If many search engine marketers were fishermen, they just might
starve. While a fisherman’s gear might help attract multiple types of fish,
there is no one-size-fits-all gear in the tackle box (Figure 2.2). Fishermen
have multiple rods, reels, and bait to choose from depending on what
type of fish they wish to catch. It is the type of fish that determines the
gear the fisherman uses.
Figure 2.2 Ranking well on a
web search engine is point-
less if you are reaching the
wrong target audience.

Likewise, it is the type of searcher that determines the marketing

strategies that search engine optimization (SEO) professionals use. Is
the target audience novice, intermediate, or advanced searchers? What
are their main goals? Searchers can be characterized by experience level,
plus their goals and behaviors. When people use the commercial web
search engines, they are trying to communicate where they want to go,
what they want to read, and what they want to do.

When people use the commercial web search engines,

they are trying to communicate where they want to go,
what they want to read, and what they want to do.
 DIFFERENT PERCEPTIONS OF QUERYING BEHAVIOR 19

Knowledgeable SEO professionals continually try to understand the

goals and characteristics of web searchers. With this knowledge, they can
help site owners build web pages that meet user expectations and satisfy
their needs. To do this, they use the scent of information—the keywords
that people type into search engines.
By increasing the scent of information using the keywords in search Note On the web, a query
listings, site owners can encourage users to click their listings and maintain is typically defined as a
their site’s search engine visibility, meeting both business objectives and question or a request for
information. If you are
users’ expectations. The scent of information is a key way to convince speaking with a web search
web searchers to bite and to keep them hooked. expert, the word querying is
In this chapter, we’ll explore querying as a positive search behavior more accurate than the
word searching for describ-
and explain some of its negative perceptions. We will also go over the
ing the primary activity
main components of a web page’s search listing and learn where the associated with search
commercial web search engines are getting this information. engines (such as typing
words into a search box).

! DIFFERENT PERCEPTIONS OF QUERYING

BEHAVIOR
As mentioned previously, SEO professionals view querying as a positive
search behavior because people are communicating where they want
to go, what they want to read, and what they want to do. People do not
always know what content is available on prospective websites. They use
a web search engine to discover new content.
Additionally, people might remember great content they read or
used previously. They might not have bookmarked the web page. They
might not have printed or written the name of the site or its URL. So
they use a web search engine to re-find that desirable content.

Keywords are a user-generated scent of information.

Therefore, SEO professionals view querying behavior as a posi-
tive search behavior. From their perspective, query formulation is an
opportunity to present searchers with desired content. The users’ scent
of information is unmistakable—it consists of the keywords that people
type into search engines. In other words, keywords are a user-generated
scent of information.
Usability professionals, on the other hand, have different perceptions
of querying behavior. They often observe users’ frustration and confusion
with search engines and search engine results pages (SERPs). They also
observe site abandonment—points at which users leave a website—due to
20 CHAPTER 2: THE SCENT OF INFORMATION AND WEB SEARCH ENGINES

querying behavior. To them, querying behavior can signify a diminished,

or missing, scent of information.
So whose perception is more accurate? Let’s look at some of these
perceptions individually to see how these two groups differ, and to see
what they have in common.

Missing Information Scent

A site search, not to be confused with a web search, is a secondary naviga-
tional aide. During usability testing on a site, participants tend not to use
a site search engine (Figure 2.3) until they’re unable to find information
by browsing. If users do not see keywords in the web page, they resort
to using a site search engine to find what they want and to create the
missing information scent. Therefore, for usability professionals, querying
behavior is a negative search behavior—a failure of site navigation.

Figure 2.3 Users might use a site search engine if they are unable to find desired
content via browsing. Therefore, site querying is often viewed as a negative search
behavior in usability terms.

However, on a web search engine (Figure 2.4), querying is viewed

as a positive search behavior because searchers are communicating what
they desire.
Figure 2.4 Web search
engine experts view query-
ing as a positive search
behavior.

Irrelevant or Poor Search Results

Many web content creators have commercial motives for achieving top
web search engine positions, regardless of how relevant they are to the
user’s request. How many times have you Googled a person’s name and
 DIFFERENT PERCEPTIONS OF QUERYING BEHAVIOR 21

seen some rather odd search listings? How many times have you clicked
a link from Google and landed on a web page (Figure 2.5) that did not
have the information you wanted? And how many times have you aban-
doned a website because the information you found simply wasn’t want
you had in mind when you began your query?
Figure 2.5 The reaction you
don’t want after a searcher
clicks a link to your website
from Google.

Usability professionals commonly observe this user frustration and

confusion. Usability professionals also experience this frustration person-
ally when they use a commercial web search engine. From a usability
perspective, they do not recognize the critical balance between business
goals and user expectations. They see too much focus on business goals
at the expense of a positive user experience.
Unfortunately, some web content creators and SEO professionals
go to extreme measures to achieve top search engine positions. Search
engine spam—which is creating pages deliberately to trick the web
search engines into offering inappropriate, redundant, or poor-quality
search results—is still used as a search engine marketing strategy. As a
result, some substandard and irrelevant listings constantly appear at the
top of Google or any search engine results. These substandard listings Note Search engine spam
is still a common marketing
lead to a negative user experience for the search engine.
strategy. Google and the
However, not all SEO professionals are search engine spammers. other web search engines
Many SEO professionals are dedicated to meeting user expectations— are constantly discovering
delivering and maintaining an appropriate information scent, beginning and removing spam content.
Check out Part 5, “Best
with the search listings on a search engine results page and ending on Practices: Dos and Don’ts of
a website. These SEO professionals not only achieve top search engine Search Engine Optimization,”
positions, they also contribute positively to brand perception and brand in Search Engine Visibility
(New Riders, 2007), for
affinity by appropriately matching user queries to web page content, details about different types
resulting in more first-time visitors and more repeat visitors. of search engine spam.
22 CHAPTER 2: THE SCENT OF INFORMATION AND WEB SEARCH ENGINES

! HOW TO PROVIDE INFORMATION SCENT

IN SEARCH LISTINGS
A user-generated scent of information begins at the commercial web
search engines when searchers type their query words into the search
box. After users click the “Search” button, search engines validate the
users’ scent of information in search results by using term highlighting,
Note A web page’s search which is highlighting, or bolding, the matching query words in the listing
listing appears in the main summaries. Search engines highlight query words in these places to help
content area of a search searchers feel more confident that they are seeing the best, most relevant
engine results page.
listings. In Figure 2.6, you see term highlighting in both the (1) main
search listings and (2) the ads in a Google SERP for the keyword phrase
thyroid cancer symptoms.
Figure 2.6 Search engines
validate searchers’ scent of
information by highlighting
query words, or keywords, in
various places in the search
engine results page (SERP).

q
w

A good information scent has visible signs, which in the case of

search results pages is term highlighting. For each main search listing, if
query words are present, they are highlighted in the (1) title-tag content,
(2) meta-tag description or page snippet, and (3) URL (Figure 2.7).
Figure 2.7 An individual
search listing for the key-
q
word phrase thyroid cancer
symptoms. Keywords are e w
highlighted in the title tag,
page snippet, and URL.
 HOW TO PROVIDE INFORMATION SCENT IN SEARCH LISTINGS 23

Searchers select the link that they feel provides the best scent of
information. Each element (identified with a number in Figure 2.7) of a
web page’s search listing provides specific information scent. Let’s look
at each of these elements in detail.

Title Tags
An HTML title tag has a specific meaning to website design and search
engines. The text between the <title> and </title> tags, is the web page
title and is known as the title tag. It looks like:
<title>The Art Institute of Chicago</title>

On an actual web page in a web browser, title-tag content displays

at the top of a screen (Figure 2.8).
Title-tag content

Figure 2.8 The title-tag con-

tent displays on browsers at
the top of the screen.

Not only does the title-tag content validate and continue the users’ Note See Jared Spool’s
scent of information, it also serves other purposes in the search listing: report “Designing for the
Scent of Information” at
" Click-through (on the search listing). The title-tag content is
http://www.uie.com/reports/
the call to action in each search listing. The title tag is format- scent_of_information/.
ted as a text link. Text links work best if they are between 7
and 12 words long, or approximately 40–69 characters, which
is long enough to boost the chances of a searcher’s query
words appearing in a link, yet short enough to find the word
easily while scanning the search results.
" User confidence. If keywords are present in the title tag, it Note Google gets title-tag
increases user confidence. Searchers are more confident that content directly from each
search engines have delivered the best results when they see web page.
24 CHAPTER 2: THE SCENT OF INFORMATION AND WEB SEARCH ENGINES

their query words displayed as a hyperlink. Searchers also

believe clicking a keyword-focused link will lead them to
desired content.
" Relevancy. All the major web search engines evaluate keyword
use in title-tag content to calculate rankings.

Meta-Tag Description or Page Snippet

Note On-the-page content A search listing’s description typically contains around 155 characters.
refers to content that is Google gets the listing description from one of three places on a web
visible on a web browser page: meta-tag description, a single occurrence of on-the-page content,
between the <body> and
</body> tags in XHTML
or multiple occurrences of on-the-page content.
code. The Art Institute’s home page (Figure 2.8) only contains graphic
images. Google and the other web search engines have a difficult time
detecting what an image is and how to describe it. So they do not use text
formatted in graphic images for a search listing. Instead, in this example,
Google is displaying a portion of the meta-tag description as the search
listing’s description.
In source code, the meta-tag description looks like the following:
<meta name="description" content="The world-renowned Art
Note This meta-tag
description contains approx-
Institute of Chicago houses both a museum and school. The
imately 270 characters, but museum contains more than 300,000 works of art while nearly
Google only displays around 3,000 students are enrolled in various undergraduate and
155 characters in the listing. graduate degree programs at the School of the Art Institute
of Chicago." />

For example, if you look at the Google listing of The Art Institute
of Chicago’s home page (Figure 2.9), you see that Google is generating
the listing description from the meta-tag content.
Meta-tag
description
Figure 2.9 Search listing for
the query words art institute.

If a web page contains keywords in the on-the-page content area,

Google might generate a listing description from one part of that content,
as shown in Figures 2.10 and 2.11:
 HOW TO PROVIDE INFORMATION SCENT IN SEARCH LISTINGS 25

Snippet from page content

Figure 2.10 Search listing for
the query words pancreatic
cancer symptoms.

Figure 2.11 Landing page

from the search listing
shown in Figure 2.10. Google
is generating this listing
description from one snippet
of text from the on-the-
page content, even though
the page has a meta-tag
description.

Or, Google might generate a listing description from multiple places

on a web page. If you look at the text in the search listing (in Figure 2.12)
and compare it to the listing’s landing page, you can see that Google is
generating part of the listing description from two snippets (see Figure
2.13 on the next page) and the meta-tag description.
Here is the source code for the meta-tag description:
<meta name="Description" content="NCI's gateway for
information about pancreatic cancer." />

And here is the Google listing:

Snippet from page content
and meta-tag description
Figure 2.12 Search listing for
the query words pancreatic
cancer trials.
26 CHAPTER 2: THE SCENT OF INFORMATION AND WEB SEARCH ENGINES

Figure 2.13 Landing page

from the search listing
shown in Figure 2.12. For the
National Cancer Institute’s
Pancreatic Cancer page,
Google is generating part of
the listing description from
multiple snippets of text from
the on-the-page content.

Tip An important tip that Google reserves the right to change listing descriptions to make
warrants repeating: To them more useful for searchers. So no one knows exactly how a web
ensure your search listings page listing will appear in search results. Therefore, to ensure your search
always contain a strong
scent of information, key-
listings always contain a strong scent of information, keywords should
words should be present in be present in both on-the-page content and meta-tag descriptions.
both on-the-page content
and meta-tag descriptions
URL Structure (Web Address)
In a search engine results page, the URL can be crucial for encourag-
ing searchers to click the link to your web page because it can reinforce
the scent of information. For example, let’s look at the URL shown in
Figure 2.13: http://www.cancer.gov/cancertopics/types/pancreatic
When we showed this URL, and only this URL, to usability test
participants and asked them what they believed this web page is about,
the answers were consistent:
 HOW TO PROVIDE INFORMATION SCENT IN SEARCH LISTINGS 27

" “It’s a web page about pancreatic cancer topics.”

" “It’s about pancreatic cancer.”
" “I can read information about the types of pancreatic cancer.”
All participants correctly stated that the URL would lead them to
a web page that contains information about pancreatic cancer. When
we asked participants to come up with other web addresses that might
help them understand the content that is available on that page, they
came up with:
" www.cancer.gov/types/pancreatic.html
" www.cancer.gov/types/pancreas.html
" www.cancer.gov/cancer/pancreas.html
" www.cancer.gov/cancer/pancreatic.html
" www.cancer.gov/topics/pancreaticcancer.html
" www.cancer.gov/topics/pancreatic-cancer.html
" www.cancer.gov/cancertopics/pancreaticcancer.html
" www.cancer.gov/cancertopics/pancreatic-cancer.html
" www.cancer.gov/cancer/types/pancreatic.html
" www.cancer.gov/cancer/types/pancreas.html
" www.cancer.gov/pancreaticcancer.html
" www.cancer.gov/pancreatic-cancer.html
" www.cancer.gov/pancreatic.html
" www.cancer.gov/pancreas.html

All of these URLs provide a strong information scent because they Tip From a usability per-
contain important keywords. Additionally, some of these URLs are easier spective, fewer characters
to read than others because they contain fewer characters. in a URL makes it easier
to read.
When we showed this next URL to test participants and asked them
what they believed this web page was about
www.cancer.gov/cancertopics.aspx?ArticleId=1080
all test participants gave the same answer, “It’s an article about cancer
topic(s).” There was no mention of the keyword phrase pancreatic cancer.
When we asked participants how confident they were that this URL would
lead to a page about pancreatic cancer, most of them had doubts.
The following URL structure generated even less user confidence:
www.cancer.gov/cancertopics.aspx?CatId=1080&ArticleID=234
28 CHAPTER 2: THE SCENT OF INFORMATION AND WEB SEARCH ENGINES

User confidence decreases when users do not see keywords in URLs.

In an ideal situation, a web address provides useful information about
a page’s content.
Let’s look at some parts of a sample URL where important keywords
can be placed:
http://www.domain.com/subroot/path/file
" Subroot: A domain name followed by a single directory. In the
following URL, the subroot is brand.
http://www.company.com/brand/
" Path: A domain name followed by an arbitrarily deep path but
not ending in the file name. In the following URL, the path
name is refrigerators.
http://www.company.com/brand/refrigerators/
" File: Anything ending in a file name other than index.html. In
the following URL, the file name is refrigerator-name.html.
http://www.company.com/brand/refrigerators/
refrigerator-name.html
Every portion of a search listing is important for increased user
confidence and clickability. If keywords are not present in a search listing,
the information scent is very poor, and searchers are less likely to click
the link in that listing. On the other hand, if there is too much bold text
in a search listing, the listing becomes too hard to read—a big turn-off.
Therefore, searchers are less likely to click that keyword-stuffed listing.

Every portion of a search listing is important for

increased user confidence and clickability.
SEO professionals and site owners should maximize each of the
three search listing elements for the desired impact. As you will see in
forthcoming chapters, some portions of an individual search listing are
more important than others. Which portion of a search listing is important
for searchers who want to go to a specific website? Which portion of a
search listing is important for searchers who want to do something?
Site owners should understand how the searcher’s intent determines
the importance of each search listing element.
 3 TYPES OF WEB SEARCH ENGINE QUERIES 29

! 3 TYPES OF WEB SEARCH ENGINE QUERIES

Since 2002, web search queries have been classified based on searcher intent.
(See Andrei Broder, “A taxonomy of web search,” SIGIR Forum, 36(2) Note The article “A
(2002): 3-10.) These three query types are called navigational, informational, taxonomy of web search”
and transactional. By understanding the three types of queries, SEO pro- is available online at:
www.sigir.org/forum/
fessionals can work with the search engines to serve up the right content F2002/broder.pdf.
to the right user for a specific query. The better you can anticipate the
user’s query intent, the better chance you have of meeting user goals and
driving business via your website.
" Navigational queries are ones in which the searcher wants to
go to a specific website, or a specific web page (usually a home
page) on a specific website. Approximately 10 to 26 percent of
web search queries are navigational.
" Informational queries are ones in which the searcher wishes
to read or view more information about a topic. Informational
queries are the most common type of search engine query,
comprising anywhere between 48 and 80 percent of web
search queries.
" Transactional queries are ones in which the searcher wishes
to perform some interaction on the web, aside from reading.
Approximately 10 to 24 percent of web search queries are
transactional.

By satisfying navigational, informational, and

transactional query needs, website owners can get a
higher return on investment (ROI) than by satisfying
transactional queries alone.
To ensure your website is successful, you should make sure your
website accommodates all query types. Users might perform multiple
types of searches to find your website. For example, searchers might
want to monitor a bid on an item for sale on eBay. Before searchers can
monitor a bid (informational), they must first go to the eBay website
(navigational).
By satisfying navigational, informational, and transactional query
needs, website owners can get a higher return on investment (ROI)
than by satisfying transactional queries alone. Therefore, it’ll pay off if
you address all query types when building and writing content for your
30 CHAPTER 2: THE SCENT OF INFORMATION AND WEB SEARCH ENGINES

websites. The next three chapters will cover the three query types in
detail, but let’s take a quick look at what user behaviors classify queries
into each of these categories.

Navigational Queries
As shown in Figure 2.14, searchers have a variety of options to accomplish
the task of navigating to eBay. They can type the URL of eBay’s home
page, www.ebay.com, into a web browser’s address bar (1). However,
a quicker way for users not accustomed to using the URL address bar
might be to type in the word ebay into search box on the Google home
page (2), or to type ebay.com into the Google search box within the web
browser (3).
Figure 2.14 Searchers
have some options to get
to eBay’s home page. If a
q
searcher uses a commercial
web search engine to navi-
gate to a site, the query is a
navigational query, and the
e
keywords used to arrive at
the web page are naviga-
tional keywords. w

Transactional Queries
Many search engine marketers tend to spend a great deal of time focusing
on transactional queries because they want to target searchers who are
ready to buy. On the surface, focusing on a specific point in the buying
cycle might seem like a great way to save time and money. However, this
strategy may cost website owners prospects and lifetime customers.

Informational Queries
Before people make a buying decision, they often begin a web search ses-
sion with an informational query to establish a frame of reference: Which
companies offer the products? How much do these products generally
cost? Can shoppers get these products only online, or can they quickly
drive to a nearby physical location to purchase this product? If there is a
nearby (offline) physical store, how can they get there?
 CONCLUSION 31

! CONCLUSION
Key points in this chapter:
" Savvy marketers should understand how people are searching
and why people are searching in order to formulate the best
search engine marketing strategies.
" When people use the commercial web search engines, they are
trying to communicate where they want to go (navigational),
what they want to read (informational), and what they want to
do (transactional).
" By increasing the scent of information in search listings, site
owners can encourage users to click their listings and maintain
their site’s search engine visibility, meeting both user expecta-
tions and business objectives.
" In web search, the scent of information consists of the key-
words that people type into search engines.
" Search engine optimization professionals view querying as a
positive search behavior.
" Usability professionals sometimes view querying as a nega-
tive search behavior because it can indicate a missing scent of
information.
" Web searchers want their information scents validated in
search engine results pages (SERPs). Web search engines, such
as Google, get the information to display in search listings
from your web pages’ content.
" Every portion of a search listing—title tag, meta-tag content or
page snippet, and URL—is important for the scent of informa-
tion, user confidence, and clickability.
" Web search queries are classified based on searcher intent:
navigational (where can I go?), informational (what can I read
or learn?), and transactional (what can I do?).
" By satisfying navigational, informational, and transactional
query needs, website owners can get a higher ROI than by
satisfying transactional queries alone.
This page intentionally left blank
CHAPTER 3
NAVIGATIONAL SEARCHES—
WHERE CAN I GO?
With a navigational query, a web searcher’s main goal is to go directly
to a website’s home page or to a specific website page. The term
navigational query refers to people using commercial web search engines
to go (navigate) to a specific website page.

Navigational queries are important to site owners because searchers

who perform them genuinely wish to visit your site. Web searchers
might want to visit your site to find specific information (such as your
company’s or organization’s phone number) or to perform a transaction
(buy, enroll, log in, and so forth). Regardless of the searcher’s final goal,
many query sessions begin with navigational intent.

Navigational queries are more common than you might imagine, compris-
ing between 10 percent and 26 percent of web search engine queries.
Clearly, a considerable number of web searchers want to go to specific
websites. Let’s look at how to determine that a search is navigational
and how to ensure your site meets searchers' expectations.
34 CHAPTER 3: NAVIGATIONAL SEARCHES—WHERE CAN I GO?

! NAVIGATIONAL INTENT
Searchers formulate navigational queries because they might not know
or remember the URL of the site. They might type the abbreviation of
a company’s name because the official company name is long and easily
misspelled. Searchers might only remember part of a URL. Or searchers
might simply find it easier to type in the company name in a search box
rather than an organization’s full URL in the address bar of a web browser
(Figure 3.1). This occurs more frequently than you might imagine.
Figure 3.1 During usability
tests, you often see users
typing the word “Google”
to arrive at Google’s home
page from Yahoo!, and vice
versa. When we ask usability
test participants why they
use a Yahoo! search to go to
Google’s home page, they
Typically, navigational queries occur more frequently on mobile
usually say that it is easier to phones and PDAs (personal digital assistants). These devices have much
just type in a single keyword smaller screens than desktop and laptop computers. Therefore, searchers
rather than the full URL.
have less space to type in keywords, and less space to view corresponding
Interestingly, many search-
ers are not even aware that search results. PDAs and mobile phones also have smaller keyboards.
they can type in a URL in a Therefore, searchers tend to type in fewer keywords with fewer characters
browser’s address bar. when they use a mobile search engine.
Additionally, the address bar is not always easily recognizable on
a mobile device. Many searchers do not even know where to type a
known web address. So they type in all or part of a URL in an area of
the screen that is familiar to them—the search field (Figure 3.2). In a
recent studies, 17 percent of English-language Google mobile queries
were for URLs; Microsoft reported a smaller percentage (7.54 percent);
Yahoo! reported an even smaller one (5 percent). For this and more
information about querying research, see “A Large Scale Study of Wire-
less Search Behavior: Google Mobile Search” by Kumvar and Baluja at
http://www1.cs.columbia.edu/~mkamvar/publications/CHI_06.
pdf; “On Mobile User Behaviour Patterns” by Milan Vojnovic at
http://research.microsoft.com/apps/pubs/default.aspx?id=70532; and
“Deciphering Mobile Search Patterns: A Study of Yahoo! Mobile Search
Queries” by Yi et al at http://www2008.org/papers/pdf/fp846-yi.pdf.
Figure 3.2 In this Apple Web search engine companies usability test continually to improve
iPhone figure, the searcher the searcher experience. To determine whether people are using web
is using the default search
search to navigate, search engines analyze keyword combinations and
engine (Google) instead of
the unlabeled address bar corresponding clicks; which search listings users click on most frequently;
to go to eBay’s home page. and whether the search session ends once users arrive at destination sites.
 NAVIGATIONAL INTENT 35

If the word combinations or characters clearly indicate navigational intent, Note Currently, search
search engines tend to place specific listings in positions one through three engines do not use the num-
of the search engine results page (SERP). Searchers rarely look past the ber of clicks (click popular-
ity) to determine search
first two or three listings for queries with navigational intent. Let’s take engine rankings, except in
a look at some examples of navigational queries. the area of personalized
search. With personalized
search, search engines are
Example 1: Home Pages able to tailor search listings
for each user based on past
Suppose a searcher wants to go to the National Cancer Institute’s home
search history, which sites
page. Instead of typing in the words National Cancer Institute in Google, the user clicks on most
she types in the abbreviation for the organization (NCI) because it is often, bookmarks, and other
easier and faster to type than the full name. She also types in the top- factors. The data from per-
sonalized search can help
level domain extension .gov because she remembers that the National search engines determine
Cancer Institute is a government website. whether a searcher’s intent
She uses Google search instead of the address bar because Google is is navigational, informa-
tional, or transactional.
her computer’s default home page, and, in the past, she successfully arrived
at the National Cancer Institute’s website using this search strategy. Figure
3.3 shows Google’s search engine results page from this query.
Figure 3.3 Because the
searcher typed in an
abbreviation and a domain
extension in the search box,
Google determined that the
searcher’s intent was most
likely to be navigational.
36 CHAPTER 3: NAVIGATIONAL SEARCHES—WHERE CAN I GO?

The query entry communicates strong navigational intent. In all

likelihood, the searcher probably wants to go to a government website
(.gov) with the abbreviation NCI. The National Cancer Institute’s canonical
domain name is cancer.gov, and Google correctly forwarded the searcher
to the appropriate website’s home page URL.
Query entries are considered to be URLs if they begin with http or
www, or if they contain .com, .org, .edu, .net, or any other common top-
level domain (TLD) extension.
What if a searcher only remembered part of the organization’s name
but still wanted to go to that specific website? A searcher might type in
part of the organization’s name, as shown in Figure 3.4:

Figure 3.4 Since the searcher did not remember whether the official organization
name was the National Cancer Institute or the American Cancer Institute, she typed
in the part of the name that she remembered, hoping that Google would deliver search
results that would lead her to the right website.
 NAVIGATIONAL INTENT 37

CANONICAL DOMAIN NAMES

A canonical domain name is the standard, or authoritative, domain name that you wish site visi-
tors to see. A canonical domain should be reduced to the simplest and most vital form possible
without being overly general.
For example, searchers can go to the National Cancer Institute’s home page via the following
URLs:
" http://www.cancer.gov
" http://www.nci.gov
" http://nci.nih.gov

In Figures 3.3 and 3.4, Google has selected the URL www.cancer.gov as the most appropriate
URL for the National Cancer Institute’s home page. For more information about canonicalization,
visit www.mattcutts.com/blog/seo-advice-url-canonicalization/.

To accommodate searchers who have navigational intent, site owners

should be proactive and purchase multiple domain names, not only for
their official company name but also for abbreviations of their company
name, established brands, and trademarked terms. Consider purchasing
domain names with common misspellings of any of these domains.
Some unscrupulous web marketers count on domain misspellings to
trap unwary users.

Example 2: Login Pages

One of the most interesting aspects of website usability is differ-
entiating what people say from what people do. During usability
tests, usability professionals encourage test participants to think
out loud to help them determine why participants make specific
choices on a web page, such as which links to click on. Usability
professionals also measure and observe actual user behavior. All
too often, people say they are going to do something (click on a
link) but then perform an action that you might not expect.
In the web search industry, users saying one thing and doing
another happens quite frequently. Look at Figure 3.5. What
type of keywords do you believe this searcher will generate? Do
you believe the keywords will be navigational, informational,
or transactional?
On the surface, the searcher’s intent is both informational
Figure 3.5 What is this user’s intent?
(because he wants to see how many frequent flyer miles he has Navigational, informational, or
accumulated) and transactional (because he will probably have transactional?
38 CHAPTER 3: NAVIGATIONAL SEARCHES—WHERE CAN I GO?

to log in to his account in order to acquire this information). However,

this particular searcher generated the following query and was given
these search results (Figure 3.6):

Figure 3.6 What the searcher said and what the searcher did were quite different.

Note Details for keyword Even though the searcher stated that he wanted to check his frequent
research can be found in flyer miles, his web search query was only one word, united. Before the
Part 2 of Search Engine searcher could check the number of miles he has accumulated, and before
Visibility, 2nd Edition, by
Shari Thurow (New Riders
he could log in to his account, he had to first go to the website where
Publishing, 2007). he could log in (in this case, the United Airlines website). He used a web
search engine to navigate to a specific site. Single-word queries might be
navigational if the organization’s brand is very well known. In fact, one-
or two-word queries are a strong indication of navigational intent.
Search engine marketing (SEM) professionals regularly research
keyword phrases to determine searchers’ intent. Keyword research
tools (available online) help SEM professionals determine how people
search and the approximate volume of keywords used per month. Data
from web analytics software and site search engines can also show how
searchers are formulating queries.
 NAVIGATIONAL INTENT 39

However, SEM professionals must look beyond this data and look
at the big picture. In this example, if SEM professionals only focused on
how many times per month people searched for the word united, they
would not understand why searchers formulate that query. They might
assume that the searchers’ only goal was to go to the United Airlines
home page.

LEARN THE KEYWORDS IN YOUR USERS’ MENTAL MODELS

During usability testing, ask test participants the simple question, “How
would you look for this information on the web?” Their answers and
actions will often reveal their intentions and the steps to complete their
desired tasks. You can see query formulation firsthand if they use the
web search engines to complete tasks. You can also witness frustration
with terminology and labeling within search listings and corresponding
landing pages. Does the content of landing pages match the searchers’
mental models?
Keyword data alone does not reveal searchers’ underlying motivations
for formulating search queries. Keyword data, combined with usability
test data, can reveal searchers’ motivations.

Example 3: Category and Channel Pages

Sometimes searchers want to navigate to a specific page within a specific Note A channel page is a
website. They do not want to go to the home page and browse the site top-level category page.
to find desired information, especially if their desired information is Sometimes, a channel page
is referred to as a depart-
buried deep within a site’s information architecture. They want to go ment page. To illustrate, the
directly to the “right” page. Sometimes, the “right” page is a category vertical hierarchy of a web-
or a channel page. site might look like: Home >
Channel > Category >
For example, let’s use the Apple website. Searchers might want to
Subcategory > Content.
go to Apple’s website for a variety of reasons. They might want to buy
something, download some music, or get technical support.
Suppose a searcher wants to see the variety of iPods that are available
before deciding to buy one. He might want to see the cool stuff he can
download on the different types of iPods before purchasing. He periodi-
cally might visit the Apple website because he knows he can download
cool stuff directly from the site. In other words, this searcher could very
easily become a repeat visitor and lifetime customer.
40 CHAPTER 3: NAVIGATIONAL SEARCHES—WHERE CAN I GO?

To find the content on the web, the searcher might type in the key-
word phrase apple ipod into Google and get the search results shown
in Figure 3.7:

Figure 3.7 Google SERP for the query words apple ipod.

Google tried to deliver the most appropriate web listings based on

these query words. Google has interpreted that the searcher wishes to
see information about iPods on Apple’s official website. A channel page
(iPod + iTunes) is the top search listing.
By using the keyword “apple” as part of the search query, the searcher
might be communicating that he wishes to view desired content on
Apple’s official website. In Google’s main search results in Figure 3.7
you see that Google has accommodated a navigational query for Apple
ipod in the number 1 and number 2 positions, presenting a channel page
in position 1, and a category page in position 2. Figure 3.8 shows the
landing page for the listing in the number 1 position.
 NAVIGATIONAL INTENT 41

Figure 3.8 Landing page after clicking the top search listing link shown in Figure 3.7,
the iPod + iTunes channel page.

Since navigational queries can lead to transactions

(downloads, purchases, and so on), search engine
marketing professionals should test transactional
querying by purchasing advertisements for some
navigational query words.
If the searcher wants to see a list of available iPods, this is a very
appropriate landing page. The searcher can easily view the variety of
available iPod types (iPod shuffle, iPod nano, iPod classic, iPod touch)
immediately at the top of the screen. Each graphic image is labeled with
appropriate text. The folks at Apple have optimized this channel page
quite effectively for this type of search query. The highlighted naviga-
tion button at the top of the screen communicates that visitors are in
the iPod + iTunes section. The available iPods are shown right away at
the top of the screen, and the page identifiers (title tags, page content,
headings, URL or web address) communicate it as well. The “you are
here” cues, which provide a sense of place, and the information scents
are very clear.
Since navigational queries can lead to transactions (downloads,
purchases, and so on), search engine marketing professionals should
42 CHAPTER 3: NAVIGATIONAL SEARCHES—WHERE CAN I GO?

test transactional querying by purchasing advertisements for some

navigational query words. Notice that in the search results page shown
in Figure 3.7, Apple has purchased advertising space for a particular type
of iPod. If searchers click on that advertisement, they are not delivered
to the same landing page shown in Figure 3.8. Instead, they are delivered
to a specific page that accommodates transactional intent. This is clear
on the landing page because pricing is available on this advertisement’s
landing page (Figure 3.9) that is not available on the channel page.

Figure 3.9 A searcher who is ready to make a purchase will want to see specific
prices or a price range.

! SEARCH LISTINGS AND NAVIGATIONAL

QUERIES
The goal of commercial web search engines is to deliver searchers to
their desired content as quickly and as easily as possible. Likewise, search-
ers’ main desire is to leave a web search engine to arrive at their desired
website (or web page) as quickly as possible. With navigational queries,
searchers do not have the patience to explore search results. In their eyes,
a query with navigational intent has a clear, right answer.
Searchers who perform navigational queries rarely look past the
first three positions, and they are not paying much attention to the
snippets in each search listing. Instead, they focus their attention on the
link, which is taken from a web page’s title tag, and the URL as shown
in Figure 3.10:

q
w
Figure 3.10 For a navigational query, the two most important items in a search listing
are (1) the title-tag content, and (2) the URL.
 SEARCH LISTINGS AND NAVIGATIONAL QUERIES 43

Once searchers determine that the web address is the right one, they
click on the hypertext link in the search listing. They do not return to the
search engine but rather stay on the site they wished to navigate to.
The URL is very important for navigational queries. If a URL is too
long, searchers tend to look past it for listings with shorter URLs. For
example, let’s compare the following URLs.
" www.companyname.com/New_York_City/New_York_
City-Restaurants/New_York_City-Restaurants-Thai/
Thai-Restaurants.html (113 characters)
" www.companyname.com/New-York-City/Restaurants/
thai.html (56 characters)
Even if the first URL were in the number 1 position, searchers Note See Marketing-
would focus their attention on the listing in the number 2 position. So Sherpa’s Search Marketing
it is best to keep URLs as short and descriptive as possible. For naviga- Guide 2008 for more details
about URL length.
tional queries, a good rule of thumb is to keep URL length to less than
65 characters, if possible.
Remember that some people who type in navigational keywords are
likely to be repeat visitors. In fact, a recent study from Google U.K. and
comScore revealed that online travel buyers change the type of keywords
as they move along the path to purchase. Of all the consumers sampled
who made a final purchase, 29 percent started with a non-branded key-
word phrase but ended with a branded keyword phrase. The branded
keyword phrase was likely an indication of navigational intent. Web
searchers want to go back to sites they have previously visited.
Google accommodates navigational queries differently for U.K.–based
search listings and U.S.–based search listings. Soccer is not nearly as popular
in the U.S. as it is in the U.K. It is unlikely that U.S.–based searchers who
use the keyword united want to go to the Manchester United website
(Figure 3.11 on the next page). However, in other areas of the world, this
site is exactly where searchers wish to go (Figure 3.12 on the next page).
Notice how Google is keeping the top positions in the U.K.–based search
results to accommodate a keyword with navigational intent.
If the Manchester United site owners want search engine visibility at
the top of Google U.S. search listings, they could purchase Google ads for
this keyword, to appear specifically on Google U.S. search results. If the
return on investment (ROI) for this navigational keyword is high, then it
is worth it for them to maintain the ad as long as it continues to deliver
ROI. However, if the commercial web search engines are delivering the
site’s home page for navigational searches, it might not be necessary
to purchase advertising for the organization name, thus saving search
advertising expenses.
44 CHAPTER 3: NAVIGATIONAL SEARCHES—WHERE CAN I GO?

Figure 3.11 Google U.K. (United Kingdom) search results.

Figure 3.12 Google U.S. search results. Notice that the Manchester United website is not listed in the top three
positions.
 OPTIMIZING FOR NAVIGATIONAL QUERIES 45

! NAVIGATIONAL QUERY INDICATORS

Your web analytics data can help you determine which keywords show
navigational intent. Some navigational query indicators include:
" Portions of URLs (such as http and www) Note At the beginning of
2008, Google announced
" Common domain suffixes (.com, .org, .edu, .net, and so forth) that the average number
of words per query has
" Company, business, and brand names
increased to four words
" Keyword phrases comprised of one or two words per query.

! OPTIMIZING FOR NAVIGATIONAL QUERIES

Here are some general guidelines for optimizing your website to accom-
modate navigational queries:
" Make it easy for the search engines to put your site’s home
page in top positions for your company name. Your site’s
home page and “About Us” pages should be optimized for your
company’s official name. Optimize more than just your home
page for informational and transactional keywords.
" The URL is very important for navigational queries. Keep URL
length to less than 65 characters, whenever possible, for pages
that typically satisfy searchers’ navigational intent.
" For pages that you know will be important for navigational
intent, keep the meta-tag description short for easy reading,
but descriptive and explanatory (Figure 3.13).

Figures 3.13 The National Cancer

Institute did an outstanding job of
writing a short meta-tag descrip-
tion for its home page, around 105
characters.
46 CHAPTER 3: NAVIGATIONAL SEARCHES—WHERE CAN I GO?

" Be realistic about search engine positioning. If you periodi-

cally monitor how search listings appear for your targeted
keyword phrases, you will learn which websites and individual
web pages stay in the top positions for navigational queries. It
might be unrealistic to expect your listing to be in the first posi-
tion if it does not clearly meet searchers’ navigational intent.
" Buy domains that have your brands, company name, and trade-
marks in them. If it is not possible due to budgetary limitations,
then at least put those items in other places in a URL using the
“URL Structure (Web Address)” section in Chapter 2, “The
Scent of Information and Web Search Engines,” as a guideline.

! CONCLUSION
Key points in this chapter:
" Navigational queries are important to website owners because
searchers who perform them genuinely wish to visit your site.
" When a searcher types in all or portions of a URL as a key-
word, it is a strong indication that the searcher’s intent is
navigational.
" To accommodate searchers who have navigational intent, site
owners should be proactive and purchase domain names for
their official organization or company name, abbreviations
of their company name, established brands, and trademarked
terms.
" If your organization does not have an easily recognized name
or abbreviation, then ensure the most appropriate ads and cor-
responding landing pages appear for company-name searches.
Apple does an outstanding job at appropriate page delivery.
" The URL structure (web address) in a search listing is very
important for searchers who wish to go to a specific website.
" Keep URL length to less than 65 characters, whenever possible,
for pages that typically satisfy searchers’ navigational intent.
" In the event that the commercial web search engines are deliv-
ering your site’s home page for navigational searches, it might
not be necessary to purchase advertising for your company
name, thus saving search advertising expenses.
" The description or snippet is not as important as the URL
structure for navigational searches. Keep the meta-tag descrip-
tion short (around 100 to 150 characters) for pages that typically
satisfy searchers’ navigational intent.
CHAPTER 4
INFORMATIONAL
SEARCHES—
WHAT CAN I LEARN?
An informational query is one in which the searcher’s goal is obtaining
information about a general or specific topic. Sometimes a searcher
wants quick information, such as how to do something. And sometimes
a searcher wants to delve deeper into a topic, and is willing to do con-
siderable research before making a business transaction.

Informational queries are the most common type of web search query,
comprising between 48 percent and 80 percent of web searches. What
do these numbers mean to website owners? If a site owner wants to get
and maintain long-term search engine visibility, then satisfying infor-
mational searches is crucial. All websites should contain informational
pages that provide specific content of user interest. Let’s look at how
to determine that a search is informational and how to ensure your site
meets searchers’ expectations.
48 CHAPTER 4: INFORMATIONAL SEARCHES—WHAT CAN I LEARN?

! INFORMATIONAL INTENT
Many web search sessions begin with an informational query and end
with a transaction. People will not purchase your products and services
if you do not provide them with enough content to make an informed
decision, especially for high-ticket items.
For example, on an ecommerce site, searchers might be interested
in locating something in the real world (such as a physical location of
a store), or they might be interested in seeing a list of available items.
Many product or shopping queries have a “locate” goal because online
shoppers want to know where they can purchase a desired product. In
fact, the plural form of a targeted keyword can be a strong indication
that searchers wish to view a list of available products.
Category pages are a type of information page because they contain
lists of available items. Figure 4.1 shows a standard format for a category
page. Another category page format is an annotated list. The annotated
list format can provide a stronger information scent because the anno-
tation text can contain additional keywords as well, reinforcing existing
keywords on the page (Figure 4.2).
Figure 4.1 A category page
is a type of information
page because it contains a
list of products that search-
ers might wish to view and
compare before making a
purchase. For high-ticket
items such as furniture, cars,
and plane tickets, searchers
typically visit many websites
to gather information before
buying.
 INFORMATIONAL INTENT 49

Figure 4.2 A topical category page on the National Cancer Institute website. The
annotated list format provides many opportunities for keyword placement.

Proper keyword selection and placement can make an annotated

category page appear more topically focused to both searchers and search
engines. In other words, the page format shown in Figure 4.2 is an effec-
tive format for people who query and for people who browse.
Let’s take a look at some more examples of informational queries.

Example 1: Quick Facts

Search sessions can be long or short when a searcher wants information
about a topic. Of course, if searchers want a quick fact, they prefer to
make the search session as short as possible.
In Figure 4.3 on the next page, our searcher is traveling to London
and wants to see what the weather is like. Notice that the top three search
listings are not official company sites. Even Weather.com’s listing is not
in the top three results, meaning that the search engine has determined
that the keyword phrase (weather in London) does not indicate a naviga-
tional goal. It is most likely an informational goal. The searcher is using
Google to find out about the current London weather, not to navigate
to a specific website.
50 CHAPTER 4: INFORMATIONAL SEARCHES—WHAT CAN I LEARN?

Figure 4.3 A search engine

results page for an infor-
mational query. Notice that
the top positions are not
dedicated to navigational
queries. These search list-
ings strongly indicate that
this keyword phrase shows
informational intent.

Searchers do not always

show informational intent with
their keyword selection. For
example, in Figure 4.4, what type
of keywords do you believe this
searcher, a Manchester United
soccer team fan located in Eng-
land, will generate? Do you
believe the keywords will be
navigational, informational, or
transactional?
On the surface, the search-
er’s intent appears to be informa-
tional. He can view the match
schedule on multiple websites.
He does not have to go to the offi-
cial Manchester United website
Figure 4.4 Is this searcher’s intent naviga-
to view the match schedule. tional, informational, or transactional?
However, this particular
searcher typed the word united in Google and received the search results
in Figure 4.5:
 INFORMATIONAL INTENT 51

Figure 4.5 What the

Manchester United fan said
and what he did were quite
different.

Even though the searcher stated that he wanted to see the match
schedule, his web search query was for only one word, united. His
actions did not match his words. He clearly wanted to go to the official
Manchester United website to view the match schedule.

Field studies and exploratory usability tests often

reveal searchers’ informational goals above and beyond
what can be inferred from web analytics data and
keyword research tools.
Moral of the story? To see such discrepancy between words and
actions, it is very important to have some one-on-one, face-to-face
contact with searchers. Observe what users do as well as what they say.
Web analytics data will only show that this type of searcher wants to go
to the official Manchester United website. The data does not show this
searcher’s informational goal. Field studies and exploratory usability tests
often reveal searchers’ informational goals above and beyond what can
be inferred from web analytics data and keyword research tools.

Example 2: Questions and Answers

One of the strongest indicators of informational intent is a keyword
phrase that is formatted as a question, such as:
" How do I get driving directions to your office?
" Where are the nearest parking garages near this restaurant [name]?
" What is the capital of Nigeria?
" What are your hours of operation?
" What does the abbreviation NMR mean?
52 CHAPTER 4: INFORMATIONAL SEARCHES—WHAT CAN I LEARN?

Question-formatted keyword phrases present a great opportunity

for multiple types of search engine optimization, particularly how-to
keyword phrases. Searchers might want to see graphic image instructions
and diagrams that they can eventually print. Searchers might also want to
watch a video that shows them how to do a procedure. One way to see
how search engines determine the best results for question-and-answer
queries is to monitor search engine results pages to see what types of
listings appear on the first few pages (Figure 4.6).

Note You can find details Figure 4.6 For many informational searches, there are multiple ways that your listing
for graphic image and video can appear on the first page of search results. Like web page listings, graphic image
optimization in Part 3 of files and video files should be optimized in order for them to appear in the top search
Search Engine Visibility, 2nd results. How-to files present a great opportunity to satisfy informational queries.
Edition, by Shari Thurow
(New Riders Publishing, A natural place to implement keyword phrases in a question-answer
2007). format is a Frequently Asked Questions (FAQs) section of a website.
However, some usability professionals do not recommend having an
FAQs section. They say that most users should be able to have all their
informational needs addressed on main content pages and category pages,
making the FAQs section unnecessary.
Nevertheless, users have questions about products, services, delivery
options, hours of operation, guarantees, and so on. And users often format
their search queries as questions. Can you imagine formatting a product
page on an ecommerce site in a question-answer, question-answer for-
mat? That format would probably confuse users. And a question-answer
 INFORMATIONAL INTENT 53

format isn’t really an appropriate format for a product page. However, Tip Other names for
an FAQs section of a website is a legitimate and accepted way to satisfy question-answer pages
include Help and Customer
question-formatted informational queries.
Service. If your target
In addition, a customer service or FAQs section is often more flexible audience is likely to not
than other sections of a website. If your customer service staff, or others understand the FAQs
who have direct one-on-one contact with your users, determines that abbreviation for Frequently
Asked Questions, use Help
customers and prospects keep asking the same question over and over or Customer Service as an
again, adding that question to an FAQs page is an easy process. alternative navigation label.

Example 3: Lists
When searchers wish to delve deeper into a topic of interest, they often
type keyword phrases that they hope will provide them with a list of
suggestions and a frame of reference for further research. Sites that
provide a topical list of resources, particularly annotated lists that easily
validate an information scent, often appear at the top of search engine
listings (Figure 4.7). Sections of a website that typically give a topical list
of resources include Links, Resources, and Tips sections.

Figure 4.7 An annotated links page on a business-to-business (B2B) website. Notice

Note On a web page,
that the page is a long one, and the site owners have clearly indicated that content is
available below the fold via the bulleted hypertext links. The scent of information above the fold refers to con-
is maintained above the fold as well as below the fold. In addition, the repetition of tent that is located near the
important keywords helps with search engine visibility. top of the page, content that
can be viewed in a browser
without scrolling.
54 CHAPTER 4: INFORMATIONAL SEARCHES—WHAT CAN I LEARN?

Tip An important tip that Another type of web page that naturally contains a list is a category
warrants repeating about page, as illustrated at the beginning of this chapter in Figures 4.1 and
many category pages from 4.2. One thing to remember about many category pages from a search
a search usability perspec-
tive: The plural form of a
usability perspective: The plural form of a query word should naturally
query word should naturally occur on this type of page within the title tag, heading, locational bread-
occur on this type of page crumb links, and introductory paragraph (if used).
within the title tag, heading,
Individual product pages should emphasize the singular form of a
locational breadcrumb links,
and introductory paragraph word, with a few exceptions. One exception: If a product page contains
(if used). links to other product pages containing the same word, providing a strong
scent of information, then searchers find it acceptable to see both the
singular and plural form of a word on a page (Figure 4.8).

Figure 4.8 This product page links to other product pages containing the word saw.
 SEARCH LISTINGS AND INFORMATIONAL QUERIES 55

Another exception is a product page that offers a set of items, such

as a set of tools. In this instance, the plural form of a word is a natural
occurrence. However, a product page that emphasizes a set of items does
not necessarily provide a good frame of reference for further research.
Therefore, a product page for a set of items will not likely rank well for
informational queries where searchers want a list of items.

! SEARCH LISTINGS AND INFORMATIONAL

QUERIES
Search engine results pages communicate a lot of information in individual
search listings, including what search engines determine to be searcher
intent. For example, suppose a searcher wishes to buy a refrigerator, a
high-ticket item. Before going to the store and purchasing a refrigerator,
he wishes to see the types of fridges available since he has not had to
purchase a fridge for a very long time. To establish a frame of reference,
he only types in a single keyword, refrigerators, as shown in Figure 4.9:

Figure 4.9 Search engine results page for the query word refrigerators.
56 CHAPTER 4: INFORMATIONAL SEARCHES—WHAT CAN I LEARN?

What might the arrangement and placement of these search listings

communicate to us? First, we see a wide variety of listing types supporting
varied intentions among various searchers. In the search listings are:
" Product review pages
" A government website
" Category pages
" Shopping search listings
" Wikipedia

Whenever you see a Wikipedia listing appear on the first page of

web search results, it is a strong indication that the search engine has
determined that the keyword (or keyword phrase) is an informational
query. Wikipedia is an informational website. In contrast, online retail
sites are primarily transactional sites.

By focusing only on transactional queries and ignoring

informational queries, many site owners are probably
losing initial sales and potential lifetime customers.
Therefore, whenever a website owner sees a Wikipedia listing appear
in search results, she should immediately think, “Are there pages on
my website that meet searchers’ informational needs?” If not, then it is
unrealistic to expect pages from your site to appear at the top of search
results for that query type. Adding appropriate product reviews, how-to
tips, and maybe an FAQs section can help your site appear in web search
results for informational query words.
By focusing only on transactional queries and ignoring informational
queries, many site owners are probably losing initial sales and potential
lifetime customers. Wikipedia listings can actually help remind site own-
ers to build more effective information into their sites. Retailer listings
can peacefully coexist with Wikipedia listings. No one ever lost a sale
to Wikipedia.
In web search results, the listing title and description/snippet are
the most important elements for searchers with informational goals.
Informational searchers want to be sure that they are clicking a link that
delivers them to desired content. So they take the extra time to read
the listing description/snippet (Figure 4.10). The URL becomes less
important to searchers because the desired information, not necessarily
the URL, is the main target.
 INFORMATIONAL QUERY INDICATORS 57

q
w
Figure 4.10 For an informational query, the two most important items in a search list-
ing are (1) the title-tag content, and (2) the page snippet or meta-tag description.

On one hand, many major search engines use the meta-tag description
when displaying a page’s (or file’s) listing. The content in the meta-tag
description should accomplish the following:
" Encourage searchers to click the link to your web page.
" Reinforce content that is already available (and visible to users)
on the page, and as a result provides and validates the scent of
information.
" Help to obtain top positions in the search results for search
engines that use meta-tag descriptions to determine rankings.
On the other hand, search engines do not always take listing descrip-
tions from meta-tag descriptions. Search engines might take the listing
description from other on-the-page content, or use a combination of the
two. For this reason, meta-tag content should reinforce the most important
keyword phrases description already available on a web page.

! INFORMATIONAL QUERY INDICATORS

Your web analytics data and keyword research data can help you deter-
mine which query words show informational intent. Some informational
query indicators include:
" Question words (how, what, where, why, who, when)
" Keyword phrases with four or more words per phrase
" Plural form of a word (indicating the desire to see a list of
items for further research)
Search engine results pages also provide strong indicators of infor-
mational intent. If you see Wikipedia consistently appear in search
results for many of your targeted keyword phrases, search engines have
determined that searchers who type in those keyword phrases want to
read information about that topic. Likewise, if you see other types of
informational websites (reference, how-to, and so on) dominate the top
search results, then search engines have deemed those keyword combi-
nations as informational.
58 CHAPTER 4: INFORMATIONAL SEARCHES—WHAT CAN I LEARN?

! OPTIMIZING FOR INFORMATIONAL

QUERIES
Here are some general guidelines for optimizing your website to accom-
modate informational queries:
" For pages that you know will be important for searchers’
informational queries, spend the extra time to write longer
(200–250 characters) meta-tag descriptions.
" Do not use the same meta-tag descriptions across an entire
site. Write unique meta-tag content for each web page, espe-
cially pages that are likely to appear at the top of web search
results for informational queries. Category pages, how-to
pages, and annotated links/resources pages are types of infor-
mation pages.
" Since search engines do not always use a meta-tag description
as the search listing’s description, make sure that other page
content (headings, paragraphs, bulleted lists, and so on) con-
tains important keywords. The double exposure of keywords
in meta-tag descriptions and keywords in the main content
area gets the best long-term results.
" The plural form of a word can be a strong indication that
searchers desire to view a list of products and/or services.
Many category and channel pages should be optimized for
the plural form of a targeted keyword.
" Do not ignore other types of optimization strategies for
informational queries. Images and videos are also desirable
responses to informational queries. Graphic images and videos
should be optimized as well as the text-based pages that con-
tain them. Local search listings can provide quick information
as well, such as hours of operation, location, and telephone
number(s).
Even if searchers do not initially click your informational search list-
ing, they do notice when your site keeps appearing over and over again
for topically related keyword phrases. After seeing your site appear for
a wide variety of keyword phrases, user confidence in your brand tends
to increase. Remember, though, user confidence can plummet with one
click if the scent of information is not validated quickly. Users remember
the sites that did not deliver, and they will not click future listings.
 CONCLUSION 59

! CONCLUSION
Key points in this chapter:
" Informational queries are important to website owners
because they are the most common type of search engine
query. Satisfying informational searches is crucial for a site’s
long-term search engine visibility.
" Many web search sessions begin with an informational query
and end with a transaction.
" All websites should contain informational pages. They provide
specific content of user interest.
" Some types of informational pages include category pages,
channel pages, FAQs or customer service pages, and pages that
contain topical lists of resources (such as Links, Resources, and
Tips sections).
" Field studies and exploratory usability tests often reveal
searchers’ informational goals above and beyond what can be
inferred from web analytics data and keyword research tools.
" Longer keyword phrases (more than three words) tend to be
informational queries; shorter ones tend to be navigational
queries.
" If question words (how, what, where, why, who, when) are part
of a keyword phrase, then the keyword phrase is an informa-
tional query.
" The plural form of a word can be a strong indication that
searchers desire to view a list of products. Many category and
channel pages should be optimized for the plural form of a
targeted keyword.
" The appearance of Wikipedia and other how-to listings in
search results is a strong indication that the query words show
informational intent by the users.
" The description or snippet is more important than the URL
structure for informational searches. Keep the meta-tag
description longer (around 200 to 250 characters) for pages
that typically satisfy searchers informational intent.
" Do not use the same meta-tag description on every page of a
site. Tailor meta-tag descriptions to reflect actual page content.
This page intentionally left blank
CHAPTER 5
TRANSACTIONAL SEARCHES—
WHAT CAN I DO?

With a transactional query, a searcher’s main goal is to perform some

sort of activity beyond merely reading. Sometimes the activity occurs
online, such as downloading software, watching a video, or playing an
online game. And sometimes the activity ultimately will occur offline,
such as purchasing bedroom furniture or going to the dentist. Approxi-
mately 10 to 24 percent of web queries are transactional.

Transactional queries are important to website owners and search engine

optimization professionals because they both hope to capture searchers
at a critical point in the buying process: right when they are ready to
buy and provide personal information (such as name, address, phone
number, email address, and so on). Let’s look at how to determine that
a search is transactional and how to ensure your site meets searchers’
expectations.
62 CHAPTER 5: TRANSACTIONAL SEARCHES—WHAT CAN I DO?

Additionally, items such as videos, sound files, slide shows, games,

and so forth can increase the stickiness of a site, encouraging site visitors
to stay on your site longer and view more content. Popular and informa-
tive interactive items can also increase a site’s external, third-party link
development, which has a direct impact on a website’s rankings.

! TRANSACTIONAL INTENT
Determining transactional intent can be tricky because searchers do not
always type the activity they wish to perform as a keyword. For example,
a person might be interested in buying some crystal water goblets but
does not type in the word buy or purchase as query words (Figure 5.1).
Figure 5.1 A searcher
who wishes to buy some
water goblets will prob-
ably not type the word buy
in the search box. Yet this
searcher might be ready
to buy. Notice that Google
has placed shopping search
results at the top of this
search results page, accom-
modating searchers with
transactional intent.

As another example, searchers might want to enroll in a class or watch

a video but not type the word enroll or watch as part of the search query.
They might want to listen to the latest song by their favorite band but
not type the word listen as part of the search query. They might want to
play a game of Sudoku but not type the word play. Searchers can show
transactional intent by using nouns (video, music, game) as well as verbs
(download, chat, enroll). The activity might occur on an actual website,
such as getting a quote for auto insurance. Or the activity might occur
offline, such as making a phone call to a local pharmacy to refill a pre-
scription. For product-oriented nouns, consider that users might have
both informational and transactional intent.
Nevertheless, an easier case occurs when searchers do type the exact
activity they wish to perform. The words download, apply, search, and find
 TRANSACTIONAL INTENT 63

are more commonly used than you might imagine. Regular keyword
research often reveals the precise action that searchers wish to take, and
these action words (usually verbs) should appear in search listings as well
as corresponding landing pages. Web pages that satisfy transactional intent
do have special features, as shown in the following examples.

Example 1: Buying Products

Product pages on an ecommerce site should satisfy transactional queries.
Since searchers with transactional intent wish to perform a specific
activity, product pages that meet their expectations should include the
following items:
" Actual query words (to validate the scent of information)
" Words associated with the desired activity (such as Add to Cart,
associated with the words buy or purchase)
" Clear call to action
" Enough graphic and textual information (such as a product
photo, product description, and price) to close the sale
All of this information should appear above the fold, so that site
visitors do not have to scroll. In other words, searchers’ most desired
information and desired transactional keywords should appear above the
fold. Figure 5.2 shows portions of a product page that meet the expecta-
tions of online shoppers with transactional intent. As an added bonus,
such pages satisfy users’ informational intent as well.
Figure 5.2 Searchers’
most desired information
appears above the fold:
product name, short product
description, price, and the
primary call to action
(Add to Cart).
64 CHAPTER 5: TRANSACTIONAL SEARCHES—WHAT CAN I DO?

A category page can also meet the needs and expectations of online
shoppers with clear transactional intent. Product names, prices, thumbnail
photos, and clear calls to action are normally portions of an ecommerce
site’s category page (Figure 5.3).
Figure 5.3 Category pages
can also meet searchers’
transactional expectations.

Category pages often appear in web search results for transactional

queries, although searchers prefer to view more detailed product pages before
making a final purchasing decision, especially for high-ticket items. Product
pages provide more screen real estate for keyword placement, too.

Example 2: Entertainment
Transactional intent does not automatically mean the desire to purchase
a product or service. People also use the web for entertainment. Recent
years have seen a global explosion in online video viewing, photo shar-
ing, and audio file downloads.
 TRANSACTIONAL INTENT 65

However, Flash videos and splash pages in search results can cause Note A splash page is a
searcher frustration. Here are a few reasons that many users dislike web page that typically
Flash videos: consists of either a Flash
animation and a redirect
" Delays the scent of information (users have to watch a video to a new page after the ani-
before they see their keywords validated) mation finishes, or a link to
skip the Flash animation
" Diminishes the scent of information (videos tend not to be (Skip Intro).
keyword focused)
" Hides the scent of information (site visitors never see their
keywords validated on a web page)
After users click a link to a web page from a search engine listing,
they generally do not want to watch a video (which might be an advertise-
ment) before they see their user-generated information scent validated.
They become even more irritated if they watch a Flash movie and never
see their information scent validated at all, or if the movie distracts them
from reaching their query goals (Figure 5.4).
Figure 5.4 This screenshot
is the beginning of a Flash
movie on a website’s home
page, with the call to action,
skip intro, barely visible at
the bottom of the screen.
What type of products or
services do you believe this
site offers?

Videos often decrease search usability by delaying,

diminishing, distracting from, or hiding the scent of
information.
Even the most experienced web users become easily irritated with
Flash videos, as the “The Cost of Assuming User Expectations” case study
illustrates.
66 CHAPTER 5: TRANSACTIONAL SEARCHES—WHAT CAN I DO?

CASE STUDY
THE COST OF ASSUMING USER EXPECTATIONS

A software client wanted to increase the search usability of his website, and he wanted to
determine why the site was not converting well. Since this client’s minimum purchase price was
over $100,000, he was not targeting small or medium-sized companies. The specific primary
persona was the head of an information technology (IT) department who made purchasing
decisions. Therefore, we only recruited people who fit this description as participants in the
usability test. All participants had considerable web experience.
People who commonly visited our client’s site had the following goals:
" Find a vendor source for products or technology licensing (transactional and
informational)
" Quickly get a datasheet on desired technology and/or product (navigational)
" Purchase or license products/technology (transactional)

Nowhere in these goals is the word video or Flash mentioned. Nevertheless, the client used many
Flash movies on the site, especially on the home page. Right away, we knew we were going to
tell our client to sink the splash page, or at least decrease the amount of screen real estate
allocated to the large Flash video. Our client insisted that their target audience absolutely loved
the Flash video. IT people love to watch cool movies, they claimed. Plus management insisted
that the use of advanced technology on a site gave their brand more credibility.
Our usability test began on the client’s home page. Here were some of the responses to the
Flash video:
" “I don’t have time for this. Quit wasting my [expletive] time….” (Participant abandoned
the task.)
" “The site wastes space. I am very busy… I want to get down to business. I want to see
things. These light waves—this is just distracting.” (Participant abandoned the task.)
" “This graphic takes up way too much space and doesn’t give me any information.
It doesn’t tell me anything and keeps up this annoying flashing when I’m trying to do
something else.” (Participant abandoned the task.)
" “What the [expletive] is this?” (Participant abandoned the task.)
" “Are you [expletive] kidding me? I don’t have time for this….” (Participant abandoned
the task.)

Unfortunately, our client did not accept our findings, even after we played audio recordings of
the sessions. Six months later, when we viewed our client’s updated website, almost all of the
Flash had been removed. Even the large Flash video on the home page was smaller, and more
room was made for keyword-focused copy.
Moral of the story? Advanced Internet users do not automatically wish to see the latest and
greatest technology on a website merely because they are advanced Internet users. Don’t
assume. Test.
 TRANSACTIONAL QUERY INDICATORS 67

! SEARCH LISTINGS AND TRANSACTIONAL

QUERIES
In terms of individual parts of a transactional search listing, transactional Note A/B split testing is a
listings are similar to informational search listings. The URL is less impor- method where a baseline
tant to searchers because the information and the associated activity is control item is compared to
a number of single-variable
the main target, not necessarily the URL. Keyword research, A/B testing, items in order to determine
and multivariate testing can show you if interactivity keywords should the best option. For exam-
appear in the listing, as shown in Figure 5.5: ple, you can compare differ-
ent font sizes in a headline
to determine which size
users respond to most.
q Multivariate testing (also
w known as multivariable test-
ing), as the name implies,
compares multiple vari-
Figure 5.5 For a transactional query, the two most important items in a search listing ables. For example, a multi-
are the (1) title-tag content, and (2) the page snippet or meta-tag description. Notice variate test can reveal the
that the activity words play, set, and try appear in the listing description, encouraging best font size, font color,
searchers to click the link to the website. and number of words in a
web page headline.

! TRANSACTIONAL QUERY INDICATORS

Your Web analytics data and keyword research data can help you deter-
mine which query words show transactional intent. Some transactional
query indicators include:
" Specific interaction verbs (buy, find, search, download, play,
view, log in, register, enroll, subscribe, join, apply, contact, chat,
and so on).
" Nouns that are associated with some type of activity (games,
movies, music, recipes, slide shows, demo, tour, quote, calculator,
software or the software name, and so on).
" File extensions for nontext files (.jpg for photos, .mp3 for
music, and .mpg for videos) and file compression (.zip for
Windows computers and .sit for Macintosh computers).
68 CHAPTER 5: TRANSACTIONAL SEARCHES—WHAT CAN I DO?

! OPTIMIZING FOR TRANSACTIONAL

QUERIES
Here are some general guidelines for optimizing your website to accom-
modate transactional queries:
" The primary call to action should be painfully obvious to
both site owners and searchers. For example, if you want
searchers to download a file, the word download should be
part of the hyperlink, and that hyperlink should look clickable
(Figures 5.6 and 5.7). The transactional scent of information
should be maintained from a search results page to the landing
page. Chapter 8, “Search Usability Is Everyone’s Job,” discusses
calls to action in greater detail.

Title URL Description

Figure 5.6 In the search listing, the call to action is seen in the title tag, the descrip-
tion (which is the page’s meta-tag content in this instance), and the URL.

Figure 5.7 Notice that the

call to action on Apple’s t w
Safari page clearly shows
searchers where they can
click to download the
browser software (1). The
call to action is also clear
in the web page’s title tag
e
(2), the graphical heading
(3), the textual heading (4),
and the URL (5).

q
 OPTIMIZING FOR TRANSACTIONAL QUERIES 69

" Include desired activity words in the title tag on key pages. For
example, if you offer a specific search page on your site, then
make sure you use the word search or find in the title, heading,
URL, and description of the page. Likewise, if you have a login
page on your site, such as an email login, make sure you use
the word login in the title, heading, URL, and description of
your page.
" Don’t assume searchers want to take an action without initiat-
ing it themselves. In other words, don’t start playing a video or
sound file unless the searcher specifically indicates that he/she
wishes to watch the video or listen to the sound file.
" Focus groups are not always the best source of information
for feedback on interactivity and multimedia because the focus
group leader, not the user, guides the interactivity. Further-
more, a focus group typically shows initial reactions to an
interactive feature, not long-term effects.

CASE STUDY
OFFICIAL FAN SITE

One of the most revealing field studies we performed was for an official Frank Sinatra fan site.
The site had many sound clips that people could listen to at their leisure…except for one file on
the site’s home page that automatically played every time the home page loaded.
The site owners thought the sound file was really cool. When fans first visited the site, they
also thought the sound file was cool. You could see that they liked it from the expressions on
their faces.
Over time, though, as we worked on iterative designs and functionality, we watched test par-
ticipants become irritated with the home page. At some point, most participants asked if they
could unplug the speakers or turn off the volume before the test commenced.
We quickly learned that sound files might impress new visitors but alienate repeat visitors. Had
we not observed one-on-one user activity on this site over time, we never would have learned
this about the site’s target audience. The site still uses sound files, but a sound clip does not
play unless a site visitor specifically requests to hear it.
70 CHAPTER 5: TRANSACTIONAL SEARCHES—WHAT CAN I DO?

! CONCLUSION
Key points in this chapter:
" Transactional queries are those in which searchers wish to
perform some sort of online or offline activity beyond merely
reading.
" Transactional queries are very important to site owners
because they capture searchers at a critical point in the buying
process: when they are ready to buy.
" Certain verbs such as download, search, find, log in, chat, apply,
buy, and so forth indicate transactional intent.
" Transactional keyword phrases do not always contain a verb.
Certain nouns such as video(s), picture(s) or pic(s), music, and
game(s) also show transactional intent because they imply
watching, hearing, and playing, respectively.
" Likewise, file extensions such as .jpg for photos, .mp3 for
music, and .mpg for videos are a strong indicator of transac-
tional intent.
" On key pages within a site, place the word that describes
searchers’ desired activity in the title tag content. For example,
an advanced search page should contain the words search or
find in the title tag content. Pages where searchers can down-
load software should contain the word download in the title tag.
" When searchers demonstrate transactional intent, the cor-
responding landing page (after clicking a search listing) should
contain their desired call to action above the fold. This call to
action should stand out on the page. In other words, searchers
should not have to work very hard to complete their desired
activity.
" Don’t delay, diminish, distract from, or hide the scent of infor-
mation by initiating an action (such as playing a video). Let
users initiate their desired activities.
CHAPTER 6
THE SCENT OF
INFORMATION AND
LANDING PAGES
After searchers click a link to your website from a search engine, they
have two choices: They can either stay on your site, or they can abandon
it. One reason searchers stay on a website is the scent of information.
If they see their user-generated scent of information (keywords) on a
website’s landing page, they believe the page will help them reach their
goals: where they want to go, what they want to learn, and what they
want to do. Searchers become confident that the landing page is giving
them what they need. The result is a great user experience, more sales
and conversions, and a positive brand impact.
72 CHAPTER 6: THE SCENT OF INFORMATION AND LANDING PAGES

Searching does not end after a person clicks a link from a search
engine results page (SERP) to a website. People still look for desired
content once they enter and browse a website, and they exhibit some
specific behaviors while they are looking. By considering these behaviors
and applying some simple techniques, you can build your website to help
users follow the scent of information.

! SCANNING BEHAVIOR
The vast majority of users do not read web pages word by word. Instead,
they scan pages for individual keywords and keyword phrases. If users
arrive on your site and do not see their desired content after a quick
scan, they will abandon the site and move on to another. Recent studies
show that users only read about 20 percent of the words on a web page.
Therefore, if users arrive on your site via the commercial web search
engines, important keywords and calls to action need to be featured
prominently (above the fold) on web pages.
Note See Jakob Nielsen’s It is not enough to call attention to keywords on your web pages.
“How Little Do Users Read?” For your site to be successful, users must perform certain actions on it.
at http://www.useit.com/ Calls to action should stand out on the page as well. What do you want
alertbox/percent-text-read.
html. The article contains a users to do after they find their desired content? Add to cart? Enroll or
link to the full academic register? Log in? Subscribe to your blog? Fill out a form?
paper. A call to action is something you wish users to do when they land on
a page. Most web pages contain multiple calls to action. As an example,
let’s use a product page from an e-commerce website.
The primary call to action on a product page is Add to cart. A primary
call to action is the one action you most wish users to do. If users are not
ready to respond to your primary call to action, what do you want them
to do instead? Abandon your website? Of course not. You want them to
stay on your site and maybe purchase a similar item. So the secondary
call to action might be to click links to alternative products. What if users
do not wish to answer the secondary call to action? What do you want
them to do? Call a customer service agent?
Identifying and prioritizing calls to action for every page on your
site might seem like an exhausting exercise. However, site visitors will
not Add to cart unless you make it easy for them. And they will not look
at additional items unless you provide links to those related items. Site
visitors need to see both keywords and the main calls to action with a
quick scan.
 SCANNING BEHAVIOR 73

How can you call attention to both keywords and calls to action on
a web page? First, you should identify and prioritize the primary, second-
ary, and third-level calls to action on each type of web page. Table 6.1
shows a simple spreadsheet layout for this process:

Table 6.1 Identifying and prioritizing calls to action on web pages

PAGE NAME CALL TO ACTION 1 CALL TO ACTION 2 CALL TO ACTION 3

Product A Add to cart Watch product Click links to

video related items
Product B Add to cart View larger image Click links to
related items
Contact Us Fill out contact Email customer Call toll-free
form service number

Second, as you design each element on a page, format the primary

call to action so that it stands out. The following list includes some ways
to draw attention to an item on a web page (ordered from the least atten-
tion grabbing to most attention grabbing):
" Dimmed text
" Italics
" Bold
" Larger font/typeface (typically in page headings)
" Border
" Color (warm colors advance and cool colors recede; eyes
naturally move to heavy color saturation and a large number
of colors, and so on)
" Graphic images (curves versus straight lines; photos of faces
or people; effects that indicate a dimension change such as
beveling, embossing, shadows, and so on)
" Sound
" Animation/movement

The secondary call to action should be subtler than the primary one,
but the secondary call to action should still be featured prominently on
a page. The third-level call to action should be even more subtle than
the secondary one, and so on. Figure 6.1 shows the formatting of the
various calls to action on a product page.
74 CHAPTER 6: THE SCENT OF INFORMATION AND LANDING PAGES

Figure 6.1 An e-commerce

site’s product page.

q
w
e

The primary call to action is Add to cart. It is formatted as a graphic

image in a warm color. The image has rounded corners, which makes it
stand out on a website that has a grid layout, and it has an icon. It is larger
than other navigation buttons. The shading and slight bevel add unique
dimension. And it is placed near the center of the screen surrounded by
some extra white space.
The secondary call to action is either to enlarge the product photo
or to take the product tour. These calls to action are easy to spot because
online shoppers always want to see a photo or a video of the product
they are purchasing. Therefore, users have come to expect a link for
such views. Additionally, photos typically contain millions of colors. The
product photo is the largest photo on the page. The curved magnifying
glass (icon) link to view the larger photo is located within the photo’s
border. Note that the magnifying glass conforms to prior experiences
among users. Therefore, the site visitor need not spend effort to “decode”
the icon. User icons match users’ expectations and prior experiences.
Users do not have to click that icon to view a larger photo. They can
click the product photo as well. Both the magnifying glass link and the
product tour button use a dimmed color, which helps the primary call
to action stand out.
The third-level call to action is to view related items. This call to
action is formatted as text links and associated thumbnail photos, which
are located in the right column of the page. Note that placement on the
right side conforms to site visitors’ expected pattern of page reading: left
to right. Thus, a third-level call to action fits best to the right of a sec-
ondary call to action. In a text-like document, task follows our left-right,
 SCANNING BEHAVIOR 75

top-down reading habit (in Romance languages). Note that Arabic and Note A performance test
Hebrew are read right-left, then top-down. Follow the expectations of evaluates the effectiveness
of site navigation, labeling
whatever language you serve. and terminology, organiza-
Once you have formatted some page prototypes, you can conduct tion of content, task flow,
a variety of usability tests to determine their effectiveness. Eye-tracking error handling, and calls to
studies can tell you whether users’ eyes are focusing on keywords and action on a website.
Performance tests can tell
calls to action. Performance tests can tell you which formats yield the you how many users com-
most clicks and the most conversions. Even free exploration tests can let plete the assigned task, how
you know if calls to action are annoying, too subtle, or just right. Let’s many steps they took to
complete the task, time
look in greater detail at a couple of usability tests you can conduct to spent, and potential road-
determine the mental models of web searchers. blocks. You will learn how to
do a free exploration test in
Chapter 9, “How To Improve
Anticipate Searcher Needs with Usability Tests Your Website’s Search
One way to determine searchers’ mental models of your web pages is to Usability.”

perform an expectancy test. An expectancy test will also reveal searchers’

expectations about your web page’s content.
To perform this test, you first will need to gather examples of listings
from the commercial web search engines, using your most important
keyword phrases shown in Figure 6.2.
Figure 6.2 Google listings
for the keyword phrase
pancreatic cancer symptoms.
These listings appear spe-
cifically for the National
Cancer Institute’s website.

If you are having a difficult time finding your site’s listings in Google’s Note If you want to find
search results, you can easily find them by performing an advanced search. the search listing for an
Type in your targeted keyword phrase followed by site: and your domain exact keyword phrase on
your site, enclose the key-
name (no space between the colon and your domain name). For example, word phrase in quotation
the listings in Figure 6.2 were located using the following query: marks. For example, if you
want to find the exact
pancreatic cancer symptoms site:www.cancer.gov
phrase pancreatic cancer
How to perform the expectancy usability test: symptoms on the National
Cancer Institute site, the
1. Present participants with one or two search engine listings, Google query will be
such as the ones shown in Figure 6.2. “pancreatic cancer symp-
toms” site:www.cancer.gov.
2. Ask participants, “What content do you expect to see if you
click this link?” Have them explain why they expect to see
specific content.
76 CHAPTER 6: THE SCENT OF INFORMATION AND LANDING PAGES

3. Ask participants, “What do you think will happen if you click

this link?” Have them explain why they believe something will
or will not happen.
4. Perform this test for navigational, informational, and transac-
tional keyword phrases.
Note Different ways of Normally, in an expectancy test, you do not tell participants whether
using the five-second usabil- their answer is right or wrong because you are simply determining their
ity test will be discussed in mental models. However, we also want to know how well or poorly the
Chapter 9, “Search Usability
and Your Site’s Success.”
landing page meets searcher expectations. So follow up the expectancy
You can find out more about test with Jared Spool’s five-second usability test, a test which is used to
Jared’s five-second test at gather initial impressions and to quickly measure how a content page
www.uie.com/articles/
performs with users.
five_second_test/
How to perform Jared Spool’s five-second usability test:
1. Show the user the landing page that corresponds directly to
the search listing.
2. Count five seconds.
3. Close the window or take the landing page away from the user.
4. Ask the user:
" Was this content what you expected to see after you
clicked the search listing? Why or why not?
" What (content) did you expect to see but did not see?
" What (content) did you want to see? Why did you want
to see it?
What you are trying to determine is whether users easily spotted
their scent of information (keywords). Were users able to see their infor-
mation scent validated on the landing page? If they did not, why didn’t
they see it? Where on the landing page did they expect to see keywords
and associated images?

If the scent of information is too weak or disappears,

people will abandon your website.
Sometimes, after the initial five-second test, you might allow test
participants to view the landing page for an additional 8–10 seconds to
gather more details. But still carefully note what they did and did not
notice within the first five seconds of viewing the page.
If a web page was optimized well for search engine visibility, key-
word usage on the landing page will be obvious. You will commonly
see keywords in the title tags, visible body text (headings, paragraphs,
 SCANNING BEHAVIOR 77

breadcrumb links, and so on), and the URL. However, you might not
always see keywords on the page because the design team might have
unwittingly blocked the information scent.
Remember, if the scent of information is too weak or disappears,
people will abandon your website. Let’s look at some ways that web
designers, and search engine marketers, might unwittingly delay, dimin-
ish, or block the scent of information.

Avoid the Iceberg Effect

On many websites, the most important content tends to get buried far
below the fold, like the proverbial iceberg, with only 10 percent of page
content visible at the top of the screen (Figure 6.3).
Figure 6.3 The iceberg
effect

Above the fold

Below the fold

When web searchers land on an iceberg page, Note Bounce rate is a term used in website analytics
it does not appear to have their desired content. that refers to the effectiveness of an entry page. More
The scent of information is missing and they specifically, it is a percentage of pages that are viewed
abandon the site. So if you find that important once—when a site visitor lands on an entry page and
then abandons the site after only viewing that entry
pages have a high bounce rate, see if the page page. It is typically calculated as:
layout might be hiding keywords. Number of single page accesses
µ 100 = Bounce rate
Some web designers and developers inadver- Total number of entries
tently hide the scent of information when they use
a file tab layout as shown in Figure 6.4 on the next page. The default file
tab displays product details. If a user clicks another file tab, the product
details content will disappear and different text will appear in its place.
Although this particular design strategy is useful for preserving screen
real estate, it can hide the scent of information—keywords.
78 CHAPTER 6: THE SCENT OF INFORMATION AND LANDING PAGES

Figure 6.4 File tabs and

their corresponding layers.
Product review content is
q
not visible to web searcher
(1) until they click the
“Reviews” link (2). If the
website owner wants
searchers to read product
reviews, the scent of infor-
mation (the “Reviews” link)
should be located at the tip
w
of the iceberg.

An even worse situation can occur if the file tabs appear below
the fold. In this situation, not only are important keywords invisible to
users, but the links that lead to their desired information disappear. To
preserve the scent of information on pages that use this layout, make
sure the most important keywords are available above the fold in the
short product description. If you cannot place searchers’ most important
keywords above the fold, then make sure the file tabs appear above the
fold (Figure 6.5). The file tab links let your target audience know that
their desired content might be available on that web page.
Figure 6.5 Keep file tabs
visible above the fold to pro-
vide an information scent.

Above the fold

Below the fold

 SCANNING BEHAVIOR 79

Another effective way to maintain the scent of information on a long Note A screen reader is
web page is to write multiple HTML headlines using keywords. At the top software that interprets and
of the web page, make sure there are hypertext links to those headings. reads (aloud) content that is
displayed on a computer
This format is highly useful for screen readers as well (Figure 6.6). screen. Computer users who
are blind or visually impaired
typically use screen readers
to navigate websites and
read site content. Visually
impaired users can call up a
list of on-the-page headings
and jump directly to the sec-
tion of the page that con-
tains their desired content.

Figure 6.6 Provide links above the fold to the content below the fold.

To maintain the scent of information, make sure that the words on

file tabs and hypertext links are reinforced in the text the users see after
they click them. If the words are not reinforced, users feel the links are
misleading.

Avoid Misleading Links

Users become frustrated if they follow a scent of information and it leads
them down the wrong path. If users believe that a website is deceitful,
they will view that brand as being untrustworthy.
Suppose you are driving to Phoenix, Arizona, to meet potential cus-
tomers. You are looking forward to meeting these prospects because they
have actively sought out your products and services. As you are driving, Figure 6.7 Road sign point-
ing to Phoenix, Arizona.
you see the sign shown in Figure 6.7.
The scent of information in this sign is straightforward. If you wish
to go to Tucson, then you drive straight ahead. If you wish to go to Phoenix
Phoenix, then you must make a right turn. The sign provides two visual
cues as the scent of information: city names (text) and arrows (images) Tucson
indicating the direction you must drive to reach your destination.
Now suppose your competitor wants you to miss your meeting Figure 6.8 Sign showing the
with your prospect. He replaces the sign shown in Figure 6.7 with the wrong directions.
sign in Figure 6.8.
80 CHAPTER 6: THE SCENT OF INFORMATION AND LANDING PAGES

The directions are reversed. Drivers who follow this misleading scent
of information will drive toward the wrong city—wasting time and gas
money, maybe even losing new customers. Even if a person notices that
the sign’s directions are reversed, he is distracted from reaching his des-
tination because of his confusion. And this is how users feel when they
use a website with misleading or mislabeled links—confused, distracted,
annoyed, and frustrated.
Many Flash sites appear to be misleading links in search listings because
searchers do not see keywords in the search listing also appearing on the
landing page (Figure 6.9). What product, service, or information does
this home page offer? Help desk software? Accounting software? Press
release services? Babysitting services? Or maybe skip intro services? In
this instance, the Flash technology is not the problem since Google has
been able to crawl Flash sites for many years. The problem is how Flash
technology is being used—web developers and site owners do not always
consider searchers’ scanning behavior when they build these sites.
Figure 6.9 This web page
formatted in Flash does not
contain any keywords.

On a website, a misleading scent of information can be a costly

mistake. Not only will it confuse current site visitors, it will also alien-
ate new visitors. Even if you did not intend to mislead site visitors, they
still might perceive your site as deceptive. People will not give you their
personal information and credit card numbers if they feel your website
is untrustworthy.
 FOSTER ORIENTING BEHAVIOR 81

! FOSTER ORIENTING BEHAVIOR

Orienting is a behavior in which people use cues to determine where they
are located. In the physical world, people might determine their loca-
tion by using a map, road signs (Figure 6.10), symbols, words, a global
positioning system (GPS), or a combination.
Figure 6.10 The stop sign,
the painted line on the
road, and the word stop
painted on the road are all
visual cues that help drivers
orient themselves at this
intersection.

Orienting is something people do without realizing it. For example,

when you ride in an elevator, the lights above the doors communicate
which floor you are on. If you see the number “3” light up, you know
the elevator is on the third floor. If you see the letter “L” light up, you
know the elevator is on the lobby level. The elevator lights are you are
here cues that help you orient yourself within a building.

The presence of easily scanned you are here cues

makes users feel your site is trustworthy and credible.
Likewise, there are many you are here cues that can help users deter-
mine their location within a website. After people click a link from a
search engine results page, they scan the web page for you are here cues.
Ideally, people should be able to orient themselves on a web page with
little or no effort. And they should be able to orient themselves within
five to 10 seconds after a page loads.
82 CHAPTER 6: THE SCENT OF INFORMATION AND LANDING PAGES

The presence of easily scanned you are here cues helps instill trust
and credibility in your site. As users click, the site should communicate
to them that they are getting closer to their goal. You are here cues can
provide that feedback, especially if they are based on query words. In
addition, consistently implementing you are here cues helps users accom-
plish their tasks more quickly and efficiently.
To determine how easily searchers orient themselves on a web page,
use the five-second usability test again:
1. Show the user the landing page that corresponds directly to
the search listing.
2. Count five seconds.
3. Close the window or take the landing page away from the user.
4. Ask the user:
" Whose website are you viewing? How did you determine
this?
" What section, if any, of the site are you viewing? How did
you determine this?
" What content is available on this page? How did you
determine this?
" How confident are you that your desired content is on
this page?
5. Ask the following questions if participants indicated that they
did not see their desired content:
" Do you believe that this page will lead you to your desired
content?
" How did you determine this?
" How confident are you that this page will lead you to your
desired content?
Figure 6.11 shows the places that people typically look at to orient
themselves within a website.
1. Title tag. In Firefox and Internet Explorer, a web page’s title-
tag content is visible at the top of the screen and in the file tab
underneath the address bar in the latest web browsers.
2. URL. In an ideal situation, the URL should reflect page con-
tent, especially for navigational queries.
3. Logo and tagline. People expect to be told whose site they are
visiting in the upper left corner of a web page.
 FOSTER ORIENTING BEHAVIOR 83

q Figure 6.11 Areas on a web

w page where site visitors scan
to establish a sense of place.

e
r
t

y
u

4. Locational breadcrumb links. These links communicate to

users where they are within a website’s hierarchy and what
page they are currently viewing.
5. Page heading. Site visitors use the page heading to orient
themselves largely due to wording, font size, font color, sur-
rounding white space, and location on the screen.
6. Main content at the top of the page. People scan text at the
top of the page for their targeted keywords. If they used a
transactional word for images (such as photos) or videos, they
expect to see non-text representations of those transactional
words as well.
7. Navigation buttons. Current page and/or section of the site is
highlighted.
You are here cues are very important to web searchers because when
they click a link from a search engine results page, they do not always
enter the site via the home page. In all likelihood, they arrive on a page
somewhere in the middle a site. If the search engine correctly interprets
searchers’ intent, then searchers should land on a page that contains
their desired content, or leads them to their desired content. Let’s look
at an example.
84 CHAPTER 6: THE SCENT OF INFORMATION AND LANDING PAGES

Figure 6.12 shows the top Google search results for the informa-
tional query breast cancer tests. Notice that both listings from the National
Cancer Institute site do not lead to the home page.
Figure 6.12 If web search-
ers click a link from a search
engine results page, they
often land on a page in the
middle of a website.

The first search listing in Figure 6.12 leads to a category page. Google
has interpreted the plural form of the word test to mean that searchers
might be looking for a list of breast cancer tests (Figure 6.13), which the
following landing page provides.
Figure 6.13 Landing page
after a searcher clicks the
link in the first search listing
in Figure 6.12. The searcher
is landing on a category
page in the middle of a site,
not the home page.

Now let’s review the questions asked during a five-second usability

test to see if and how users were able to determine their location within
the National Cancer Institute’s website.
" Whose website are you viewing? How did you deter-
mine this?
Participants said they were viewing the National Cancer
Institute’s website. They specifically mentioned the logo and
 FOSTER ORIENTING BEHAVIOR 85

text in the masthead area as key indicators that they were

viewing a web page on the National Cancer Institute site.
" What section, if any, of the site are you viewing? How did
you determine this?
Participants believed they were in the Cancer Topics area
of the site because they saw the navigation button labeled Can-
cer Topics highlighted. They also believed they were in a breast
cancer area of the site because of the large page heading.
" What content is available on this page? How did you deter-
mine this?
Participants believed they were viewing a page about breast
cancer tests, the very keywords that they typed into Google.
The large page heading, the logo, and the organization name
all led them to believe that their desired content was available
on this page.
" How confident are you that your desired content is on
this page?
All participants were confident that their desired content
was on the page, or at least on the site. This particular land-
ing page contains easy-to-find links to guide people to more
detailed information on other pages within the site. The pres-
ence and formatting of the secondary heading Mammography
helped increase user confidence (Figure 6.14).

Figure 6.14 When test participants were shown the National Cancer Institute’s cate-
gory page about breast cancer tests for five seconds, almost all participants were able
to establish where they were within the NCI website.

Websites that facilitate scanning and orienting help searchers reach

their goals more quickly and efficiently; increase user confidence, trust,
and credibility; and can help sites achieve and maintain top search
engine positions. However, not all behaviors that people exhibit when
they look for content indicates a positive user experience. Some search
behaviors indicate a negative user experience due to a lost scent of
information.
86 CHAPTER 6: THE SCENT OF INFORMATION AND LANDING PAGES

! POGO-STICKING: NEGATIVE SEARCH

BEHAVIOR
Note For details on pogo- Pogo-sticking is a browsing behavior where users jump up and down
sticking, see “Galleries: The the hierarchy of a website to find their desired content. If users exhibit
Hardest Working Web Pages considerable pogo-sticking behavior, they are less likely to purchase,
on Your Site” by Jared Spool
at http://www.uie.com/
download, or answer any primary call to action on your website. Here
articles/galleries/. Spool is an example of pogo-sticking behavior.
found that the more users Suppose a searcher wants to purchase a digital camera. He is viewing
exhibited pogo-sticking
a category page labeled “Digital Cameras.” This category page contains a
behavior on a website, the
less they purchased. list of thumbnail photos and corresponding text links to individual product
pages. One camera looks promising. So he clicks the link to view details
about the camera. He doesn’t see exactly what he wants and clicks back
to the category page. He sees another camera that he likes. So he clicks
the link to a different product page. The camera is too expensive, so he
clicks back to the category page. And so on. This searcher is exhibiting
pogo-sticking behavior, jumping back and forth between a category page
and individual product pages as Figure 6.15 illustrates.
Figure 6.15 A searcher
clicks back and forth
between a product page
and category pages,
pogo-sticking to find the
information they want.

Usability testing is crucial to determining whether users are pogo-

sticking, because web analytics data can be misleading. Many analytics
packages are designed to calculate the average number of page views
per visitor, but the packages are not set up to determine if the increased
page views are due to pogo-sticking behavior.
Note A clickstream is a Clickstream analysis might help you identify pages where pogo-sticking
recording of what a site occurs. To help users succeed in getting their desired content, site own-
visitor clicks while browsing ers should understand where people are pogo-sticking on a site, and why
a website.
people are pogo-sticking. Usability testing will give you those answers.
 CONCLUSION 87

Usability testing is crucial to determining whether

users are pogo-sticking, because web analytics data
can be misleading.
Remember, management might be happy to see increased page
views. But if site visitors are pogo-sticking, they are having a negative
user experience. Most users buy products without pogo-sticking at all.

! CONCLUSION
Key points in this chapter:
" Users only read about 20 percent of the words on a web
page. Users don’t read; they scan web pages for their targeted
keywords. So don’t make users work too hard to find desired
content.
" Conduct expectancy tests to determine web searchers’ mental
model of your landing page after they view various search
listings.
" Web searchers want see their query words and calls to action
featured prominently on web pages.
" Identifying and prioritizing calls to action for every web page on
your site is important for converting site visitors into buyers.
" Make the primary call to action the most noticeable call to
action on a page. The placement and formatting of secondary
and third-level calls to action should be more subtle.
" Web designers and search engine marketers might delay,
diminish, or block the scent of information by hiding desired
content below the fold and in invisible layers.
" Many Flash sites appear to be misleading links in search listings
because searchers do not see keywords in the search listing
also being used on the landing page.
" Orienting is a behavior in which people use cues to determine
where they are located with little or no effort.
" Site visitors should be able to orient themselves within five to
10 seconds after a page loads.
" Pogo-sticking is a browsing behavior in which users jump up
and down the hierarchy of a website to find their desired infor-
mation. This behavior can be an indication of a lost informa-
tion scent.
This page intentionally left blank
CHAPTER 7
SEARCH USABILITY AND
YOUR SITE’S SUCCESS
This chapter illustrates how improving search usability can benefit web-
sites that sell products and services as well as those that sell advertising
on their web pages. We’ll also look at how you can use search usability
to support customers of existing products and services and how you
can improve the bottom line when you reduce employees’ time on the
company intranet by improving the findability of content.

Additionally, we examine common measures of success and web ana-

lytics that might look easy to interpret, but can be deceiving if you don’t
understand the thinking behind users’ actions.
90 CHAPTER 7: SEARCH USABILITY AND YOUR SITE’S SUCCESS

! SEARCH USABILITY METRICS BY SITE TYPE

From a business perspective, a website can have a variety of specific
primary goals, such as increasing conversions. For example, the primary
goal of an e-commerce site is to increase online purchasing. Similar is true
for nonprofit websites (increase memberships) and business-to-business
websites (generate more leads). The primary business goal for news and
information websites is to increase page views so they can charge more
for advertising.
The following sections outline business models that involve making
financial and traffic assumptions. The take-away from the models is the
thinking behind the numbers, not the numbers themselves. Adjust the
models to calculate ROI that more accurately reflects your website’s
potential. A model is a representation of a process designed to aid in
understanding and/or predicting results. The following formulas are
an example of one type of business model to predict daily profit of an
e-commerce website.
(5,000 visitors/day µ 2% conversion) µ $20 sale = $2,000 gross sales
Gross sales µ 20% profit = $400 profit
When developing your assumptions for your models, keep in mind
that the more accurate your values, the more accurate your annual rev-
enue prediction will be. Your predicted annual revenue will dramatically
compound any poorly chosen monthly values used in your model. For
example, if your monthly revenue is off by $10K, your annual revenue
will be off by $120K. Keep this in mind as you serve up the predicted
annual revenue to management or to your client.

Selling Products and Services

with Search Usability
Let’s take a look at how we leverage search usability (SEO and web
usability) to contribute to the bottom line for websites with a primary
goal of selling products or services.
Selling products and services includes both commercial tangibles like
sneakers and intangibles like cell phone services. Nonprofits have just
as much to gain as commercial sites from the “products and services”
model if you consider your items, such as annual memberships, an
intangible item. Let’s first take a look at factors that will determine the
ROI of your search usability investment for selling products, services,
and memberships.
 SEARCH USABILITY METRICS BY SITE TYPE 91

For example, to calculate the monthly revenue in the products and

service model, you use the following formula:
(Monthly unique visits µ Page view to conversion ratio) µ
Average purchase = Monthly revenue

" Page view to conversion ratio: The page view to conversion

ratio states that of X number of unique visitors who land on a
page that allows for a transaction (such as an add to cart page)
you can expect Y to make a transaction. If you were to apply
this to a website that sells ear muffs, it would look something
like this: of 100 unique visitors who land on the ear muffs
page, you can assume one visitor will buy the ear muffs. The
assumed page view to conversion ratio in this case is 100 to 1,
or 1 percent.
" Current and prospective traffic: This is the traffic you cur-
rently have and incremental gains to project future traffic.
In this example, we will use 100,000 as the baseline for the
monthly unique visits and incremental gains of 5 percent to
estimate prospective traffic generated by SEO efforts.
" Monthly unique visits: You use this number to establish a
baseline for monthly unique visits. You can establish this num-
ber by looking at your web analytics. When establishing your
baseline, be sure to take into account seasonality and other fac-
tors that might skew your data. You should establish a number
that will help predict accurate marketing results. Data that is
atypical and not representative of your monthly activity will
result in an error-prone predictive model.
Historical data from your website is the best option for Note SWAG is an acronym
building your models. If don’t have historical data to work for Stupid Wild Ass Guess,
with, try to establish a baseline using third-party research data Smart Wild Ass Guess, and/
or, if available, data from a comparable sister site. Worse case or Scientific Wild Ass Guess.
Ideally, your SWAG will be
scenario, you might have to SWAG your monthly unique visits closer to the scientific end
before you can generate your own historical data. of the spectrum. If you find
yourself in a SWAG situa-
" Average purchase: If your website only features one product,
tion, try using a range of
priced at $50, then your value of average purchase will be easy numbers that shows a
to figure out. If you have other variables, take a look at what conservative view and an
your web analytics are telling you, talk to marketing to establish aggressive view.
a value of an average purchase you can use to predict results.
" Monthly revenue: Monthly revenue is the result of multiply-
ing predefined quantities:
(Monthly unique visits µ Page view to conversion ratio) µ
Average purchase = Monthly revenue
92 CHAPTER 7: SEARCH USABILITY AND YOUR SITE’S SUCCESS

When you plug in the values of 100,000 monthly unique

visits, a 1 percent page view to conversion ratio, and a $50
average purchase, the monthly revenue looks like this:
(100,000 µ 1%) µ 50 = $50,000

" Annual revenue: You calculate annual revenue by multiplying

your monthly revenue by 12.
Now that we have covered the factors, let’s take a look at Table 7.1 to
see what we can expect in terms of annual revenue from an e-commerce
website by increasing traffic using SEO.
The model in Tables 7.1, 7.2, and 7.3 uses the example of an
e-commerce site that sells a tangible item at an average purchase of $50.
“Average purchase” can easily be replaced with “average cost of service”
for an e-commerce site that sells intangibles or “average cost of member-
ship” for a nonprofit site whose organization is supported by member
contributions. “Average purchase” can also be substituted with “average
value of lead” for websites with the primary goal of delivering leads to
a sales force. You can modify the headings in these tables according to
your business model.

Table 7.1 Products and services model with SEO improvement

CURRENT AND MONTHLY PAGE VIEW TO AVERAGE MONTHLY ANNUAL
PROSPECTIVE TRAFFIC UNIQUE VISITS CONVERSION RATIO PURCHASE REVENUE REVENUE

Current traffic 100,000 1.00% $50 $50,000 $600,000

Plus 5% 105,000 1.00% $50 $52,500 $630,000
Plus 10% 110,000 1.00% $50 $55,000 $660,000
Plus 15% 115,000 1.00% $50 $57,500 $690,000

Note A 5 percent increase The only factor that changes in Table 7.1 is the monthly unique visits.
in SEO traffic resulting Even at a 1 percent conversion rate and an average purchase of $50, you
in an increase of 5,000 see the increase of $90,000 ($690,00 – $600,000) in annual revenue with
monthly unique visits may
not be realistic for smaller
a 15 percent increase in SEO traffic.
e-commerce websites. This Using the same model, let’s take a look at annual results when you
doesn’t mean that your site improve the search usability of your website by making it easier for users
won’t benefit from SEO
to find desired products and complete tasks such as add to cart.
efforts, it just means that it
might take longer to achieve In Table 7.2, the factor that changes is the page view to conversion
your ROI. ratio. At an incremental gain of .25 percent, retaining the average purchase
of $50 and the 100,000 monthly unique visits results in an increase of
$450,000 ($1,050,000 – $600,000) in annual revenue when comparing an
increase of 1.00 percent to 1.75 percent page view to conversion ratio.
 SEARCH USABILITY METRICS BY SITE TYPE 93

Table 7.2 assumes that there is no increase in the monthly unique

visits from SEO efforts, but rather the increase in annual revenue
hinges solely on the page view to conversion ratio that was boosted by
improving the search usability of the page. The boost in page view to
conversion ratio might be a result of making the “add to cart” button
more apparent by
" Removing distracting elements from the page
" Increasing the size of the “add to cart” button
" Using a more eye-catching color for the “add to cart” button

Table 7.2 Products and services model with increased usability

CURRENT AND MONTHLY PAGE VIEW TO AVERAGE MONTHLY ANNUAL
PROSPECTIVE TRAFFIC UNIQUE VISITS CONVERSION RATIO PURCHASE REVENUE REVENUE

Current traffic 100,000 1.00% $50 $50,000 $600,000

0% 100,000 1.25% $50 $62,500 $750,000
0% 100,000 1.50% $50 $75,000 $900,000
0% 100,000 1.75% $50 $87,500 $1,050,000

You’ve seen how increasing traffic via SEO efforts and web usability,
as separate initiatives, gives us a lift in revenue. But what about when you
use them in concert to implement a holistic search usability program?
Table 7.3 shows the potential gain from increasing monthly unique
visits and page view to conversion ratio together. Looking at the most
optimistic view of increasing traffic by 15 percent to 115,000 monthly
unique visits and page view to conversion ratio of .75 percent from 1.00
percent to 1.75 percent, you see an increase of $607,500 in annual rev-
enue as compared to our baseline of $600,000 ($1,207,500 – $600,000).
In the more conservative scenario of increasing traffic by 5 percent to
105,000 monthly unique visits and page view to conversion ratio of
.25 percent from 1.00 percent to 1.25 percent, you see an increase of
$187,500 ($787,500 – $600,000) in annual revenue as compared to the
baseline of $600,000.

Table 7.3 Products and services model with holistic approach of both usability and SEO efforts
CURRENT AND MONTHLY PAGE VIEW TO AVERAGE MONTHLY ANNUAL
PROSPECTIVE TRAFFIC UNIQUE VISITS CONVERSION RATIO PURCHASE REVENUE REVENUE

Current traffic 100,000 1.00% $50 $50,000 $600,000

Plus 5% 105,000 1.25% $50 $65,625 $787,500
Plus 10% 110,000 1.50% $50 $82,500 $990,000
Plus 15% 115,000 1.75% $50 $100,625 $1,207,500
94 CHAPTER 7: SEARCH USABILITY AND YOUR SITE’S SUCCESS

The best scenario is to increase quantity of SEO traffic and improve

the search usability of your website. Not all websites will be able to scale
results as dramatically as Tables 7.1, 7.2, and 7.3 illustrate. Websites that
have more modest traffic will still see results from implementing a holistic
search usability initiative, but it will take longer to realize the ROI.
In many cases, when you improve both SEO and web usability you
achieve a multiplicative effect that compounds benefits instead of an
additive effect. In other words, search usability is often greater than the
sum of its parts.

Selling Ads with Search Usability

Online ads come in various shapes, sizes, and technologies. One thing
they have in common is the advertising model used by many news and
information websites. The advertising model is CPM (cost-per-thousand)
in which the cost of advertising is based on the number of times an ad
loads on a web page. We’ll take a look at some examples based on a CPM
of $1 to illustrate our how dramatic your results can be at a low CPM.
Depending on the site, the typical CPM can be well over $1.
Let’s take a look at how to leverage search usability to contribute
to the bottom line for websites with “products and services” advertising
models to sell online ads. Let’s first take a look at the factors that will
help determine the ROI of our search usability investment.
For example, to calculate the monthly revenue in the selling ads with
the search usability model you use the formula:
(Monthly page views µ Average page views) µ CPM = Monthly revenue

" Page views: A page view, or page impression, is when a single

web page loads in a browser as a result of a user clicking a link
or a user typing a URL directly into the browser address bar.
" Average page views: Page views are commonly thought of in
terms of average page views per visitor. Average page views
are the average number of pages a user views during a website
session on a given website.
" CPM: CPM, or cost-per-thousand, is an advertising model
in which an advertiser is charged per thousand times their
ad loads on a page. The M in CPM represents the Roman
numeral for 1,000.
" Monthly revenue: Monthly revenue is the result of multiply-
ing our predefined quantities:
(Monthly page views µ Average page views) µ CPM =
Monthly revenue
 SEARCH USABILITY METRICS BY SITE TYPE 95

When you plug in the values of 100,000 monthly unique

page views, a 2.25 average page view, and a $1 CPM, the
monthly revenue looks like this:
(100,000 µ 2.25) µ $1 = $225,000

" Annual revenue: You generate annual revenue by multiplying

the monthly revenue by 12.
Let’s take a look at what you can expect from search usability in
relation to ROI now that you know the factors that make up your CPM
model. Table 7.4 shows the annual revenue results for an information
website by increasing page views via SEO efforts.

Table 7.4 CPM model with increased annual revenue via SEO efforts
CURRENT AND MONTHLY UNIQUE AVERAGE CPM MONTHLY ANNUAL
PROSPECTIVE TRAFFIC PAGE VIEWS PAGE VIEWS REVENUE REVENUE

Current traffic 100,000 2.25 $1 $225,000 $2,700,000

Plus 5% 105,000 2.25 $1 $236,250 $2,835,000
Plus 10% 110,000 2.25 $1 $247,500 $2,970,000
Plus 15% 115,000 2.25 $1 $258,750 $3,105,000

Table 7.4 shows us an increase in annual revenue by increasing the

number of pages viewed via SEO efforts. The only factor that changes in
Table 7.4 is the monthly page views. Note the increase in annual revenue by
increasing the monthly page views by 5, 10, and 15 percent. Even at a CPM
of $1 the increase in the annual revenue is $405,000 ($3,105,000 – $2,700,000)
between current traffic and a traffic increased by 15 percent.
Using the same model, let’s take a look at the annual results when
you don’t increase the SEO traffic but increase the average page views
by improving the search usability of the website.
In Table 7.5, the only variable that changes is the average page
views. Using an incremental gain of .25 page views, you see an increase
of $900,000 ($3,600,000 – $2,700,000) in annual revenue when comparing
2.25 average page views to 3.00 average page views.

Table 7.5 CPM model with increased average page view via search usability
CURRENT AND MONTHLY UNIQUE AVERAGE CPM REVENUE FOR ANNUAL
PROSPECTIVE TRAFFIC PAGE VIEWS PAGE VIEWS TRAFFIC REVENUE

Current traffic 100,000 2.25 $1 $225,000 $2,700,000

0% 100,000 2.50 $1 $250,000 $3,000,000
0% 100,000 2.75 $1 $275,000 $3,300,000
0% 100,000 3.00 $1 $300,000 $3,600,000
96 CHAPTER 7: SEARCH USABILITY AND YOUR SITE’S SUCCESS

Tip Many people assume The more difficult your site is to navigate and locate relevant infor-
their sites have good infor- mation the easier it will be to increase average page views via search
mation architecture and usability. Sites with better information architecture and stronger scents
that users can easily find
their desired information.
of information might find themselves fighting for that additional page
Don’t assume; get feedback view from users, while other sites might see a relatively quick increase
from your users via usability in page views due to the poor information scent their site had before
testing.
implementing search usability. Looking at the dramatic increase in annual
revenue, it’s difficult to argue against fighting for that additional page
view no matter what your situation is.
Table 7.5 assumes that there is no increase in the monthly unique
visits from SEO efforts, but rather the increase in annual revenue is 100
percent attributable to increasing the average page view by improving
the search usability. Items that could have contributed to the boost in
average page view include:
" Improving the findability of content by improving the scent of
information
" Featuring related links on content pages to keep users engaged
" Improving the categorization of content
" Improving the readability of your content with headers that
more accurately reflect your content
" Improving the scanability of your content with bullets and
numbered lists
Now, let’s take a look at Table 7.6 to see the increase of annual
revenue in a CPM advertising model when you use a holistic approach
to search usability and SEO.

Table 7.6 CPM model with holistic approach of SEO efforts and search usability efforts
CURRENT AND MONTHLY UNIQUE AVERAGE CPM REVENUE FOR ANNUAL
PROSPECTIVE TRAFFIC PAGE VIEWS PAGE VIEWS TRAFFIC REVENUE

Current traffic 100,000 2.25 $1 $225,000 $2,700,000

Plus 5% 105,000 2.50 $1 $262,500 $3,150,000
Plus 10% 110,000 2.75 $1 $302,500 $3,630,000
Plus 15% 115,000 3.00 $1 $345,000 $4,140,000

Table 7.6 shows the potential increase in annual revenue from

increasing both monthly page views and average page views. In the most
optimistic view shown on the last line of Table 7.6 (which includes a
15 percent increase in SEO traffic to 115,000 monthly unique visits and
increased average page views of .75 from 2.25 to 3.00), you see an increase
 SEARCH USABILITY METRICS BY SITE TYPE 97

of $1,440,000 ($2,700,000 – $4,140,000) in annual revenue as compared to

the baseline on the first row. The more conservative scenario of increasing
traffic by 5 percent to 105,000 monthly unique visits and an average page
view of 2.50 shows an increase of $450,000 ($3,150,000 – $2,700,000) in
annual revenue when compared to the baseline in the first line.

If users can’t find what they are looking for they will
eventually seek out another source for information.
If your advertising model is based on page views and CPM, you
might think pogo-sticking could be beneficial for your advertising model
because it can increase page views and you could charge more for adver-
tising. Pogo-sticking may increase page views, but any financial rewards
accumulated as a result would only be temporary. The more your site’s
organization makes users pogo-stick, the more difficult it will be for users
to find what they’re looking for. If users can’t find what they are looking
for they will eventually seek another source for information. Addition-
ally, website owners typically don’t link to sites they find frustrating to
navigate, so naturally occurring inbound links will become less frequent
and inbound links will become more difficult to acquire.
Some of the user goals that are essential to news and information
websites include being able to find important information quickly and
easily. You can increase page views and resulting revenue by increasing
accessibility to these items by improving the scent of information.

Supporting Existing Products and Services

The average company loses 10 percent of its customers every year. For
every customer a company loses, it needs to acquire a new customer just
to maintain customer numbers. The cost to acquire a new customer can Note Cross-selling is the
be upwards of five times greater or even more than retaining a current act of selling an additional
product or service to an
customer.
existing or interested cus-
Additionally, companies have a better chance of cross-selling to an tomer. An example of a
existing customer than to a former customer or to someone to whom cross-sell on the web would
they have never sold before. Therefore, it’s a profitable strategy to keep be a website that sells
sneakers and features a call-
as much of your current customer base as possible. out for “socks that go great
In an attempt to keep costs down and reduce the number of cus- with these sneakers.”
tomers who “walk out” each year, companies implement customer Another common cross-sell
technique used on the on
retention programs. One customer retention strategy is to offer superior web is when websites say,
support on existing products and services to prevent losing customers “People who bought product
to competitors. X also bought product Y.”
98 CHAPTER 7: SEARCH USABILITY AND YOUR SITE’S SUCCESS

As you know from being a consumer yourself, support comes in

variety of forms. Phone and online are the primary ways you get cus-
tomer support. Phone support can cost a company a lot of money as
the more calls that come in to a call center the more people are needed
to answer those calls. This is also true with click to call and click to chat
technologies, as these technologies require human resources. A website,
however, can serve multiple service requests at any time, day or night.
The more a customer’s support is resolved online the less need there is
to staff a call center or customer service department. That could mean
significant savings for a company’s bottom line. Search usability efforts
can help control operational expenses by reducing the number of phone
calls that customer service receives.
Let’s take a look at the factors that determine the ROI of search
usability investment to lower operational costs of supporting existing
products and services.
For example, to calculate the monthly cost of calls with the search
usability model you use the formula:
(Monthly unique visits µ Calls to customer service ratio) µ
(Customer service hourly rate µ handle time) = Monthly cost of calls

Tip You can determine the " Calls to customer service ratio: This is the ratio of the number
visit to call ratio of your site of site visits to calls an e-commerce website can expect to receive.
by putting a unique cus- A reasonable ratio is 15 percent, based on our experience.
tomer service or call center
number on the website. " Customer service hourly rate: An hourly rate, such as $50, is
Track the number of calls the cost of providing the customer service. Depending on your
that come in through the business model, your customer service might be higher or
unique number and compare
lower. Adjust the dollar amount as necessary. We’ve assumed
them to the visits your web-
site is receiving to determine that the average handle time (AHT) with a customer is a half
your visit to call ratio. hour. The half hour handle time (.5) is baked into the formula.
" Monthly cost of calls: Monthly cost of calls is the result of
Note Call centers and cus- multiplying predefined quantities:
tomer service centers use (Monthly unique visits µ Calls to customer service ratio) µ
average handle time (AHT) (Customer service hourly rate µ handle time) = Monthly cost of calls
as a metric to measure the
average duration of a trans- When you plug in the values of 100,000 monthly unique
action made over the phone. visits, 15 percent calls to customer service, an hourly rate of
AHT includes hold time, talk
$50, and an AHT of .5, the monthly cost of a call looks like this
time, and necessary tasks
that follow the transaction. (100,000 µ 15%) µ ($50 µ .5) = $375,000
AHT is a driving metric
when projecting call center " Annual cost of calls: You generate annual cost of calls by
staffing levels. multiplying the monthly cost of calls by 12.
 SEARCH USABILITY METRICS BY SITE TYPE 99

Let’s take a look in Table 7.7 at how search usability can lower
customer service expenses by reducing phone support. This could apply
to an e-commerce site, a personal banking site, or just about any other
website that costs the business more to have a customer or prospect pick
up the phone than it does to have them self-support online.

Table 7.7 Improving SEO traffic increases the customer support costs
CURRENT AND MONTHLY CALLS TO CUSTOMER SERVICE MONTHLY ANNUAL COST
PROSPECTIVE TRAFFIC UNIQUE VISITS CUSTOMER SERVICE HOURLY RATE COST OF CALLS OF CALLS

Current traffic 100,000 15% $50 $375,000 $4,500,000

Plus 5% 105,000 15% $50 $393,750 $4,725,000
Plus 10% 110,000 15% $50 $412,500 $4,950,000
Plus 15% 115,000 15% $50 $431,250 $5,175,000

Looking at Table 7.7, you see that increasing traffic to a website also
increases the customer service operational costs (or customer support
department costs). SEO efforts might be considered a failure despite
increasing traffic by 15 percent if the operational costs associated with
the increase in traffic don’t outweigh the monetary benefits of our SEO.
To put it another way, if every additional user that SEO drives to the site
is worth an average of $1 of revenue but costs the business an average
of $2 in operational expenses, SEO might be considered a failure due
to the uncalculated associated operational costs to your client. This is
something site owners typically don’t track or anticipate.
SEO results might result in expenses the business didn’t account for
or isn’t interested in supporting. Some of these expenses might include
operational expenses, added head count in the call center or customer
service department, and associated management costs needed to support
the additional calls despite the positive revenue stream.
The question then is how do we use search usability to lower
Tip To lower incoming
operational costs to support the bottom line? The first step is finding out phone calls, put support
what information users are trying to locate that they can’t find online information on areas of the
or simply isn’t available online. Talk to customer service managers, site where you can add
help content under FAQs,
interview customer service representatives, and listen in on customer
on glossary pages, and
support calls to discover what type of website content would be ben- “how to” pages.
eficial to users.
Let’s take a look at the annual cost of calls when you improve search
usability. Table 7.8 uses an incremental decrease of .25 percent in calls to
customer service. Note the effect it has on the annual cost of calls.
100 CHAPTER 7: SEARCH USABILITY AND YOUR SITE’S SUCCESS

Table 7.8 Increasing search usability sufficiency by .25% lowers operational costs
CURRENT AND MONTHLY CALLS TO CUSTOMER SERVICE MONTHLY ANNUAL COST
PROSPECTIVE TRAFFIC UNIQUE VISITS CUSTOMER SERVICE HOURLY RATE COST OF CALLS OF CALLS

Current traffic 100,000 15.00% $50 $375,000 $4,500,000

100,000 14.75% $50 $368,750 $4,425,000
100,000 14.50% $50 $362,500 $4,350,000
100,000 14.25% $50 $356,250 $4,275,000

Table 7.8 assumes that there is no increase in monthly unique visits,

but rather an improvement in search usability that results in a modest
.75 percent decrease in calls to customer service from 15 percent to 14.25
percent as shown in the most optimistic scenario shown in the last line
of Table 7.8. The decrease in call volume of .75 percent saved the com-
pany $225,000 ($4,500,000 – $4,275,000) annually when you compare
the most optimistic scenario located on the last line with the baseline
located on the first line.
Table 7.9 shows what happens when you increase SEO traffic and
reduce call volume. It’s not as cut and dried as the previous models
because in this case; instead of SEO and usability working in concert,
they are actually working against each other from an ROI perspective.
The more traffic you drive, the greater your operational costs. The
better your search usability, the lower your operational expenses. The
best strategy for increasing revenue is to increase traffic and efficiency
by increasing both search and usability efforts. We refer to this as search
usability. This model shows a correlation between the increase in the
annual cost of calls and increase in traffic despite the decrease of calls to
customer service. The annual cost of calls will vary based on the values
used to generate that dollar amount.

Table 7.9 Increased SEO traffic and reduced customer service calls
CURRENT AND MONTHLY CALLS TO CUSTOMER SERVICE MONTHLY ANNUAL COST
PROSPECTIVE TRAFFIC UNIQUE VISITS CUSTOMER SERVICE HOURLY RATE COST OF CALLS OF CALLS

Current traffic 100,000 15.00% $50 $375,000 $4,500,000

Plus 5% 105,000 14.75% $50 $387,188 $4,646,250
Plus 10% 110,000 14.50% $50 $398,750 $4,785,000
Plus 15% 115,000 14.25% $50 $409,688 $4,916,250

You might find that an increase in traffic yields operating expenses

that your search usability can’t lower to a desired cost. On the other
 SEARCH USABILITY FOR EMPLOYEE INTRANETS 101

hand, your search usability might lower your operating expenses and/or
cost-per-conversion to a level that allows for extra money in one budget
to be used to increase traffic through other search engine marketing
initiatives such as pay-per-click.

! SEARCH USABILITY FOR EMPLOYEE

INTRANETS
Intranets are probably the furthest thing from most SEO’s minds as Note Every minute that
intranets have nothing to do with commercial search engines. Intranets employees spend trying to
do, however, have everything to do with website users and the amount of navigate a clumsy interface
is a minute spent away from
time they spend completing tasks on the site like looking up a colleague’s work that contributes to the
phone extension or a change of address form for human resources. Such bottom line.
actions make intranets a prime candidate for search usability.
Intranets are frequently thrown together as side projects. As long
as users can log in to them, intranets are typically considered a success. Tip In the absence of
What is usually not considered is the lost time employees spend trying actual data, many managers
just double the average
to use the haphazardly constructed website. Let’s see how you can use
salary of a given group of
search usability to lower operating expenses by implementing search workers to obtain the
usability into the company intranet. “loaded labor rate.”
There is much that goes into building an intranet other than lower-
ing time on task. This is not the book for designing intranets. However,
we wanted to show the correlation between the ease of use in finding
information on an intranet and lowing business expenses.
For example, to calculate the monthly cost of intranet use by increas-
ing employee efficiency you use the formula:
(Number of employees µ Hours of intranet usage per month) µ
Average hourly rate = Monthly cost of intranet use

" Number of employees: We used 100 for the number of

employees in this model. An organization smaller than
100 employees probably wouldn’t get the cost/benefit out of
having an intranet. An exception would be a company that has
a large portion of its employees working remotely who don’t
have easy access to network drives or company documents.
" Hours of intranet usage per month: We begin our assump-
tion at one hour.
" Average hourly rate: This is the average hourly rate of an
employee. Your average hourly rate of an employee might be
different depending on your business model.
102 CHAPTER 7: SEARCH USABILITY AND YOUR SITE’S SUCCESS

" Monthly cost of intranet use: Monthly cost of intranet use is

the result of multiplying predefined quantities.
(Number of employees µ Hours of intranet usage per month) µ
Average hourly rate = Monthly cost of intranet use
When you plug in the values of 100 employees, 1 hour of
use, and $80 as the average hourly rate, the monthly cost looks
like this:
(100 µ 1) 80 = $8,000

" Annual cost of intranet use: You generate annual cost of

intranet use by multiplying the monthly cost of intranet use
by 12.
Table 7.10 shows how increasing the efficiency of employee intra-
net use can quickly begin lowering operational costs by increasing the
efficiency of employees.

Table 7.10 Decreasing cost by increasing employee efficiency

NUMBER OF INTRANET USAGE AVERAGE MONTHLY COST ANNUAL COST OF ANNUAL
EMPLOYEES PER MONTH (HRS) HOURLY RATE OF INTRANET USE INTRANET USE REVENUE

100 1.00 $80 $8,000 $96,000 $2,700,000

100 0.75 $80 $6,000 $72,000 $2,835,000
100 0.50 $80 $4,000 $48,000 $2,970,000
100 0.25 $80 $2,000 $24,000 $3,105,000

Table 7.10 is an illustration of the old adage “time is money.” By

making information easier to find on the company intranet, you reduce
the time employees lose navigating the intranet and give them back time
that can be better spent contributing to the bottom line.

! COMMON SEARCH USABILITY GOALS

Note Number one SERP We just took a look at some of the most common primary goals based
rankings can actually hurt a
on the type of site. Now, let’s take a look at common search usability
company’s reputation if
users can’t find the informa- goals that are shared by all types of websites.
tion they are looking for on
a SERP listing. If users can’t
find their desired informa- Top SERP Rankings
tion on your site, the num- We need to rethink the perception of number one SERP rankings.
ber one ranking contributes
to a poor user experience
Number one rankings are only valuable if they lead users from the
and equates to a bad brand SERPs to users’ desired content on your website, resulting in an action
experience, which devalues that directly or indirectly contributes to your business’s bottom line.
the brand.
 COMMON SEARCH USABILITY GOALS 103

Only when you can tie your SERP listings back to conversions will the
true value of your SEO efforts be understood.

We need to rethink the perception of number one

SERP rankings.
Imagine you are a client considering hiring one of a couple SEO
firms. Firm A shows results that detail they achieved page one rankings
for 10 very competitive terms in a matter of six months, resulting in a
25 percent increase in natural traffic.
Firm B showed similar increases in natural traffic but also reports Note A conversion might
that the increase of traffic contributed to over $150,000 in new sales in be buying a product or
six months and 90 percent of those purchases were by new visitors. Of downloading a whitepaper.
It might also be getting
those new visitors, 30 percent returned to make another purchase on users to a certain type of
the site, while an additional 10 percent joined the email list. Additionally, page so you can charge
based on historical client data that Firm B was able to communicate, more for advertising reve-
nue. You’ll need to define
we can count on 3 purchases averaging $65 over a period of 12 months
what a conversion is for
from 15 percent of email subscribers. your business.
Most companies would prefer to hire Firm B due to their understand-
ing of ROI and accountability.
Your ability to track your results might be limited by technology
such as web analytics or by marketing’s ability track efforts.
Tip Be conservative with
SEOs should work with marketing to develop metrics that will help
your marketing goals. It’s
tie SEO efforts to conversions. This communicates to marketing profes- better to under-promise and
sionals that SEOs are making an attempt to be financially accountable over-deliver than over-
for their actions. promise and under-deliver.

Bounce Rates
A bounce happens when a user arrives on your website and leaves after
viewing only one page. The act of the user leaving is the bounce. If 100
users land on page A and 40 of them abandon the site without navigating
to another page, page A would have a bounce rate of 40 percent.
There’s a concern among SEOs as to how or if commercial search
engines account for bounce rates. Search engines do consider bounce
rates in their algorithms. Exactly how much weight the search engines
give bounce rate is unknown. Regardless, when measuring your website’s
bounce rates, gathering user feedback is more important than chasing search
engine algorithms. We’ll address gathering user feedback in Chapter 9,
“How to Improve Your Website’s Search Usability.”
Bounce rate alone may not tell the whole story. A high bounce rate
may indicate a poor user experience, or to the contrary that your users
104 CHAPTER 7: SEARCH USABILITY AND YOUR SITE’S SUCCESS

are finding what they are looking for and can quickly move on from
your site.
Examples of a positive bounce rate for an e-commerce site include
users who are comparison shopping. Users can compare the prices of the
same pair of shoes from three different sites within minutes and return
later to complete a purchase. In this case of comparison shopping, the
bounce rate is high but the users found exactly what they were looking
for in a very short amount of time.
Other tasks that contribute to a positive high bounce rate include
ones that can be accomplished quickly, such as locating a phone number,
finding the location of a business, and other tasks that can be completed
from an entry page.
A negative bounce will happen when your SERP listing doesn’t match
the content of its respective page or your users’ intent or if your users
can’t find the information they are looking for. You’ll receive a negative
bounce when your scent of information is weak, if users don’t feel like
they have arrived at the right page for their query, or they don’t find links
to their desired content from the entry page.
Instead of trying to lower or increase a bounce rate per se, look to
improve (increase or decrease) a bounce rate based on your users’ goals.
You can improve your overall bounce rates by making your SERP listings
accurately reflect their respective page content. The more relevant your
content is to your users’ intent, the better your bounce rate will be, be it
higher for pages that immediately help users complete a task like finding
a phone number, or lower for pages like category pages that lead users
to sought-after information.

Time-on-Site and Page Views

Time-on-site is a web analytic metric that measures the total time a user
spends on a website. Page views are the number of pages a user navigates
to during his time on a website.
Many web professionals tend to think the more time-on-site and
the more page views, the better their website is performing. But just
like bounce rate, our interpretation of the time-on-site and page view
numbers might be wrong.
You don’t know the reason one user spent two minutes on your
website viewing three pages and another spent three minutes on your
website viewing five pages. Web analytics data is just that: data. The
numbers don’t tell you why users acted they way they did. You might
like to think that your content is engaging your users and that’s why
they are spending the amount of time on our site viewing the number
 CONCLUSION 105

pages they are viewing. However, the opposite might be true. A longer
time-on-site and more page views per visitor might indicate users pogo-
sticking on your site, not being able to locate their desired information,
and as a result generating a longer time-on-site than desired.
Conversely, less time-on-site and fewer page views are frequently
seen as indicators of a website not engaging users. But this might also
be false if users are finding exactly what they are looking for and can
leave the site satisfied after only spending 30 seconds and two page views
on the site.
The problem is that time-on-site and page views, like bounce rate,
only give enough information for us to speculate why users are spend-
ing the time on the site. If you really want to understand the behavior Note We’ll look at how to
behind time-on-site and page views, you need to perform usability testing, determine the why in
Chapter 9, “How to Improve
which is covered in Chapter 9. Only then will you begin to understand Your Website’s Search
the “why” behind the numbers. Usability.”

! CONCLUSION
Here are the key points of this chapter:
" Businesses make money by increasing revenue or lowering
operating expenses. Search usability can be used to support
both of these objectives.
" Use realistic numbers when developing your search usability
models. Miscalculations in monthly estimates will be multi-
plied 12 fold in your annual results.
" It’s best to use historical data when developing your search
usability models. If you don’t have historical data, use third-
party data if available or SWAG it.
" The most important factor in developing your models is set-
ting client and management expectations correctly. If neces-
sary, use a range of numbers to project your results.
" The more difficult it is to navigate and find information on
your website, the greater the ROI you can achieve by improv-
ing your search usability.
" Many businesses lose 10 percent of their customers each year.
Search usability can reduce customer attrition by making prod-
uct and support information more findable on the website.
" Search usability can increase employee productivity by making
intranet content more findable.
106 CHAPTER 7: SEARCH USABILITY AND YOUR SITE’S SUCCESS

" A number one SERP ranking might be hurting your com-

pany’s reputation if users can’t find their desired content after
clicking it.
" Bounce rates can be easily misinterpreted. A high bounce
rate might mean users found what they were looking for on
a single page and were able to move on from your site.
" Time-on-site and page views, like bounce rates, can be easily
misinterpreted. A longer time-on-site doesn’t mean a better
user experience. It could mean that issuers are trying harder
to find their desired content. The same is true of page views.
" Conduct usability testing to understand the behavior that
drives web analytic data such as bounce rates, time-on-site,
and page views.
CHAPTER 8
SEARCH USABILITY IS
EVERYONE’S JOB
As a search professional, you have a unique opportunity to touch everyone
who works on or has interest in a website. It is in your best interest to
educate everyone so they can help you accomplish your search usability
goals. Having access to many of the players such as writers, developers,
marketers, and designers also makes you a prime candidate to infuse
search usability into the website.

We are not suggesting that one search professional can change an

entire corporate culture overnight. That would be a delusion. But you
can have big impact. This chapter introduces you to the stakeholders
in your site’s search visibility and what you need to know about each
one’s perspective.
108 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

! THE HOLE IN THE FENCE: A STORY

ABOUT PERSPECTIVE
Jimmy used to watch baseball games through a 2-inch hole in the fence
near third base. He would watch a man, standing next to a base, looking
straight ahead for about a half hour and then suddenly run off, only to be
replaced by someone else who did the same thing. The only difference
between the men was that they dressed a little differently.
Every once in a while, someone would run by and stop on the base,
and then after a few minutes run out of view. The runner never did stay
very long.
Occasionally, the baseman would catch a ball and throw the ball.
Sometimes Jimmy would hear the loud roar of a crowd.
Jimmy’s perspective on a baseball game was limited by the view he
had through the hole in the fence (Figure 8.1).
Figure 8.1 Jimmy’s limited
perspective of a baseball
game.

How is Jimmy’s perspective on baseball similar to building websites?

The greater perspective you have about the players involved, the better
you can address business interests, user needs, and the search usability
of your website.
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 109

! IT TAKES A VILLAGE TO BUILD FOR

SEARCH USABILITY
Many skills contribute to designing for search usability. Each skill is vital
to designing findable content and a usable interface.
Everyone who is part of the process has the ability to affect search
usability positively or negatively. The following sections describe some
of the roles that people play in an organization. You may know these
roles by different names; the labels aren’t as important as the influence
each role has on search usability. You’ll know who’s responsible for what
in your organization when you see it. Here’s what you need to know
about how each affects your site’s search usability.
Some web professionals may fill more than one of the following roles
in their day-to-day job. They might be a graphic artist and a copywriter.
Or they might be an information architect and a usability professional.
For the sake of clarity, we will segment out the roles and examine how
the skills associated with each role affect search usability. Let’s look
at how graphic designers, copywriters, marketing communications
specialists, usability professionals, management, market researchers,
information techs (IT), and information architects (IA) affect a website’s
search usability.

Graphic Designers and Search Usability

Graphic designers make web pages look attractive and inviting. They are
also responsible for presenting data clearly, creating consistent graphic
styles, and visually communicating functionality. The choices graphic
designers make are critical to search usability as their visual executions
affect how users scan, read, and browse web pages.

Graphic design techniques to get users’ attention

Graphic design techniques will either attempt to get users’ attention or
allow a page element to fade into the background. The more aggressive
the technique, the more likely it is to grab users’ attention—and the less
attention users will give to other elements on that page.

The choices graphic designers make are critical to

search usability as their visual executions affect how
users scan, read, and browse web pages.
110 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

Not everything can be the most important thing

We have all seen pages where it looks like every element is screaming
for attention.
Everything cannot be the most important thing on the web page.
Home pages are usually the biggest casualty of the “everything is impor-
tant” disease. This line of thinking also finds its way into subpages.
By making everything look equally important, the message you
are sending to users is that that nothing is important, which is the exact
opposite of what you want. Additionally, the resulting web page often
looks cluttered, which can irritate and confuse site visitors.

Everything cannot be the most important thing on the

web page.
Miscommunication is only the beginning of your problems. If you
do not create a visual hierarchy, users cannot locate what they came to
your site to find. Many will abandon your site and go to your competi-
tors’ sites.

How to create a visual hierarchy for search usability

As a guideline, it’s best to be conservative when designing to get users’
attention. Start with less aggressive techniques and work your way up
to stronger ones. You can monitor how your efforts change user behav-
ior by looking at your site analytics and by getting user feedback with
usability testing.
Remember, the following are general guidelines, not absolutes.
Bold, underlined, and colored text
Using bold, underlining, and different-colored text are subtle techniques
that grab the users’ attention. These techniques work well in the body
of a page and make for easy scanning of copy, links, and subheads.
You can draw users’ attention to keywords that keep them engaged
by bolding, underlining, and varying text colors. Consistently feeding
users the scent of information will increase how much time they spend
on your site and increase the page view count.
One of the most common mistakes by graphic designers on web
pages is not letting site visitors know what text is clickable and what is
not. In the sample on the left in Figure 8.2, which text is clickable? Is it
the word download? Is it the words free white paper? Is it the phrase How
to Maximize Your Email Campaigns? In fact, the text does not look click-
able at all. In the sample on the right, the clickable text is much clearer.
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 111

Remember, users determine the text they will click before they move
their mouse.
Figure 8.2 Which example
Download our free white paper Download our free white paper clearly tells users what text
about “How to Maximize Your about “How to Maximize Your is clickable?
Email Campaigns.” Email Campaigns.”

Test to see how implementing bolding, underlining, or different Note Save underlining for
colors increases or decreases time on site and average page views. links. Some sites underline
and add color. Some only
Graphics underline links in the body
Using well-designed graphics can grab users’ attention. Your eyes are of the page. Regular users
are more likely to learn the
naturally drawn to objects that look nice. Hollywood knows this. When conventions than users who
was the last time you saw a movie filled with plain-looking people? Plain- only go to a site once or
looking people movies do not sell. Pretty people movies? They sell. twice. Still, if you want users
to be certain that text is a
The same is true on the web. Graphics, if they are done well, look
link, underline the text.
more visually appealing than HTML and grab your attention faster than
bolding, underlining, or color variations. Your eyes naturally go from
areas of heavy color concentration to lighter color concentration.
Sometimes, users click graphic images more than they click text, and
vice versa. A savvy graphic designer knows when it is appropriate to use
graphic images only, text only, or a combination of text and graphics to
provide the strongest information scent and the best conversions.
In Figure 8.3 and Figure 8.4, a simple change from text to a graphic
image results in a much higher click-through and conversion rate.
Figure 8.3 The call to action
Compare Prices in this
camera ad is formatted as
a simple text link.

Figures 8.4 By changing

the Compare Prices link to a
graphic image and making it
stand out with a background
color and rounded corners,
more users will respond to
the desired call to action.

Sound and motion

Have you ever received a musical birthday card—the kind that plays a
happy birthday jingle when you open it? You do not even notice how
much the enclosed check is for. You’re too busy listening to the music.
112 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

Sound and motion grab your attention on a web page as quickly as

a catchy little jingle from a birthday card does. If you really want to get
your users’ attention, pull out all the stops and use a Flash movie with
sound. Because of the attention they demand, both sound and motion
communicate to users that they are at the top of your hierarchy.
A word of caution, however: If you are going to use sound, give
users explicit control of the volume and ability to turn it off. Better yet,
default the sound to off and let users turn the sound on at their discre-
tion. Sound blaring out from the speakers generally annoys users and
just might embarrass them at work.
If you are going to use Flash, be sure the technology promotes your
message and doesn’t detract from it. Animation, when used appropriately,
will attract users to take your call to action. It can also help users achieve
their goals. However, if animation distracts users from completing their
goals, then it will irritate them and often cause them to leave the site.
Additionally, Flash is not a favorite with the commercial web search
engines. Even though they can crawl Flash, most web pages that tend to
rank well over time are not built completely in Flash.

Web Copywriters and Search Usability

You do not read information on the web the same way you do in print.
Web users primarily scan copy, look for trigger keywords, and click the
graphics or text that carries the scent of information for content that
strikes their interest.
You read, in the traditional sense, after you find detailed information
that matches a query and want to know more about a topic.
Your copy either encourages users to continue on to deeper pages
of your site, or the copy confuses them and they use the back button to
return to the SERP to refine their query.

Don’t let your copy lose the scent of information

Site visitors use copy cues and links on the web to navigate the same way
drivers use street signs to navigate on the road. If you do not provide the
right cues in your web copy, users lose the scent of information and get
lost just like drivers get lost if there are no street signs to follow or if the
street signs are poorly labeled.
Let’s take it from the driver’s point of view. Imagine you are driving
in unfamiliar territory. You have no global positioning system (GPS),
Google Maps, or access to directions of any kind. You are going to have
to drive on local roads and major highways to reach your destination,
and you depend on street signs to guide you.
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 113

It would be helpful if the local roads had signs that prompted you
when you needed to make a turn, reading “This way to your destination”
with an arrow pointing. Think of these prompting signs as the equivalent
to hyperlinks on your website pages and your listing in the SERPs.
The road you turn down should confirm you made the correct turn
by providing a confirmation sign. It should read “This way to your des-
tination” with an arrow that now points straight ahead. Think of these
Note An HTML anchor link,
confirmation signs as the equivalent of headers that confirm you landed or jump link, is a link that
on the right page from a link or a subhead that confirms you landed on takes a user from one part of
the right section of a page from an anchor link. a page to another part of the
same page. It can also be
Think of the highway signs as the equivalent of the overall scent used to take a user to the
of information. If the driver isn’t constantly fed confirmation during his middle of an HTML page
journey, he might get confused and go back a mile or so to try to pick up from another page. A com-
mon use of an anchor link is
the scent of information, similar to a user who clicks the back button.
the “return to top” link found
Let’s see how to help users get to their destination by providing at the bottom of web pages
them a scent of information with copy. or sections on a web page.

Identify critical copy areas for including the information scent

Have you ever seen on TV how doctors and nurses communicate with
each other during surgery (Figure 8.5)? It goes something like this:
" Doctor: “Scalpel.” (The doctor asks the nurse for a scalpel.)
" Nurse: “Scalpel.” (The nurse hands the scalpel to the doctor.)
" Doctor: “Clamp.” (The doctor asks the nurse for a clamp.)
" Nurse: “Clamp.” (The nurse hands the clamp to the doctor.)

Figure 8.5 This is not what

the doctor ordered.
114 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

On the web, users ask for things by clicking words and images. A
web page answers by displaying information that contextually matches
the hyperlinked text or image the user clicked. If the web page doesn’t
return the appropriate content, the user can get lost.
When users click a link, they need to see the words, or a close varia-
tion of the words they clicked below areas of the destination page to
help orient themselves.
Take a look at Figure 8.6 to see how the CIA’s careers page satisfies
a Google query for cia careers.

f
w

r
t
u through d

Figure 8.6 A Google query for cia careers results in the CIA’s careers page.

The scent of information also applies to pictures and captions. Photo

captions are second only to headlines for readership in newspapers. They
are also big attention grabbers on the web. If you are using photos on
your pages, be sure to use relevant, informative, precise, and keyword-
rich captions.
Putting the appropriate copy in specific parts of the web page
will either help users follow the scent of information or hinder them.
Table 8.1 analyzes how the elements on the CIA careers page satisfy
the Google query.
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 115

Table 8.1 CIA Career page elements.

PAGE ELEMENT COPY COMMENTS

1 Title tag Careers at CIA–Central Good

Intelligence Agency
2 Heading Careers at CIA Good
3 Subheading Doesn’t exist on this page. Adding a subheading using keywords on the page
could improve scanning. It might also give commercial
search engines additional content to index.
4 Introductory paragraph View our available career Good
opportunities by visiting
the following pages:
5 Bulleted lists Various job titles Good
(see hypertext links
below for more)
6 Captions Doesn’t exist on this page. Adding a picture to the body of the page that comple-
ments the content contextually could help users orient
themselves. Adding a caption to the picture would give
commercial search engines additional content to index.
7 Hypertext link Analytical Positions Good
8 Hypertext link Clandestine Service Good
9 Hypertext link Language Positions Good
10 Hypertext link Scientists, Engineers Good
and Technology
11 Hypertext link Support Services Good
12 Hypertext link View All Jobs Good
13 Breadcrumb links CIA Home > Careers > Good
Careers at CIA
14 Image alternative text NA The image near the header is a CSS background image
that doesn’t allow for alternative text. An image that
used keywords in the alternative text would give search
engines indexable content and give visually impaired
users additional content to help them interpret the page.
15 URL https://www.cia.gov/ Good
careers/jobs/index.html

Reflect the users’ language

Reflecting the users’ language in the text of a web page in the areas
commonly scanned and also on the search engine results pages will
help users
" Navigate search engine results pages (SERPs)
" Navigate web pages
" Orient to the page’s content
116 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

" Accomplish tasks

" Scan more easily
" Reduce pogo-sticking
" Reduce back button use
Copywriters should have access to the results of keyword research
to understand what words and phrases users use in their queries when
searching on commercial search engines and scanning, foraging, and
browsing on your website.
If possible, web copywriters should observe usability tests, talk with
focus groups, and have access to other market research noting the words
users use to describe products and tasks. Be aware that different users
may search differently:
" Younger users may search differently than older users.
Research indicates that users over age 60 tend to use links
more than search. Also, these users construct less useful search
criteria than younger users, thus taking longer to complete the
task, or failing altogether.
" Users with more subject matter experience may use different
words to search than users with less.
" Users from one region may search differently than users from
another region. For example, here are three different names
for what is essentially carbonated sugar water:
" Pop in the Midwestern area of the United States
" Soft drink in Australia
" Soda in New Jersey

Format copy for search usability

Formatting copy is as important to search usability as the keywords
used in your web copy. Consider the following writing guidelines for
search usability:
" Write in a journalistic style using an inverted pyramid with the
most important information at the top of the page and above
the fold.
" Use short paragraphs, two to three sentences each, similar
to newspaper paragraphs. To aid scanning, consider using a
header for every two to three short paragraphs, and for every
paragraph longer than five to seven sentences.
" Mix paragraphs with long, short, and medium-length sen-
tences. Keep within a 20-word maximum for a sentence.
" Use bullets in lieu of paragraphs to help users scan content.
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 117

" Use numbered steps if you are writing instructions, such as

recipes for cooking.
" Present information to users in snack sizes. Avoid serving up
large portions of text.
" Get to the point with your copy. Avoid using overblown exces-
sive hype.
" Left-hand justify your copy.
" Make body copy hypertext about 7 to 12 words long.

Marketing Communications
and Search Usability
Marketing communications professionals know every touch point, includ-
ing a website, has the potential to build a relationship with customers.
This makes marketing communications a prime candidate to help infuse
search usability because it can have a big impact on branding.

Bad search usability equals bad brand experience

A poor search usability experience gives you a similar feeling to what you
get when you are left on hold listening to “Your call is very important
to us. Please hold and your call will be answered in the order in which
it was received.”
Do you really feel your call is important after you have been on hold
for 10 minutes? No.
This type of experience devalues the brand.
Users feel the same way about poor website experiences. The more
users are forced to muddle through your website not finding what they
are looking for, the more your website communicates a negative brand
experience.

Four user behaviors to tip you off that your

website might be degrading your brand
User behaviors that can tip you off that your site is degrading your
brand are
" Back button use Note Jared Spool refers
" Pogo-sticking to the back button as the
“button of doom.”
" Site search use
" Site maps (wayfinder) use

Each of these behaviors communicates that users can’t find what they
are looking for by using the primary navigation. If you discover your users
118 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

exhibiting these behaviors, there’s a good chance that your information

architecture doesn’t make sense or your website’s scent of information
is not strong enough.
Note Brand equity refers to A website that doesn’t help users find what they are looking for is
the value a consumer places a website that devalues your brand equity—brand equity that you have
on your brand. built up through the years with thousands or millions of dollars spent on
radio, billboards, TV, pay-per-click advertising, and direct mail.
Note Learn more about In addition to devaluing branding, back button use, pogo-sticking,
back button use, pogo- and the use of supplemental navigation items like site search and site
sticking, and site search maps contribute significantly to decreased conversions.
behavior from Jared Spools’
July 2008 webinar entitled
“The Scent of a Web Page:
What can you do to improve users’ success rate?
The Five Types of Naviga- By watching some specific areas, you can improve your users’ success
tion Pages” located in the rate on your site.
virtual seminar catalog:
www.uie.com/events/ Reduce back button use and pogo-sticking
virtual_seminars/. Jared
Spool is the CEO of User
You can reduce back button use and pogo-sticking by improving the scent
Interface Engineering of information by using keywords consistently throughout key areas of
(www.uie.com). a website such as headers, breadcrumbs, and clearly defined links.
Reduce site searches
The reason site search contributes so heavily to failure is because most
searches fail users. Site search design generally does not receive the same
attention and budget that other portions of the site get.
aTest your site search. Make sure it is delivering the most helpful
content it can to your users during what may be a very vulnerable time
on your site.
Create a user-friendly site map
You need to move beyond site maps designed solely to give commercial
search engines access to your web pages and work to improve your site maps
to better serve users who might be one click away from abandoning.
User-friendly site maps are easy and cheap to build and implement,
and you hardly have to change your site other than adding a link to the
utility navigation and/or the footer navigation.
A user-friendly site map should provide links to the most important
areas of the website. These areas should include a combination of category
pages and product or information pages that are critical to accomplish-
ing user and business goals. Links should be annotated to describe what
users will find on the respective pages and in subdirectories. The link
annotations should also provide keyword-rich content for the search
engines to spider.
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 119

Smaller websites might be able to get away with providing links to

every page on the site. Larger sites might need to limit site maps to the
most important pages and/or provide multiple site maps in each major
section of the site.
Communicate crisis information
Crises don’t happen every day, but when they do they often strike fast
and without warning.
Depending on the severity of the crisis you may need to
" Post an announcement on the home page that links to your
press release.
" Replace your homepage with a single announcement on your
home page.
" Post a press release in the news section of your website.

In the event of a crisis, be sure the press and anyone else interested can find
" What the company has to say about the crisis.
" Who to contact in the event of a crisis. It could be a PR agency
or an internal employee.
" Company name, location of the company headquarters, web-
site of the company, and executive team.
The last thing you want is the media finding information about your
crisis on someone else’s site or blog because you weren’t prepared.
Acting fast so the media and your customers can find information
on your website can improve your users’ experience, your site’s search
usability, and may actually be an opportunity to build brand trust by
proactively making an effort to get the users to their desired content,
even if it is about a crisis.

Usability Professionals and Search Usability

Usability professionals keep the users as their focal point when develop-
ing a product or service. Usability professionals conduct usability tests
to be sure users can use a website effectively, reducing the time it takes
to learn an interface and to complete tasks. Usability professionals pick
up where search engine optimizers (SEOs) leave off, working to keep
users on a site and getting them to their desired information.
In general, web usability professionals do not take into account two
key items regarding search usability: querying behavior and keyword
research.
120 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

Querying—a behavior usability professionals ignore

Many usability professionals start testing from the home page or an
interior page. Usability professionals typically don’t take into account
the querying behavior performed on commercial search engines that
brings the majority of users to a site. This is a mistake.
Figure 8.7 shows web analytics data for a particular site. This pie
chart shows that the majority of users found the site by entering a query
into a commercial search engine. Why would usability professionals not
consider the behavior of most of its site visitors?
Figure 8.7 Web analytics
data clearly shows most
site visitors reached a page
within this site through a
search engine.

Usability professionals need to look at site analytics to see how users

are reaching the site. What they need to look for is
" Percentage of users who find the site via commercial search
engines
" Entry pages
" Keywords that brought users to entry pages
" Bounce rates
" Conversions attributed by path

The goal is to determine the type of query used to find a particular

page. By understanding the type of query, be it navigational, informational,
or transactional, you can better anticipate the intent of the users and give
them access to information that is most relevant to their query.
This is where linking among pages of your website becomes critical,
as you need to try to address the interests of multiple types of users.

Maximizing an information query

If it is the users’ intent is to find information, you should not only pro-
vide the information they seek, but provide access to additional relevant
information to keep users on the site. Other types of information pages
include
" Frequently Asked Questions (FAQs), Help, or Customer Service
" Press releases
" Tips pages
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 121

For example, as shown in Figure 8.8, the National Cancer Institute

provides links to various related information from the main Pancreatic
Cancer category page, including information on
1. Pancreatic cancer treatments
2. Prevention of pancreatic cancer
3. Screening for pancreatic cancer
4. The page also features “General Cancer Resources” in the
right-hand column, which gives access to more general, but
still highly relevant, information.

Figure 8.8 The category

page for Pancreatic Cancer
at the National Cancer
Institute’s website provides
strong scents of information
that encourage site visitors
to continue viewing pages
within the Pancreatic Cancer
section. Links are reinforced
with keyword-focused anno-
tations and relevant hyper-
text links.

r
e

Analyzing the quality of search engine traffic

Traffic from the commercial web search engines is user initiated, pre-
qualified, and task-based. Therefore, users arriving at a site via a search
engine should be more interested in your content than users who landed
on your website by clicking links out of curiosity from other websites
or found your site through social networking websites.
122 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

If a good portion of your search engine traffic is bouncing off the

page soon after arriving on your website, you may have to take a good
look at your entry pages and queries that brought users to your website
to find out why users are bounding off. Web analytics may only tell you
half the story. You may have to usability test to get a better idea of why
users aren’t finding what they are looking for.
Ask these questions:
" Scent of information. Are you using the same keywords on
your landing page, be it the home page or interior pages, as was
displayed in the SERP? And if you are, can users easily find it?
" User expectations. Did you meet your users’ expectations of
content that would be on the web page based on which SERP
listing they clicked?
Note A/B and multivariate You can measure the results of your changes in your web analytics
testing are the preferred by comparing two prior months of data or running A/B or multivariate
test methods for comparing testing to see if you were able to
two prior months of data.
These are sometimes " Lower bounce rate of your targeted users from this page
referred to as a pre-post test " Increase page views for these users (but not at the expense of
and can be influenced by
seasonality, news, and time- pogo-sticking)
sensitive promotions. " Increase conversions attributable to this entry page

Which is it—informational query or transactional query?

It can be difficult to differentiate between informational and transac-
tional queries when looking at web analytics. As a result, you’ll need to
anticipate multiple mental models depending on the query used to find
your web page(s).
In Figure 8.9, the data from Google analytics shows some ambigu-
ous keyword phrases that could qualify for either informational or
transactional queries.
Figure 8.9 Take a look at
the keyword phrase
eco-friendly vacation packages.
Was the searcher looking
for information about avail-
able vacation packages? Or
was he looking to book a
trip? Is it an informational
or transactional query? It’s
a tough call.
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 123

If your information pages are found as a result of transactional

queries, make calls to actions easy to find to help users stay on the
information scent.
Commercial web search engines are not perfect. For example, some-
times informational pages, such as this category page from the National
Cancer Institute website in Figure 8.10, appear in search results along
with transactional pages. Nevertheless, the main calls to action on this
category page are easy to find:
1. The hypertext link to the booklet “What You Need to Know
About™ Cancer of the Pancreas” is located in the center
of the page, and the link is underlined to indicate that it’s
clickable.
2. Likewise, if searchers want information about pancreatic can-
cer treatments, the labels, annotations, and hypertext links are
clearly indicated in the “Treatment” section of the page.
Figure 8.10 National Cancer
Institute informational page
can match a transactional
query.

If your transactional pages are being found for informational queries,

be sure to give sufficient access to other informational pages. Provide a
clear scent of information to related pages that users will find of inter-
est, as shown in Figure 8.11 on the next page. If this e-commerce site’s
124 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

product page appears in search results for an informational query, it helps

guide searchers to other circular saws available on the site.
1. The location breadcrumb link provides both a sense of place
and an information scent should searchers wish to view a list
of available circular saws.
2. The hierarchical hyperlink in the product details section also
guides users to the circular saws subcategory page.
3. The thumbnail photo links and corresponding hypertext links
to other product pages provide a very strong, relevant scent of
information to specific product pages of interest.
4. The global navigation label guides searchers even further up
the site’s hierarchy to a main category page.

Figure 8.11 This type of

product page appropriately
accommodates both infor-
mational and transactional
search engine queries.
q

e
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 125

Keyword research—the overlooked user language

The second factor web usability professionals discount is keyword
research. Using the users’ language is touted in the web usability com-
munity through various research methods including one-on-one inter-
views, focus groups, and surveys, but keyword research is not a widely
promoted method.
Web usability professionals should familiarize themselves with the
paid and free keyword research tools and other keyword research methods
used by search engine optimization professionals.
Using keyword research tools, you’ll see the most popular keywords Tip Here are a few
users use to query, keywords usage trends, and variations of keyword URLs to help you begin
your keyword research:
phrases users favor. https://adwords.google.com/
Some tools that SEO professionals use include Google Adwords select/KeywordToolExternal,
Research Tool (free), and Keyword Discovery and WordTracker (both www.keyworddiscovery.com,
and www.wordtracker.com
of which have free and paid versions).

Early, often, and together

Usability and search professionals should work with each other early
and often during the build and launch of a website. If you wait until the
end of a project to implement search usability, chances are you going to
have to re-engineer part or all of the website. No one likes to hear that
their $50,000 website isn’t paying the bills.
If you are a usability professional coming into a project, be sure to
ask if there is a search professional working on it. Search professionals
should ask who is handling the web usability. Sometimes search and
usability professionals are working on the same website and don’t even
know it.
The earlier you bring search usability into your project the more
likely your website will be successful.

Management and Search Usability

According to a study published in the journal Psychological Science, some-
one’s ability to consider another perspective decreases in proportion to
the level of power in a given relationship. This explains why some of us
have a difficult time selling to upper management.
Just because management frequently thinks they are right doesn’t
mean they are. Let’s take a look at how management’s way of thinking
affects search usability.
126 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

Management often doesn’t manage to ask what’s wanted

Note During a white ele- Have you ever been to a white elephant gift exchange party? A white
phant party, numbers elephant party is where everyone brings a wrapped unwanted gift they
(picked out of a hat before have received and ultimately walks away with someone else’s unwanted
the game) determine the
order in which participants
gift—a gift that’s hopefully more desirable than the one they brought.
pick gifts. If you have never Many of the crummy gifts that find their way to the white elephant
been to a white elephant party come from well-intentioned gift-givers, such as family or close friends.
party, we highly suggest it.
People who would swear they know whom they are buying for.
It’s a great time and a good
opportunity to unload a So, why so many crummy gifts? Crummy gifts happen when people
crummy gift. If you’re lucky, incorrectly or inaccurately assume the preferences of others. Sure, a
you might even go home surprise is always nice, but a gift we are sure to like beats a bad surprise
with a slightly less crummy
gift than the one you any day of the week (Figure 8.12).
brought!

Figure 8.12 They could have at least asked me what I wanted

instead of guessing…

Poor search usability is very similar to receiving a crummy gift. The

difference is that your users don’t politely smile and tell you they love
your gift and rewrap it in hopes of later unloading it on someone else
at a white elephant party. Web users go elsewhere, and you are left to
figure out where they went and what they wanted.
If you do not take the time to understand your users, you can expect
they will abandon your site and go to your competitors’. As a result, a
good portion of your website maintenance will go to correcting your
lack of user understanding.
Not understanding your users will also cost you a lot of money in
lost opportunity. Figure 8.13 shows bounce rates, average time on site,
and percentage of new visitors from a website that management thought
was running great because they assumed they knew their users. That was
until they put web analytics on the site.
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 127

Figure 8.13 Only 3 percent to 4

percent of users return to this site.

The results in Figure 8.13 show a bounce rate of almost 80 percent.

This means that almost four out of five users who hit the website imme-
diately leave. Of the 20 percent who stay, they are staying for less than
45 seconds. The percentage of new visits (almost 97 percent) tells us the
vast majority of users are new visitors. Only 3 percent to 4 percent of
users returned to the site.
To sum it up, almost 100 percent of this site’s visitors are leaving in
under 42 seconds and are never coming back. This is an example of a
website that was built under the assumptions that management could
build a site without any feedback from users.

Know your users before you build

The more you know about your users the better your site will be. What
you want to know before you build is
" What type of information are your users looking for? (How
to satisfy an informational search)
" What do they expect to do on your site? (How to satisfy a
transactional search)
" What motivates users to use your website? (How to engender
trust and emotions)
You have more freedom to make changes in the beginning of the
design process than later in the process. Find out as much as you can
about your users in the beginning so you can apply as much as you
can into the design before the code is written and design freedom is lost
or severely limited.
Try to get as much information as possible from users in the begin-
ning of the project. Use interviews, observations, focus groups, surveys,
and combinations of these data-gathering methods. It will save you hours,
money, and headaches later.

What do your users expect?

There is no way to account for all the types of behaviors users will display
on the search engines or on your website. Search usability is an itera-
tive process. Just like your website, search usability is never 100 percent
complete. There is always something to test. Something to improve.
Something to measure.
128 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

CASE STUDY
INTERNATIONAL TRAVEL SITE

Websites that “use the users’ language” often generate a higher return on investment than
sites that don’t. There are a variety of ways that website owners can determine the “users’
language,” including many keyword research tools that the web search engines provide. Search
engine optimization (SEO) pros use these tools all the time to determine how web searchers
find their client’s official site and their client’s competitors’ sites.
Without direct, one-on-one contact with actual site visitors, keyword research tools can lead
business owners and SEO professionals down the wrong path. Here is an example.
During usability testing on an international travel site, we tested the site’s information architec-
ture (categorization and labels) in multiple languages: American English, British English, French,
and Spanish. In particular, we were determining how the keyword phrase student housing was
used in different cultures.
Online keyword research data showed the word for student (in French, étudiant or étudiants)
as a good navigation label because students typically have different income levels and needs
than other groups of people. The keyword research data showed that many French-language
searchers used the word étudiant or étudiants to find accommodations for student travelers
on a limited budget. Therefore, our client wanted the word étudiant implemented in several
different navigation labels.
Usability testing generated some interesting responses. Almost every time the French phrase
for student housing appeared on a web page, test participants noticeably paused and looked
confused. When we asked why they hesitated, they said that the phrase was not appropriate
for a travel website. To them, student housing or student accommodations meant dormitories,
where students can live while attending college.
The French participants then placed the student-related information groupings in one of two
piles:
" Discard. Participants felt that the information label and/or grouping did not belong on
the website at all.
" Do not know. Participants were unsure whether the information label and/or grouping
belonged on the website.

Over 90 percent of French test participants placed student-related information in the discard
pile. In fact, usability tests in the other three languages had nearly the same outcome.
Even though keyword research tools are one means of discovering the users’ language, they should
not be the only way to determine the text on navigation labels and other information scents.
SEO professionals often apply keyword research data to a website and end up creating a
site architecture that is confusing to both search engines and site visitors. It is hard to make
correct inferences without watching users attempting to complete a task you’ve given them.
Usability testing combined with keyword research data provides the most accurate navigation
solutions.
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 129

It is very difficult to design a simple site the first time out of the gate
that meets or exceeds the goals of users and the business.
Why you might not get your website right the first time:
" There are no perfect user interfaces.
" There are no perfect information architectures.
" There are no perfect designs.
" There is no average user.

There are, however

" Smart user interfaces
" Smart information architectures
" Smart designs

How to be smart:
" Get feedback from users early in the process, even before
design.
" Apply your findings in your initial design.
" Test your design with a candidate user and sample tasks and Note Testing one user can
get feedback. make the site better for
everyone.
" Repeat.

How management’s decisions affect web usability

We once worked with a company whose flagship product was Internet
fax. Internet fax is a service that allows you to send and receive faxes over
the Internet through your email just like you would send an email. No
fax machine is needed. It’s a very convenient service.
Unfortunately, the company was losing money quarter after quarter.
They decided to be in the document management business despite the fact
that the majority of revenue was generated by Internet faxing. Document
management comes in many flavors. For the sake of this story, know that
Internet faxing is not the same as the document management.
Part of the initiative to be in the document management business
was to position the website to speak to investors. The company made
investors the primary target audience of the website with no regard for
customers or prospects.
Management decided that they would no longer refer to their faxing
services as fax. Instead, they renamed their fax service to a very arbitrary
name. The name they chose didn’t mention fax or have any connection
to fax whatsoever.
The website soon fell out of natural search results. Users who had
formerly located the site via navigational searches due to referrals were
130 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

confused because they couldn’t find the fax services they heard about
from their colleagues and friends. These users abandoned the site.
Users who located the website by means other than search engines
were not able to find the fax service they were looking for. They weren’t
able to orient themselves on the home page because they never found the
word fax anywhere on the page. These users abandoned as well.
The users who didn’t immediately bounce off the home page tried
using the site search. Zero results for fax were found. Other users tried
pogo-sticking around the site, unable to find what they were looking
for. They abandoned too.
Management broke the big rule of using the users’ language in hopes
of being something they were not to investors. And as a result, manage-
ment alienated their prospects and achieved the self-fulfilling prophecy
that the company could not be sustained on Internet fax revenue alone.
The company was sold a few years later.

Market Researchers and Search Usability

Usability testing is misunderstood. If you tell a market researcher you
are interested in usability testing, they will probably come back with
something like:
“Great, I’ll set up a focus group right away!” or “Great, I’ll send out
a survey right away!”
So, what is the difference between usability tests and focus
groups?
Focus groups, surveys, and other research methods are very valu-
able, but they are not usability tests. Market research will not tell you
why users do the following:
" Browse on your website
" Pogo-stick on your website (Figure 8.14)
" Forage on your website
" Scan or read your website’s content
" Use your website’s search functionality (site search)
" Orient on your website
" Perform a query on a commercial search engine
" Click the back button
" If they use site search
" If they use your site map
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 131

Only with usability tests will you get a full picture of how and why
users search on the commercial search engines and on your website.
Figure 8.14 Site visitors
often exhibit pogo-sticking
behavior but do not realize
that they do it.

Why focus groups aren’t usability tests

Focus group participants may tell you that they want specific information
and functionality on your website, but you really do not know if that’s
true until you usability test.
In one study, participants taking part in a focus group on how con-
sumers might evaluate health information on the web suggested that
they wanted to know
" The source of the health information published on the website
" If the information came from an individual or an institution
" If the information was based on science or an individual’s
experience
Corresponding usability tests told a story contrary to the statements
made in the focus groups. Researchers found
" Participants did not actively search for who stood behind the
sites.
" Participants did not actively search for how the information
had been compiled.
The conclusion: People say one thing, but do another. Therefore, do
usability testing if you want to know how users will use your site.
132 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

Why surveys aren’t usability tests

Imagine asking your customers the following survey questions:
1. When searching on Company A’s website, I tend to pogo-stick:
" Strongly Disagree
" Disagree
" Neutral
" Agree
" Strongly Agree
2. I lose the scent of information on Company A’s website:
" Never
" Almost Never
" Sometimes
" Frequently
" Always
3. Rate Company A’s site on its ability to help you orient yourself:
" Poor
" Below Average
" Average
" Above Average
" Outstanding
4. When on Google I primarily perform:
" Navigational queries
" Informational queries
" Transactional queries

Trying to get information from users about their search behavior just
isn’t possible without usability testing. If you want to know how users
search on the web, you need to watch them use the web.

IT and Search Usability

An IT group could be referred to by many different titles depending on
your organization. You may know them as programmers, developers,
software engineers, or information technology (IT). They are a very
valuable resource that is often asked to stretch beyond their skill set in
ways that frequently wreak havoc with users.

Building websites using iterative design

The iterative design model was developed in response to unhappy users
and exceeds budgets of traditional design. Using the iterative design
model, you observe users using the website in many states of develop-
ment, including rough paper prototypes, graphic mockups, and function-
ing interfaces. The iterative design model refines the website building
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 133

process by getting constant feedback from users through multiple rounds

of prototyping, user testing, and analysis.
Developers found the most successful websites get feedback from
users from the beginning of the project. They also found that websites
built using the iterative design model were more likely to go live on time
and within budget.
Even with the success of iterative design, many developers prefer to
build websites the traditional way, delaying search usability efforts until
the project is live. The old model of throwing requirements over the wall
to IT and getting a finished product back in three months doesn’t work. It
never really did. By iteratively testing your search usability efforts before
going live, you can make improvements, thereby achieving greater results
for your client or employer and delivering a better user experience to your
users. Design must be iterative if search usability is to succeed.

Losing the scent of information

The scent of information can be lost in many ways. In the world of devel-
opers, it happens through user errors, URL structure, and site search.
Losing the scent—error messages
What’s the use of driving traffic to a website only to have users abandon
halfway through the checkout process because of an insufficient error
message?
Poorly written error messages results in frustrated users and the loss Note When writing error
of the scent of information. If users cannot figure out what happened messages, write in English,
because of an incoherent error message, they abandon. Programmers French, Japanese, or
whatever language your
are typically the ones who end up writing the error messages. They do users speak. Whatever you
so as an afterthought when working in the code. So what happens when do, do not write in tech-ese.
users encounter a poor error message?
If you’re really lucky, your users will incorrectly blame themselves
for triggering an error message, feel dumb and pick up the phone, use
click to chat, or click to call to have a real person help them.
If you’re really, really lucky, your users will incorrectly blame themselves
for triggering an error, feel dumb, muddle through the error message,
figure out what went wrong, and proceed through the checkout.
However, if your site is like most websites, you will not be so lucky,
and users will abandon at an alarming rate.
Losing the scent—URL structure
IT isn’t concerned with URL structure as long as the URL serves up a
web page. This is true for almost everything that IT touches in terms of
web usability. As long as it works technically, IT is moving on to a new
project. It doesn’t matter that users cannot use it.
134 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

URL structure communicates to users the organization of your web

site. Some sites do this better than others.
Let’s take a look at www.marthastewart.com, specifically the URL
for Martha’s Valentine’s Day recipes. You can tell by looking at the URL
that the page is about Valentine’s Day recipes but it takes a while.
Actual URL: http://www.marthastewart.com/valentines-day-holiday-
recipes?lnc=f804758a65cee010VgnVCM1000003d370a0aRCRD&rsc=
leftnav_holiday_valentines-day
A cleaner URL with less code gibberish would communicate impor-
tant information much quicker.
Proposed URL: http://www.marthastewart.com/holidays/valentines-
day/valentines-day-holiday-recipes
This proposed URL communicates the information architecture and
gives users a sense of what they might find on the site.
To savvy users, the proposed URL communicates that in order to
get to “Valentine’s Day recipes” you have to click through “holidays”
and then “Valentine’s Day.”
The proposed URL communicates that the site might offer information
on holidays other than just Valentine’s Day, because “holidays” is part of
the URL string. The URL also communicates that there might be more
information on Valentine’s Day than just recipes because “valentines-
day-holiday-recipes” is a subdirectory of “valentines-day.”
From a SERP point of view, a more descriptive URL is more intui-
tive for users who are trying to choose from the SERP listings. Compare
the URLs in Martha Stewart’s Valentine’s Day recipe page to Hershey’s.
The Hershey’s URL is more descriptive and contains keywords that
give a good scent of information. The Martha Stewart URL contains
similar keywords, but also contains a string that detracts from the scent
of information by adding unnecessary noise.
Figures 8.15 Martha
Stewart’s complex URL.

Figures 8.16 Hershey’s

descriptive URL.
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 135

Losing the scent—site searches

Using site search doesn’t have to mean losing the scent of information,
but it commonly does. According to Jared Spool, once users turn to site
search their failure rate rises to about 70 percent as opposed to 47 percent
without using site search.
So who owns this 70 percent abandon rate? The responsibility of site
search usually falls under IT, not because it’s an IT job, but because IT or
someone else thinks it’s an IT job. In reality, the primary owners of site
search should be information architects as they are the ones responsible
for the organization of information.
You should consider a few things when creating your site search. First,
consider if you should provide site search at all. If you cannot implement
site search well, you might want to consider skipping it.
If you decide to use it, consider the following:
" Most importantly, try to limit the use of site search. Work
with an information architect on the labels used in primary
and secondary navigation and a usability professional on con-
textual and related links. The more obvious your navigational
elements are, the less your users will use site search and other
supplemental navigation.
" Watch users use your site. At what point do they use site
search? That’s the point where they have lost the scent of infor-
mation. Try to correct the problem to reduce site search use.
Research has shown that following links (instead of searching)
exposes users to more pages they can use. Also, task success
goes up (assuming the link structure creates a clear scent leading
to the goal). On the other hand, some tasks benefit from search:
finding a book using a title or author; or finding a phone num-
ber and address using a person’s name, and so on. These would
be called “exact searches.” Computers serve these goals well.
" Work collectively. Get IT and information architects to work
collaboratively on site search to determine what results show,
the formatting of those results, and the keywords used in the
site search results.
" Observe user interaction with search results. Similar to
their use of commercial search engines, most users do not
go beyond the first page of the search results. Test how users
respond to more than the typical 10 results per page. Change
the number of search results to 20, 30, or 40 results per page.
" Find out what your users are searching for. Mine the data
from your site search to make your website and site search
better. Consider incorporating the language used in site search
136 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

into your website’s information architecture as labels and in

tips, FAQs, and other content pages of your website. Make
sure these new pages show in your site search as well.
Note For more information
and resources to help you
" Anticipate user intent. Many queries entered in site search
build site search, see are going to be informational searches. What other types of
Information Architecture for information can you present that is relevant to their query?
the World Wide Web, Third
Edition by Peter Morville When done well, site search can reduce site abandonment and get
and Louis Rosenfield users back on track.
(O’Reilly Publishing, 2006).
Avoiding unnecessary features
As a rule of thumb, the more features something has the more difficult it is
to find those features and the more difficult it is to use those features.
So what’s the problem, you ask? Developers love features. They love
to put features on websites that users have no interest in.
At what point do more features become unnecessary? Before adding
more features to a website, ask yourself a few questions:
" What is the benefit to the users?
" Do the users value the benefit?
" Did the users express interest in the feature?
" Will the user value the feature on your website?

Features are only cool if users think they’re cool. Users may find
features annoying and distracting. Avoid worshipping the cool. Focus
on the useful and relevant.

Information Architects and Search Usability

Information architects organize the web pages to make content easier to
find. Their work is the foundation of any website and search usability.
Information architects are not graphic artists or programmers. They
aren’t terribly concerned with how a website looks. They do care about
how a website functions, however they do not code the functionality.
Information architects’ work includes
" Defining navigational elements and labels
" Organizing sequence and content of web pages
" Determining breadth and depth of a website navigation scheme
" Defining the appropriate titles for pages
" Developing an optimal framework that supports findability
and search usability
" Developing wireframes (blueprints of web pages) in PowerPoint
or other tools
 IT TAKES A VILLAGE TO BUILD FOR SEARCH USABILITY 137

Depending on the individual and organization, information archi-

tects might also cover
" Task analysis
" Interaction design
" Real-time responsiveness goals
Note If you find yourself
" Usability evaluation, usability testing
going straight to the com-
" Assessing conformance to usability standards puter to design, you’re prob-
ably skipping a few very
Information architects simplify complexity, determine navigational important steps in the web
controls, and understand what information is needed for users to accomplish design process. You could
tasks and find information. Information architects are responsible for not turn a computer on for
days and still be designing a
" Reducing the number of clicks needed to complete a task or to website. You can do all of
find information your upfront work such as
research, information archi-
" Reduce the number of windows needed to complete a task or
tecture, and wireframing
to find information without a computer. You
" Reduce the learning curve of websites so users can complete only really need a computer
when you get into graphic
tasks and find information more efficiently design and coding.

It’s all uphill without good information architecture

One of the biggest and most common mistakes made when building
websites is when graphic designers go straight to Adobe Photoshop or a
similar graphics program and start designing. This is like a construction
company starting to construct a building without a blueprint.
Do you think someone hands a construction company a truckload of
bricks and tells them to start building? Of course not. Then why do web
designers begin building with no information architecture or wireframes
to guide their decision-making? They shouldn’t, but it happens.

Bad information architecture will cripple your search

usability efforts.
The danger of going right into Adobe Photoshop at the beginning
of the design process is that a good-looking design can cover up inferior
information architecture. Bad information architecture will cripple your
search usability efforts.
In addition to covering up bad information architecture, heading straight
into graphic design at the beginning of the process frequently takes the
attention away from the information architecture as the team becomes
consumed with conversations about colors, fonts, and photos.
138 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

Look and feel, and the emotions evoked from images, are very
important, but those shouldn’t be pursued at the expense of the website
information architecture. More thought and discussion is typically put
into a photograph that can be easily swapped out than the backbone of
the site—the information architecture. This needs to change if search
usability is to succeed.

ARE TWO PAGES BETTER THAN ONE?

A SHORT STORY ABOUT INFORMATION ARCHITECTURE

A colleague of ours told us this story. He was hired as an SEO consultant but knew a thing or
two about information architecture. He noticed some architectural inconsistencies and spam-
ming concerns on the site and brought them up to the marketing manager responsible for
managing the website.
SEO consultant: “Did you know you have two of the exact same pages on your site?”
Client: “No, we do not.”
SEO consultant: “Yes, you do.” (Emails the client the two URLs)
Client: “Oh, right. I think it’s OK.”
SEO consultant: “The pages are being used incorrectly. The higher level category page should
be a landing page that serves as the table of contents for this section. It should be used as a
pathway to get users to their desired content.”
Client: “Uh-huh.”
SEO consultant: “The other page is one of five content pages that houses the actual informa-
tion under the section. It’s a parent-child relationship. It’s what people expect from a website
and is how your other sections are set up. The way it’s currently set up will be confusing to
the users.”
Client: “Users will figure it out.”
SEO consultant: “Users will figure it out? Have you ever read a book that had the same page
published twice?”
Client: “No. Why would you print the same page twice in book?”
SEO consultant: “You wouldn’t. You shouldn’t print the same page twice on a website either.”
Client: (Long pause) “We’ll change the page.”
If you want to get users to the information they want, you need to set up your information
architecture to make that happen. Our colleague’s experience illustrates how having your feet
in multiple disciplines can help your client build a better website.
 CONCLUSION 139

Information architecture versus quality assurance

In many organizations, more time is spent on quality assurance (QA) than
preparing a website for good information architecture. If your informa-
tion architecture and your wireframes are not thoroughly thought out,
you will be wasting your time proofing a site that isn’t worth proofing
to begin with.
This misbalance between IA and QA is partially due to
" The lack of understanding what an information architect is
" The lack of understanding how an information architect’s
work affects website design and search usability
" An unqualified need to see a tangible graphic design by
management
" Unrealistic timelines that result in projects that exceed budgets
and compromise quality
Top-loading your process with surveys, interviews, thesaurus
research, web logs, usability tests, and keyword research will culminate
in a solid information architecture that will act as your foundation for
search usability.

! CONCLUSION
Key points of this chapter:
" Search usability is everyone’s job. Many times there are no
clear lines to define when an SEO professional’s job begins and
ends and when a graphic designer, copywriter, or information
architect’s job begins and ends.
" It’s all about perspective. The more you understand other web
professional disciplines the better you can manage, implement,
and measure search usability.
" Not every page element can be the most important thing on
the web page. Page elements must be created with a visual
hierarchy. If not, everything will appear to be equally visually
weighted and search usability will suffer. Use design principles
to consciously evoke a “top down” flow or hierarchy.
" By using the users’ language in critical areas of web pages and
formatting copy appropriately, writers can keep feeding users
the scent of information.
" A poor search usability experience equals a poor brand experi-
ence. Improve your users’ brand experience online by reducing
their use of the back button and pogo-sticking.
140 CHAPTER 8: SEARCH USABILITY IS EVERYONE’S JOB

" If you are going to use supplemental navigation like site maps
and site search, put the enough resources behind them to
make them work for your users.
" Crises happen. Plan to manage your brand online by consider-
ing how users will look for information in times of crisis.
" Usability professionals need to stop ignoring querying behav-
ior. They must also include keyword research when they talk
about using the users’ language.
" You can expect your users who find your site in search engines
to be more targeted and better-quality traffic than other
sources of online traffic.
" Instead of assuming you know what users want from your
site, ask them. The more you know about your users in the
beginning of the process, the less reworking you will need to
do later.
" Focus groups aren’t usability tests. Surveys aren’t usability
tests. Usability tests are usability tests. Ask people who could
be your site users to show you how to complete a given task.
Ask them to “think out loud.” Make notes of their experiences.
Tell them “we are not testing you, we are testing the design.”
Take notes on problems.
" Interactive development, showing your site to potential users
at each stage, leads to successful search usability.
" Losing the scent of information can happen in many ways. In
the world of developers it results from user errors, URL struc-
ture, and site search.
" More features mean more usability problems for users. More
features also mean reduced success in finding what users are
truly looking for.
" Search usability begins with good information architecture.
CHAPTER 9
HOW TO IMPROVE
YOUR WEBSITE’S
SEARCH USABILITY
This chapter gives you web usability tests you can conduct to improve
your website’s search usability and increase conversions. It also addresses
myths and misconceptions about search and web usability.

You don’t need to go through a rigorous recruiting process to find par-

ticipants to run the usability tests described in this chapter. If you have
research personnel and budget that you can leverage to get feedback from
users, great. If you find yourself in an “if you want something done right,
you have to do it yourself” scenario, try talking with a friend, neighbor,
or colleague to get a different perspective on your website.

The idea is to get valuable, actionable feedback to continually improve

your website’s search and web usability. Let’s take a look at a few usability
tests and other techniques that will help.
142 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

! IDEAL TEST PARTICIPANTS

Before you can conduct any usability test, you’ll need test participants.
You want to find people who are representative users of your website. For
example, if you’re trying to sell tennis shoes on your site, test people who
wear tennis shoes. If you are testing a travel website, choose participants
who book flights and/or car rentals online.
Some websites might be more difficult to recruit for than others. If
you can’t find ideal test candidates, ask the next-best candidates to “role
play” your users. Give them some background or create a story to help
them identify with the part. The more detailed you can make your story,
the better your participants will be able to play the role.
If you test participants that you feel are less-than-ideal candidates,
consider the qualities that make them not ideal and factor those qualities
into your assessments of your test findings. Not being able to find the
ideal test participants shouldn’t prevent you from testing your site.
Generally speaking, there are a few things that make some test can-
didates better than others, whether they are your ideal test participants
or not.
Qualities you want in a test participant
" Someone who is comfortable thinking out loud.
" Someone who isn’t familiar with your site, assuming you need
to test for first-time user interaction.
" Someone who is familiar with your site, assuming you want to
improve your website for repeat visitors.
" Someone who will be able to understand the language.

Qualities you want to avoid in a test participant

" Someone who is trying to please you and doesn’t want to hurt
your feelings.
" Someone who doesn’t use the web (email use is not the same
as web use). You are not designing for the newbie who is just
getting familiar with scroll bars and radio buttons.
" Someone who finds your site easy because he or she has
extensive domain knowledge—more than most users do. His
or her performance on your site might overshadow issues for
less-knowledgeable users.
" Someone who doesn’t have the minimal amount of domain
knowledge that you need your user to have. This user will be
stumped at every turn on your website.
 REVERSE CARD SORT TEST 143

Additionally, you’ll also want to avoid web developers and web

designers in your test group. Developers will probably view your site as
a challenge and try to figure it out, and web designers will tend to tell
you how to redesign your site rather than showing you how they would
actually use it.
Be sure to communicate to your test participants that it’s the website
that is being tested, not them. They will still feel like they are the ones
being tested. Inevitably, you will get at least one user per test group who
asks, “How’d I do?” Tell them they gave great feedback, and thank them
for their time.
Additionally, tell the participants not to worry about hurting your
feelings. Tell them that testing is a way to learn how people interact with
the website so that you can improve it for everyone who uses it. You can
expect to pay up to $100 an hour for your participants’ time. This might
seem like a lot to some people, but it’s pennies compared to the time it
takes when your test participants don’t show, and you need to start your
recruiting from scratch all over again. If you are testing with co-workers,
offer to buy them lunch or give them movie tickets. If you are a small
business owner in need of participants, offer dinner to family members
and friends in exchange for being a test participant.

! REVERSE CARD SORT TEST

The reverse card sort test will help you fit new content into an existing
website in a way that makes your products and services easy to find for
search engines and for users. This test will also help you assess whether
adding new content is a natural fit for your existing website, or if the
content is a patch for a compromised information architecture.
Most usability tests need about five to 10 participants per group
to generate meaningful results. Some research suggests that you need
upwards of 15 users for a reverse card sort test. Generally speaking, you
know you have tested with enough users when you stop hearing feedback
unique to each participant and start hearing consistent, similar feedback
from your test group. For reverse card sorting, you are looking for trends
in how users organize the information. An 80 percent or better accuracy
rate is a reasonable goal for reverse card sorting.
Before we get to how to apply the reverse card sort test to an existing
website, let’s first take a look how to conduct the test using cards. You can
use index cards or pre-formatted labels for your reverse card sort testing.
Use what you have handy and/or feel most comfortable using. If you
hand-write the labels on the cards, make sure your writing is legible.
144 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

In Figure 9.1, notice the predefined categories at the top that include
colors, flowers, and fruit. We’ve also included a few blank cards in case
participants come up with a new (and possibly better) category or cat-
egory label. Also, notice the nine large cards below the category cards
that include red, peach, green, and so on.
Under which category would you expect to find each of the nine
cards? Would you prefer to create your own category labels?
Figure 9.1 The layout of a
reverse card sort test before Reverse Card Sort Test
a test participant catego-
Categories:
rizes the cards. Note the
predefined categories and Colors Flowers Fruit Blank
cards
blank cards at the top and
the predefined cards with
labels grouped below.

Red Peach Green

Cherry Lemon Orange

Blue Violet Rose

Figure 9.2 shows two possible results of the reverse card sort test.
One test participant classified orange as a fruit and another test participant
classified orange as a color. When there is more than one category that a
test participant identifies a card with, they might look to you to confirm
their choice or might tell you that they would put the card under more
than one category. You might have had the same thoughts as you were
trying to organize the cards in Figure 9.1. As the conductor of the test,
you should encourage test participants to choose a category and remind
them there is no wrong answer. You want them to commit to a category
so you understand how they would group the information.
 REVERSE CARD SORT TEST 145

Figure 9.2 Orange was clas-

sified as a color by one test
Colors Fruit Flowers Colors Fruit Flowers
participant and as a fruit by
another.

Red Cherry Rose Red Cherry Rose

Blue Peach Violet Blue Peach Violet

Green Lemon Green Lemon

Orange Orange

Figure 9.3 is another example of how a test participant categorized

the cards. In this case, the test participant thought that cherry referred to
cherry blossom, peach referred to peach blossom, and lemon referred to
lemon blossom. They saw no need to use the fruit category.
Figure 9.3 Some participants didn’t
even use the fruit category.
Colors Flowers

Red Cherry Rose

Blue Peach Violet

Green Lemon

Orange
146 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

Let’s apply the reverse card sort test to the National Cancer Institute
website in Figure 9.4. In lieu of colors, flowers, and fruit, use the top-
level navigation tabs as main categories of the website.
Figure 9.4 In a reverse card
sort usability test, you do
not have to show the entire
web page. You can, as in this
example, only show partici-
pants the main categories. Looking at the National Cancer Institute primary navigation in
Figure 9.4, where you would expect to find the following information:
1. Latest information about breast cancer treatment
2. Cancer information for physicians and other healthcare
professionals
3. Calendar of events
4. Most common types of cancer
5. Risk factors for pancreatic cancer
You generally encounter one of three scenarios while conducting the
reverse card sort test:
" Scenario 1: Everyone places the content under the same pri-
mary navigation link. This scenario is ideal and is also the most
unlikely. Few websites are this cut and dried.
" Scenario 2: The majority of your users place the content
under one primary navigation link while the others place the
content under other primary navigation links. This is the most
likely situation.
" Scenario 3: No matter how many users you test, you don’t see
any trends in how your users organize the content. This occurs
with sites that have poorly labeled primary navigation buttons.
You have nothing to worry about if most of the responses fall into
Scenario 1, when there is consensus among your test participants as to
where they would look for content. Simply add the content under the
category your test participants chose.
In Scenario 2, the majority (as note earlier, an 80 percent accuracy
or better is a reasonable goal) of your users are telling you where they
expect to find the content. Listen to the majority and place your content
where they believe it belongs.
You’ll need to perform an additional step for the remaining users who
looked for your new content under the other primary navigation links.
Ask these users on which page under the primary navigation links they
 ONE-ON-ONE FIELD INTERVIEWS 147

selected would they expect to find the content. These users are telling you
which pages should include a cross-reference link to the new content to
get them back on the scent. For best results, annotate the cross-reference
with keywords to increase the scent of information.
If you find yourself in a Scenario 3 situation, chances are you need
to reexamine your site’s information architecture as the reverse card
sorting isn’t yielding meaningful results.
The reverse card sort test can also help organize content at a sec-
ondary level. Imagine you want to apply keyword and corresponding
content to existing Frequently Asked Questions (FAQs) pages. Imagine
you have a preexisting list of FAQs organized under five categories. You
want to add additional FAQs to your pages but you aren’t sure how to Note A card sort test
organize them. Using the reverse card sort test, show your users the five allows website users to
FAQ categories and ask them under which FAQ category they would group site content in a way
that makes sense to them.
expect to locate the new questions. The idea is that if your web-
Many search engine optimization (SEO) professionals find them- site is organized in a way
selves working with clients who do not want to change their website that makes sense to your
users, they will have an eas-
architecture but still want to implement SEO by adding new optimized ier time locating information
content. The reverse card sort will help organize new content if your on your site. For more infor-
website architecture already makes sense to your users. If the website mation on card sorting go to
structure doesn’t make sense to your users, consider reengineering the http://www.usability.gov/
design/cardsort.html or
architecture as users will have a difficult time finding information on http://en.wikipedia.org/
your site and adding new content will just add to the noise. wiki/Card_sorting.

! ONE-ON-ONE FIELD INTERVIEWS

Use one-on-one interviews to ask users what type of information they
want from your website, and what they want to do when they get to
your website. By understanding the type of information and functionality
your users expect from your site, you can begin to build a customized
keyword list that will help users find your site on the commercial search
engines for information and transactional queries.
One-on-one interviews give you the potential to build a keyword list Tip One-on-one interviews
that is unique to your audience and your website. One-on-one interviews aren’t meant to be a substi-
also give you the opportunity to ask follow-up questions. You can clear tute for online keyword
research tools. Use one-on-
up any misunderstandings and find out more about your users’ needs one interviews in conjunc-
and interests, potentially generating more content for other page types tion with online keyword
such as FAQs and how-to pages. research tools to create a
unique list of keywords to
During the interview, make a list of the keywords users say out loud
optimize your website.
when they express themselves. Note how they talk about performing a task
and the type of information they would like to find on your website.
148 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

After the interview, compare your user feedback with the information
and functionality your site currently features and the priorities attributed
to your content and functionality when you first built your site. Is your
site matching up to users’ expectations or is there a disconnect between
what your users expect and what your website offers?
One-on-one user interviews can very well be the beginning of cre-
ating new categories, subcategories, FAQs, tips, and other content-rich
pages and functionalities for your website.
You can walk away from one-on-one interviews with a competitive
edge if you listen to the subtle and sometimes not-so-subtle differences
in how your users express what they want to find and what they want
to do on your site.

! FUNCTIONAL SALIENCE TEST

Note The functional Functional salience tests help usability professionals determine and
salience test was adapted prioritize the functionality of software applications. The functional
from Eric Schaffer’s salience test was born of the necessity to limit the number of options
Institutionalization of
Usability: A Step-by-Step
available in a software application. Users frequently say yes when offered
Guide. Find more of Eric functionality, resulting in an application that has too many functions and
Schaffer’s usability tests at controls to make it useful.
https://www.informit.com/
You can use a variation of the functional salience test to determine
articles/article.aspx?p=
170791&seqNum=5 the most relevant page interlinks available to site visitors from any one
page, including contextual links, upsell links, and alternative links. This
test should exclude global and local navigational items, as those navi-
gational elements should be constant across the site and/or categories
where appropriate.
Begin the test by describing the page in question to a user. For the
purpose of this example, let’s use a product page that features running
shoes. You could verbally describe the page as featuring a pair of running
shoes for the more serious runner who typically runs three or more miles
a day, totaling 20-plus miles a week. Many of the runners who use these
running shoes run in 5K and 10K races, and half and full marathons.
Ideally test participants should be representative of an audience
who would find your page useful. Don’t ask someone who has little
or no interest or knowledge in the page topic or product, as you’ll get
skewed data. After you describe your page, give your participants a list
of interlinks they might have available to them. Tell them that they can
choose only three of those options from the page in question. In the case
 HOME PAGE EXPECTANCY TEST 149

of a product page that sells running shoes, the options you would offer
to test participants might look like this:
" Similar running shoes
" Running jackets
" Running shorts
" Running socks
" Running pants
" Running training tips
" How to train for your first marathon
" Find a race near you

Just like software users, chances are your website users will want more
information than is good for them and will want to design a site that has
everything at their fingertips, resulting in a website that is overcrowded
and full of noise—dramatically decreasing the ability of site visitors to
search and find information.
By forcing test participants to select the information that is most
important to them, we can limit the number of links to the most impor-
tant ones. As a result, we will be creating a site that meets users’ mental
models and a site with content that is more easily found.

! HOME PAGE EXPECTANCY TEST

Use this test to make sure you are properly communicating content Tip It’s best to conduct a
that is accessible from the home page. Document the content the user preliminary test with a co-
expects as well as the media type and formatting such as a PDF or a worker to get the kinks out
of your test before testing
video. The closer you can match your content and execution to your with participant(s). Any co-
users’ expectations, the better the experience will be for them. The home worker will do. What you
page expectancy test will need to be done in two steps: are doing is testing your test
to make sure you haven’t
1. Click affordance. Print out your home page and give it to test missed anything in your
participants. Ask them to circle everything they believe to be a preparation. This will ensure
hyperlink on the page. a smoother execution of the
test with your participants.
2. Expected content. Ask the participants what information they
would expect to find or what they would expect to do if they
were to click one of the links they circled.
150 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

Step 1: Click Affordance Test

Note Don Norman was Affordance is the visual indication of how an object functions—in other
the first to apply the term words, what it is for. In an ideal world you would be able to perceive an
affordance to user experi- object’s affordance without having to handle the object but simply by
ence in The Design of
Everyday Things.
looking at it.
For example, there is no need to pick up a pencil to know that it
affords writing, or a hat to know that it affords wearing on your head.
If you have to ask yourself, “What is this for” or “What does this do,”
the affordance of the object is not clear and will require more cognitive
processing to determine its function.
One way affordance applies in the hypertext medium of the World
Wide Web is through the perceived clickability of web page elements.
Page elements that are clickable should look clickable. It’s just as important
that items that are not clickable don’t look clickable as we don’t want
users randomly clicking around the page, getting frustrated, getting
angry, and eventually abandoning.
The importance of click affordance was illustrated by one of our
participants being tested on a website that didn’t underline links. She said
angrily, “Why don’t websites underline links any more? I don’t understand
why they change something just to change it without making it better.”
She said this just before she abandoned the assigned task.
Test for click affordance with paper prototypes, more advanced
graphical prototypes, or even a page screen shot. Whatever you do, don’t
test with a functioning website. Affordance tests aren’t to determine if
a user notices if the arrow turns into a hand when the user moves the
mouse pointer over something that is clickable, but that’s exactly what
you’ll be testing if you use a functioning website. Test to see what users
perceive to be clickable elements on the web page. The only way to
do this is to get users away from the computer so they aren’t tempted
to rollover page elements to see if the arrow changes. In their natural
environment, users determine if something is clickable or not before
they move the mouse, not when they rollover something. Again, it’s the
perception of what is clickable that you are testing.
Begin the click affordance test by giving the test participants the
paper prototype, graphic prototype, or hardcopy screen shot. Ask them
to circle everything on the page they think is clickable. Chances are, your
participants won’t circle every actual link. They might even circle some
page elements that are not links. Some participants will circle big blocks
of content; others will be very specific and will circle only one or two
 HOME PAGE EXPECTANCY TEST 151

words. You might have to gently ask some of them to clarify what they
think is a link and what isn’t a link.
Don’t lead your participants by acting surprised when they don’t circle
something you think is obviously a hyperlink. Also, don’t ask “Don’t you
think the big graphic on the right is a link?” or other types of questions
that will taint your results.
Looking at the National Cancer Institute page (Figure 9.5), what
would you think is clickable?
Figure 9.5 A content page
from the National Cancer
Institute’s website did well
in a click affordance test.
www.cancer.gov/newscenter/
pressreleases/ALTTO

The National Cancer Institute website did well in an affordance test,

but did experience the problems shown in Table 9.1.

Table 9.1 National Cancer Institute’s click affordance test results

POTENTIAL PROBLEM AREAS LINK OR NO LINK

“En español” in the left column Link, but the users didn’t think so
The icons to the left of the Link, but the users didn’t think so
print and email links Tip Anything your par-
Images above Posted: 02/29/2008 Not a link, but the users thought so ticipants don’t circle that
is a hyperlink should be
noted. Make changes to
these elements to make
If users miss a hyperlink, they miss it. Note it silently to yourself or them look more clickable.
have an observer write it down, and move on. If hyperlinked elements
don’t look clickable, they
Let’s take a look at Figure 9.6 on the next page to see how Google
fail the affordance test.
communicates clickability.
152 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

Figure 9.6 Google makes no

mistake in communicating
what’s a link in the SERP. If
it’s a shade of blue and it’s
underlined, it’s a link.

Step 2: Expected Content Test

After your users return your paper prototype, graphic prototype, or
hardcopy screen shot, ask them what they would expect to find or be
able to do after clicking each of the “links” they circled.
What users tell you they expect to find after clicking is what they
will be scanning for on the page that follows their click. If users don’t
find what they are looking for on that page, they will lose the scent of
information.
Users may also circle items that are not links. You’ll want to know
what page elements your participants thought were links and what they
expected from clicking those page elements. This may be an opportu-
nity to make a page element a hyperlink or you may need to rethink the
visual treatment of the page element as it is incorrectly communicating
clickablity.
Some items to keep in mind as you administer the test:
" Did your participants expect to begin transacting with your
site by clicking any of the links? Did they think they were
going to download a whitepaper, play a game, or submit
a form?
" Did your participants expect information that isn’t currently
on the site? Would they be disappointed if they were to click
the actual link and find information other that what they
anticipated would be there?
" What type of technology did your participants expect to
deliver the information or transaction? Did they expect a
pop-up, a new page, a new website, a PDF, a Flash applica-
tion, or Ajax application?
The closer you can match your web links to the expectations of your
participants the better the chance you have of meeting your website
goals and not losing users.
 EIGHT-SECOND USABILITY TEST ON CONTENT PAGES 153

! EIGHT-SECOND USABILITY TEST ON

CONTENT PAGES
The eight-second usability test is adapted from Jared Spool’s five-second
usability test with three notable differences:
" Jared’s test takes five seconds; ours takes eight.
" The second difference is where the testing begins. Jared begins
on the content page, starting his five-second count when the
user first sees the page. We begin our test in the SERP, and
start the eight-second count after the user clicks the SERP
listing and the viewable portion of the content page is loaded.
We’ll explain this in detail in a minute.
" The third and final difference is the goals of the tests. Jared’s test Note You can find out more
is designed to understand users’ initial impression of a content about Jared’s five-second
page. Our test is designed to do that too, but also to understand test at www.uie.com/
if they feel like they are on the right page after they click the articles/five_second_test/

SERP listing, if the information they want is on the resulting

page, and if they can establish a sense of place on this page. We
give the user three more seconds to allow for download time as
well as the time it takes to cognitively process and determine a
sense of place and note keywords on the resulting page.
It’s very important that you don’t tell users you are only going to
give them eight seconds to look at the page. If you indicate that you will
be removing the page after eight seconds, they may try to cram as much
information as possible, skewing your test results.
Here’s how to perform the eight-second usability test:
1. Ask users to use a search engine that you are fairly confident
Tip Eight seconds may
will show your site in the SERPs.
seem like an eternity to you,
2. Give them the query to submit. but it will feel very brief to
your participants. Give them
3. Tell them to look for your company’s website in the SERP lists a full eight seconds. If you
and click it once they find it. are testing on dial-up con-
4. Once the resulting page from your site loads, begin to count, nection or have download
speed considerations, give
silently to yourself, to eight. your users up to 12 seconds.
5. Close the window after you reach eight seconds. Be consistent—every partic-
ipant should be given the
6. Ask the users same amount of time to
" Is this the information you expected to see on this page? evaluate a page. Don’t make
" What page were you on? it obvious to the users that
you are giving them a spe-
" What words do you recall being on the page? Note any
cific amount of time, or you
keywords they mention. Compare the keywords the parti- may distract them and get
cipant mentions with the keywords in your SERP listing. skewed data from the test.
154 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

Minimally what you are looking for is that users feel the SERP listing
brought them to the appropriate page on your site.
Figure 9.7 The search
engine query that brought
users to this page was
surfrider chapters. In eight
seconds, can you tell if this
page is the appropriate
page for the query surfrider
chapters? http://surfrider.
org/chapters.asp

! OBSERVE NAVIGATIONAL WEB SEARCHES

Watch users perform a navigational search on a commercial search engine.
Ask participants to perform a navigational search for your company in
their favorite commercial search engine. Be sure there is a good chance
of the users finding your site in the intended commercial search engine.
This shouldn’t be an exercise in futility.
This test may prove especially helpful to companies and organizations
whose names use common and/or competitive words and if their com-
pany is searchable using abbreviations or multiple name variations.
When you ask the user to perform the search:
" Don’t ask them to perform a “navigational search.” They prob-
ably won’t know what that is.
" Don’t tell them the name of the company you want them to
search for. If you do, they will type exactly what you just asked
them to search for.
" Don’t tell them the domain of they site you want them to find.

Instead say something like, “Find our company on the Internet using
your favorite search engine.” A generic statement like this is not leading
 REVIEW SITE SEARCH DATA 155

and leaves it to the users to decide which commercial search engine to

use and what words to use to search for your company.
Note how the participants search for your company’s website:
" Do they use your proper company name in their query?
" Do they use an abbreviation or an acronym in their query?
" Does your site come back in the SERP across a variety of com-
pany name spellings, abbreviations, and so on?
" If they use abbreviations, does your SERP listing use those
abbreviations too?
How is your site listed in the SERP? Remember, the users are going
to look for their validated query in the SERP.
Try searching for high point or true north in your favorite search engine.
Without a qualifier like furniture, high point could mean one of hundreds
of navigational searches. Similar is true of true north.
The search engines may not be able to anticipate all name variations Note The longer your com-
your users might use in their queries. Consider buying your company’s pany name, the more likely
name variations via pay-per-click advertising, especially if your company that users will edit, shorten,
or abbreviate it in some way
name uses competitive or common words in its name that might make when they perform a navi-
it too difficult to satisfy all user queries with a natural search. gational search.

! REVIEW SITE SEARCH DATA

Site search queries frequently contain the last words a user searched for
before abandoning your website.
By viewing queries in your site search logs, you’ll gain an insight
into how users are thinking about your content and the keywords users
are searching for and not finding on your site.
Your query data will fall into three categories:
" Data that is completely disconnected from the content of your
site. Be careful not to misinterpret junk queries as valid queries.
" Valid user queries that your site doesn’t satisfy.
" Valid user queries that your site does satisfy.

It’s the last two items that we want to take a closer look at. The second
category (valid user queries that your site doesn’t satisfy) requires you to
make some decisions. Just because users are looking for this information
doesn’t mean you should provide it, if it doesn’t complement your busi-
ness goals. If you decide you want to act on the user site search query,
you’ll need to decide where to build out the content so your users will
be able to find it.
156 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

The third category in the list (valid user queries that your site does
satisfy) suggests that your content is not easily found by primary means
of navigation even though it does exist on your website. Users might not
be able to find the content because the content is buried too deep in the
architecture, or because links to the content are not catching the users’
eye, or because navigation labels aren’t easy to interpret.
You can improve the findability of this content by conducting usability
tests to detect where they lose the scent of information. First, choose a
query from your site search data (remember, this data includes the site
search queries your site users composed after losing the sent of information).
Second, ask your test participants to find the content that best represents
the site search query. Note the paths they take to find the content. How
do those paths compare with the paths you expected them to take?
Be aware of where your users lose the scent of information on your
website. Indications of users losing the scent of information include hesi-
tation in clicking hyperlinks, not knowing what content to expect from
clicking hyperlinked copy and graphics, and excessive clicking that results
in frustration. Ask your participants to think out loud as they search for
content. Participants verbalizing their thoughts and actions will clarify
what they are they thinking as they attempt to locate the content.
The report in Figure 9.8 shows the number of searches and num-
ber of relevant results from a WordPress blog. Some queries are more
relevant than others. Hangers, for example, isn’t relevant to the content
on the site whatsoever.
Figure 9.8 Changes to the
site should not be made to
accommodate such irrele-
vant queries such as hangers.
 BRAND PERCEPTION TEST 157

! BRAND PERCEPTION TEST

Branding covers a wide range of attributes including graphic elements like
logos and corporate fonts. But branding is much more than graphics—
it’s the emotional connection someone feels for a product or company.
Website usability, be it good or bad, affects how users think and feel
about your brand.
What first comes to mind when you hear these brand names:
" Apple
" Starbucks
" Google
" McDonald’s
" Amazon

You probably have some strong feelings about these brands if you
fall into the target market for these companies.
Let’s take Apple, for example. Apple’s TV campaigns tout its superiority
over PC. Apple users probably find the ads smart and funny while a PC
user might find the ads arrogant and insulting. Apple is positioning itself
as a superior product in a way that resonates with its target market.
The idea behind brand positioning is to get you to think about a
product in a way the manufacture wants you to, in order to make an
emotional connection with you before you walk into a store, go online,
take the car for a test drive, or whatever action you take with a product
or service.
Every touch point either enhances or degrades the product positioning
and the emotional connection you have with a brand. This also includes
websites. Unlike many other touch points, however, you can measure
how websites affect branding pretty easily and relatively cheaply by simply
asking users about their experience on a website after usability testing.
First, you’ll need to get a baseline on the current emotional connection
your participants have with your brand and then see if their interaction
with the website enhanced their emotional connection, degraded that
emotional connection, or didn’t move the needle either way. Tip The ease of finding
information and ability to
This can be done formally or informally. If you are trying to get a
complete a task will affect
general feel of how your website is affecting your brand, you can do this brand perception. The
verbally with participants. If you’re looking to use the results as part of stronger the scent of infor-
a pitch to get more attention put on your search and web usability initia- mation, the stronger your
users’ emotional connection
tives, you should do this more formally and roll up the data into a report will be with your website
with graphs that catch the attention of your management or client. and brand.
158 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

What you’ll need

" A customer or prospect familiar with your brand.
" Tasks for testing the usability of your website.
" A survey designed to measure branding. We’ll look at some
sample survey questions in a minute.
What you do
" Before usability testing, give the participants the branding
evaluation that asks them how they perceive your brand.
" Perform your choice of usability test(s) with your participants.
" After the usability testing, give the participants the same
branding evaluation that asks them how they perceive your
brand after taking your usability test.
" Evaluate how/if using your website changes your users’ per-
ception of your brand.
Ideally, you want participants to rate your company’s brand higher
on the post-usability branding evaluation. An increased rating would
show that your website is enhancing your brand. Minimally, you want
participants to rate your company’s brand the same on the pre- and post-
usability branding evaluation. If this is the case, take comfort in knowing
that your website isn’t hurting your brand. Unfortunately, your website
isn’t helping your branding initiatives either. If you receive lower ratings
on your post-usability branding evaluation, you have some work to do
on your website.
Be sure the questions you ask are written in a way that the partici-
pants know they are evaluating how the website affects their perception
of the brand.
For example:
Rate how you feel about COMPANY A before and after interacting with
COMPANY A’s website.
" Before interacting with COMPANY A’s website, I would rate
my feelings as
1) Very Weak 2) Weak 3) Neutral 4) Strong 5) Very Strong
" After interacting with COMPANY A’s website, I would rate my
feelings as
1) Very Weak 2) Weak 3) Neutral 4) Strong 5) Very Strong
Some other questions you might want to ask in your branding evalu-
ation, using a similar format as the previous prompts, include
" Rate COMPANY A’s trustworthiness before and after interact-
ing with COMPANY A’s website.
 CHECK YOUR ERROR MESSAGES 159

" Rate COMPANY A’s reliability before and after interacting

with COMPANY A’s website.
" Rate your likelihood of recommending COMPANY A to a
friend or family member before and after interacting with
COMPANY A’s website.
You can add an open-ended question at the end of the second survey
to help the users summarize their experience after using your website and
get more insight into how they feel about your website, such as, “How
would you describe your experience on COMPANY A’s website?”

! CHECK YOUR ERROR MESSAGES

One bad error message may be all that stands between your user and
a conversion on your website. Typically error messages come in three
flavors:
" Poorly written cryptic techno-language that nobody
understands
" Poorly written error messages that were meant to be helpful
but aren’t
" Well-written error messages that properly explain the error
and help users get back on track
Ideally, all of your error messages should fall into that last category.
You might be able to get a document that lists the error messages
and the associated page and field(s) that trigger them from your IT group.
Chances are, you’ll need to go through your site and perform a post-
launch QA, creating errors and documenting as you go along.
Your error messages should be so clear and concise that your users can
match them up to the field on your web page without having to trigger
them by submitting the actual form. One way to confirm if this is true
is to bring your error messages in front of users on individual flash cards
and ask them to match up the error with the field on the page that would
trigger it if the form were submitted. If your users can correctly tell you
which field triggers the error message, it’s a well-written error message.
For example, the following is part of an error message that the Yahoo
Directory delivered because the credit card type didn’t match the 16-digit
credit card number: “Invalid Payment Instrument Data”
Have you ever been asked when checking out at a store, “Is that a
debit card or a payment instrument?” Didn’t think so.
Here’s a proposed error message: “The credit card type you selected
doesn’t match the credit card number you entered.”
160 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

While you’re at it, check your formatting.

Formatting the error message so that the message is physically near
with the field causing the error is another form of scent of information.
Proper formatting and writing of the error message will greatly reduce
users’ need to scan and search the page.

! FREE EXPLORATION TEST

Have you ever watched someone use your website? It seems so obvious,
but many of us have never seen someone actually use the websites we help
create. Watching a user freely explore your website will open your eyes
to stumbling blocks that you may have never considered otherwise.

Have you ever watched someone use your website?

It seems so obvious, but many of us have never seen
someone actually using our website.
Here’s how you conduct a free exploration test:
" Find a participant who matches your site’s user profile. Ask if
they would be willing to show you how they would use the
site to purchase, find, or learn about X, Y, Z.
" Tell them you want to see how they would use the site. Ask
them to think out loud as they click around.
" Ask them questions when something interesting happens.

Sounds simple enough, right? The idea is to not direct users to do

any specific thing. You can direct them later with other tests. For now,
let the users self-direct. You’re looking to see how they would naturally
use the site—as naturally as they would with someone looking over their
shoulder asking them questions, that is.
As your participants are going through your site, ask them why they
are doing certain things. If they aren’t using portions of the navigation,
ask them why. You may find out that they didn’t even notice the naviga-
tion or that the navigation didn’t look like navigation to them.
Other items to take note of or ask include
" If they exhibit pogo-sticking behavior (going back forth
between pages using one page as an anchor).
" If they quickly scroll down a page scanning for something, ask
them what they are looking for.
" Ask them if they know what page they are on after they click
through a few pages.
 DECONSTRUCTING COMMON MYTHS AND MISCONCEPTIONS 161

" If they use site search, site map, breadcrumbs, or other forms Tip As with any usability
test, be clear to your partici-
of secondary navigation, ask them if they normally use those pants that they are not the
types of navigation. If they do, ask why. If they don’t, ask them ones being tested—it’s the
why they used it on your site. website that is being tested.
Let them know that they are
Showing interest and being curious about your users, without being providing valuable feedback
overbearing, will yield valuable feedback that you can use to improve that will help make the web-
your website’s search usability. site better for everyone.

! DECONSTRUCTING COMMON MYTHS AND

MISCONCEPTIONS
Some managers and clients look at search and usability professionals
as a means to a quick and permanent solution for a website problem,
assuming that one round of SEO changes or one usability test will fix it.
They don’t understand that what search and usability professionals do is
a process that involves constant change based on user feedback.
Following are some common myths dispelled to help you set expec-
tations with your managers and clients and get your search and web
usability efforts moving in the right direction.

Myth #1: There Is a Magic Bullet

There is no magic bullet or secret sauce that will solve all your search
and web usability problems.
The best any web professional can do to address search and web
usability problems is
" Continue to build your toolkit with techniques to better diag-
nose search and web usability problems appropriately. Is it a
navigational, information, or transactional query? Do we need
to perform usability testing or not?
" Know when to tap into the individuals who are most appropri-
ate to address the search usability issue of the day. For exam-
ple, call an information architect if users are getting lost in the
website. Call a writer or a graphic designer if a call to action
isn’t getting users’ attention.
" Test your website with your users early and often.

Addressing search and web usability issues can sometimes be a

slow, long-term process. Just know that the discoveries you find and the
seemingly small changes you make can lead to big gains in your website’s
search usability.
162 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

Myth #2: There’s No Such Thing as

Perceived vs. Actual Website Speed
A friend of ours went skydiving for the first time last year. As his instructor
was giving the details on the jump, he mentioned that the jumpers were
going to free-fall for about 45 seconds before the parachute opened.

The instructor replied, “If you’re enjoying the free-fall,

45 seconds is not long at all. If you end up hating the
free-fall, 45 seconds feels like forever.”
Having no prior experience with jumping out of an airplane, our
friend asked, “Is 45 seconds a long time to free-fall?”
The instructor replied, “If you’re enjoying the free-fall, 45 seconds
is not long at all. If you end up hating the free-fall, 45 seconds feels like
forever.”
Similar is true with perceived website speed and download time. If
users are enjoying their time on your site, finding information quickly,
accomplishing tasks, they will think your site is fast. If they can’t find
what they are looking for, or they have trouble completing a task, they
will think your site is slow.
Some people find this difficult to believe. Instead of working on
perceived download time they work on actual download time. We’re
not saying to abandon efforts to decrease the actual download time of
your site. We absolutely encourage it. Clean up your code, optimize
your images, and upgrade your servers. The quicker you can get content
to your users the better. But it doesn’t stop there. You need to work on
perceived download time, too.
Here are some tips to improve perceived download time:
" Help users follow the scent of information from page to page.
If your users can find information more easily on your site
than your competitors’, even if it takes longer, they will feel
like they are getting around quickly and will think that you
offer a faster site.
" If users can get through a process on your site, such as filling
out a form or adding something to cart, with fewer complica-
tions and ambiguities, they will have to do less cognitive pro-
cessing and will think your site is faster than your competitors’.
" Lower user frustration levels and abandonment rates by
improving content on help pages and FAQ pages, improving
 DECONSTRUCTING COMMON MYTHS AND MISCONCEPTIONS 163

site search results, creating annotated site maps, and providing

ambiguity-free error messages.
You still may not believe in perceived download time. Don’t take
our word for it. Try a little experiment on yourself. Try staring at the
wall for one minute. Then watch a one-minute video of your choice on
YouTube. Which one felt faster?

Myth #3: The Competition Does It Right

Don’t assume that because your competition does something on their
website that it’s a good idea. Page one SERP listings give the impression
that a website is good simply because it has achieved page one results.
Often this is not the case.

Blindly keeping up with the Jones’ website is no

guarantee that it will make your website better.
One company we worked with was determined to have six links in
their primary navigation—no more, no less. They came to this conclusion
by looking at other industry websites found in the top SERP listings, some
of which featured six links in their primary navigation. To them, six was
the magic number. They rationalized that by using six primary navigation
links they would make their website simple and easy to use.
User testing of the competition’s website showed that the six links in
the primary navigation buried content, requiring unnecessary clicking,
making the desired content difficult or impossible to find. The client’s
assumptions, based on their own opinions and not on user testing,
proved incorrect.
Blindly keeping up with the Jones’ website is no guarantee that it
will make your website better. Don’t assume the way a technique was
executed on someone else’s website will work for your site. Validate your
assumptions with user testing.

Myth #4: You Can Design a “Good”

Site Without Feedback from Users
This is a close cousin to “The competition does it right.” The difference
is instead of assuming that the competition knows your users better than
you do, you’re assuming that you know your users enough to not talk
to them and usability test.
164 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

The lack of user-focused and usage-focused design makes people do

funny things. People start to assume incorrectly about their users. Those
incorrect assumptions equate to wasted money, lost business opportunity,
and talented people quitting to go work for someone else.

It’s really not that difficult to put together a

mediocre site.
The contrary happens when you bring your site in front of users. You
will see where you lose users, when they become frustrated, and where
they abandon your website. The experience will change your perception
of how people get around and use your site.
It’s really not that difficult to put together a mediocre site. There are
plenty of software applications and tutorials online that will help you
technically put together a website. This explains why there are so many
mediocre websites. You need to interact with people similar to your users
if you want to create a good website.

Myth #5: People Will Use Your Site

Since It’s the Best of Its Kind
Some people think that if their site is the best in its category, there is no
reason to usability test their website. The client’s justification: “If you
were a user, which site would you use to buy X?”
Some site owners have a point. Their website is much better than
some of the terrible sites their competition serves up to users. Still, it’s
risky to assume that your product is so desired that people will do any-
thing, including muddling through a clunky interface, to get it.
Case in point: A friend of ours canceled her trip because the websites
she was using to book her airfare were too hard to use and she decided
not to go.
It wasn’t that she didn’t want to go visit her friend; it was that she had
other options and she decided, with the help of getting frustrated with a
website or two, that she preferred one of the other options instead. Like
a road trip that didn’t involve finishing an online checkout.
Next time you think, “If I were a user, I would buy from my site
because it’s the best site out there, albeit still a bad site” think about
the hundreds of options users have that might be more desirable than
frustratingly muddling through your website not finding the information
they want or being unable to finish a task.
 DECONSTRUCTING COMMON MYTHS AND MISCONCEPTIONS 165

Myth #6: You Lose Users After Three Clicks

Most websites are built with a very poor scent of information. As a result,
users don’t find what they are looking for, get lost, and abandon after
three clicks. It doesn’t have to be this way.
If you effectively string the scent of information throughout your
website, users will keep clicking until they’ve decided they have gotten
what they need. They will decide when to leave because they’ve found
what they were looking for. If their experience was a good one, there’s
a chance they will be back.

We are led to believe that users abandon after three

clicks. It doesn’t have to be this way.
Not only will your website see an increased number of returning
visitors, but your web analytics will also show you
" Lower bounce rates
" Increased time on site
" Increased page views

WEB ANALYTIC TERMS

Bounce rates—A bounce happens when a user leaves your website or a page on your site to go to
another website. Bounce rate is the percentage of users who leave, or bounce off, your website.
Time on site—As you might have guessed, time on site is the amount of time a user spends on
your site. It is commonly thought of in terms of average time on site. Time on site can result
from either of two reasons: time spent in puzzlement, or time spent in enjoyment. Only usability
testing will show you the true reason.
Page views—Page views means the number of pages a user views during her or his time on your
site. Page views are commonly thought of in terms of average page views per visitor. Page views
are not “hits.” Many people still think a hit is a page view. A hit is an antiquated web metric and
is not the same as a page view.

What’s even better: These happy users will tell others about your site
via links pointing to your site from websites and blogs, social networking
sites like Facebook, and social bookmarking sites like del.icio.us.
To sum it up, a good scent of information equals happy users, which
results in more quality inbound links and, as a result, better search engine
rankings.
166 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

Myth #7: The User Is the Problem

We can’t tell you how many times we’ve heard from website builders
regarding how users get around their site and that they are doing it
wrong. Essentially, what they are saying is that the user, not the website,
is the problem.

Contrary to what some people think, the user isn’t the

problem. The problem is the website, and it should be
changed.
That sentiment is often expressed in one of the following fashions by
website owners, developers, designers, and others working on the site:
“They [users] should do it this way.”
“I would do it this way.”
“This is the best way to do it.”
Contrary to what some people think, the user isn’t the problem. The
problem is the website, and it should be changed. You can be apathetic
and ignore your users until they go away, or you can be empathetic and
help your users, and they will eventually make your site a success.

Myth #8: The User Is Always Right

Just because the user isn’t the problem doesn’t mean the user is always
right.

A tomato is never going to be a vegetable no matter how

many users wanted to find tomatoes under vegetables.
Case in point: In a usability test, when asked to locate the tomato
page of a fruits and vegetables website, 60 percent of users initially went
to the vegetable section. About 33 percent of those who did not find the
tomato page under vegetables looked for tomatoes under fruits.
A tomato is never going to be a vegetable no matter how many users
wanted to find tomatoes under vegetables.
You may find that users look for information on your site in an area
that obviously shouldn’t house the information, like a tomato page
under the vegetable category. One way to remedy this type of situation
is to list a blurb on a page about tomatoes under vegetables. Something
to the effect of:
Tomato You can find more information on tomatoes under fruits.
(The above link takes users to the tomato page under fruits.)
 DECONSTRUCTING COMMON MYTHS AND MISCONCEPTIONS 167

By taking into account the high percentage of users who think a

tomato is a vegetable, we are educating and helping users find their content
without changing the definition of a tomato, without creating duplicate
content, and without losing users who know that a tomato is a fruit.

Myth #9: Usability Tests Are 100% Accurate

Usability tests arguably produce the best type of feedback you can get.
They are a one-to-one, in-person, direct method of research that gives
the administrator a chance to ask follow-up questions to clear up any
questions he or she might have from the test session.
So why are usability tests not 100 percent accurate?
Test subjects, just like students in school, know they are being tested.
As a result, they tend to try a little harder than they would in their natural
environment at home or work.

No natural environment, be it work or home, is 100

percent distraction free of kids, spouses, colleagues,
phones, instant messaging, texting, or pets.
Most people will abandon the task much sooner in their natural envi- Tip Increase your usability
ronment than your test subjects in a testing environment. Additionally, the testing accuracy by testing
testing environment is usually designed to be distraction free. No natural the appropriate number of
test participants. You’ll
environment, be it work or home, is 100 percent distraction free of kids, know you are finished test-
spouses, colleagues, phones, instant messaging, texting, or pets. ing when you start hearing
test participants say the
same thing. For most tests
Myth #10: Users Will Figure Out 5–10 users per group will be
What Is a Link by Clicking It sufficient.

Just because you aren’t putting white links on a white background doesn’t
mean you aren’t hiding links from users. By not making clickable items
look obviously clickable (affordance) you are obscuring their function
and the scent of information along with it.

Don’t force users to read your mind. If something

is clickable, make it look clickable. If something is
not clickable, don’t make it look clickable.
Users decide what is clickable before they move their mouse, not
when their mouse happens to rollover a page element and the arrow
becomes a hand.
168 CHAPTER 9: HOW TO IMPROVE YOUR WEBSITE’S SEARCH USABILITY

Don’t force users to read your mind. If something is clickable, make it

look clickable. If something is not clickable, don’t make it look clickable.
This philosophy of making things represent their functionality should
be carried across all page elements.
The more you make users think about how your site functions, the
quicker they will tire and you will lose them.

Myth #11: Usability Guidelines Will

Make Your Site Successful
Tip The Design of Sites by Not every usability guideline is applicable or will work for your site as
Douglas K. van Duyne, each website and task flow is unique. You must usability test to see what
James A. Landay, and Jason works for your site and your task flow. Only then will you understand
I. Hong is a good place to
review web usability
why users respond the way they do to your website.
guidelines. A colleague of ours did all the “right things” to her shopping cart
with the intention of improving her conversion rate. Unfortunately,
she didn’t usability test her changes. Her first round of feedback on the
changes came from live users who abandoned her shopping cart at a rate
of 2.5 times that of her original design.
Without fully understanding the reasons why all the “right things”
didn’t work she reverted back to the previous design.
Following web usability guidelines is a good place to start, but they
are only one tool in the search usability toolbox. Nothing beats testing
your own site.

Myth #12: The Web Is Yours

David Weinberger, author, philosopher, and marketer, sums up companies’
view of the web like this: “… [T]he big mistake businesses make isn’t
the economic or technical. It’s the psychological or (dare I say?) spiritual.
They’re not getting the biggest ‘it’ there is to get: The web isn’t theirs.
The web is ours.” (From the essay “Explaining the Passion that Powers
the Web,” published in Best Practice: Ideas and Insights from the World’s
Foremost Business Thinkers.)
What Weinberger is saying is that while it may be your website, it’s
the users’ experience you need to design for. If not, your website and
possibly your business will fail.
 CONCLUSION 169

! CONCLUSION
Here are the key points of this chapter:
" Using the reverse card sort test to add well-organized structure
to an existing site will increase your search visibility and make
it easier for users to find the content they are looking for.
" To meet users’ expectations and reduce abandons, ask users
what content or functionality they would expect from clicking
every link on your home page.
" Watching users perform a navigational search on a commercial
search engine will show you how/if users locate your site in
the SERPs.
" Users turning to site search results can show symptoms of
your website’s lack of scent of information.
" Bad error messages can make or break conversions. Check
to see if yours are written in a language your users will
understand.
" There is no magic bullet. There is no secret sauce. It’s a lot of
hard work to improve your website’s search and web usability.
" Perceived download time is just as important, and maybe more
so, than actual download time.
" Just because the competition does something on their site
doesn’t mean you should do it on your site.
" If you want to design a good website that meets users’ goals
and business objectives, you must talk to your users.
" Keep the scent of information strong and people will follow it
further than three clicks deep.
" Users typically aren’t the problem when they can’t find infor-
mation; more likely, your site needs to be changed.
" Conduct usability testing if you want to know how easy it is
for users to find information on your website.
" If something is clickable, make it look clickable. If it isn’t click-
able, don’t make it look clickable.
" Blindly following usability guidelines won’t make your site suc-
cessful. Combining usability guideline with user testing will.
" It may be your website, but it’s your users’ experience. Treat
your users with respect and they will treat you well.
This page intentionally left blank
 INDEX 171

! INDEX

Symbol Baluja and Kumvar, 34

“best of its kind” sites, myth about, 164
" (quotation) marks, using with keywords in, 75
bold text, using, 110–111
bounce rates
A calculating, 77
defined, 165
A/B split testing
overview of, 103–104
comparing prior months of data with, 122
brand degradation, evidence of, 117–118
defined, 67
brand equity, defined, 118
above the fold
brand perception test, conducting, 157–159
explained, 53
breadcrumb links
keeping file tabs, 78
locations of, 124
transactional information, 63
using in orienting behavior, 83
actions versus words, 51
Broder, Andrei, 29
Add to cart
browsing
impact on search usability, 92–93
defined, 3
including as call to action, 72, 74
behavior, 4
ads, selling with search usability, 94–97
business goals versus user expectations, 21
advertising, cost of, 94
business models
affordance, defined, 150
developing assumptions for, 90
AHT (average handle time), use by call
centers, 98 formula for, 90
anchor link, defined, 113 selling ads with search usability, 94–97
annotated lists, using with category pages, 48 selling products and services, 90–94
annual revenue supporting existing products and services,
97–101
calculating, 92, 95
“button of doom,” Back button as, 117
increase for selling ads, 95–97
buttons, association with lobby, 8–10
Apple’s Safari page, call to action on, 68
attracting users. See calls to action
average page views, increasing, 96 C
call centers, use of AHT by, 98
B calls to action
on Apple’s Safari page, 68
B2B website, annotated links page on, 53
defined, 72
Back button, reducing use of, 117–118
172 INDEX

calls to action (continued) clicks, myth about, 165

drawing attention to, 73 clickstream, defined, 86
evaluating effectiveness of, 75 colored text, using, 110–111
example of, 65 company roles
formatting as text links, 111 graphic designers, 109–112
levels of, 72–75 information architects, 136–139
making stand out, 72 IT (information technology), 132–136
prioritizing, 73 management, 125–130
title-tag content as, 23 market researchers, 130–132
for transactional queries, 68–69 marketing communications, 117–119
cancer. See National Cancer Institute website usability professionals, 119–125
canonical domain names, explained, 37, 39 web copywriters, 112–117
captions, applying scent of information to, 114 competition does it right myth, 163
Card, Stuart, 8 content pages, performing eight-second usability
cards, using in reverse card sort test, 143–147 test on, 153–154
Case Studies copy
The Cost of Assuming User Expectations, 66 formatting for search usability, 116–117
Information Architecture, 138 reflecting users’ language in, 115–116
International Travel Site, 128 retaining scent of information for, 112–113
Official Fan Site, 69 copy areas, identifying for information scent,
category pages 113–115
best practice for, 54 copywriters, advice for, 116
channel pages as, 39 The Cost of Assuming User Expectations case
study, 66
as information pages, 48
CPM (cost-per-thousand) model, applying to ads,
for navigational intent, 39–42
94–95
and transactional intent, 64
crisis information, communicating, 119
using annotated lists with, 48
cross-selling, defined, 97
channel pages
cues, using in orienting behavior, 81–85
defined, 39
customer service expenses, lowering, 99–100
example of, 40
customers
for navigational intent, 39–42
losses per year, 97
choices, influences on, 13
supporting, 97–98
CIA career page elements, 114–115
click affordance test, conducting, 150–151
click popularity, use of, 35 D
clickability, indicating in Google, 151–152 department page, defined, 39
clickable text, indicating, 110–111 download time, improving perception of,
clicking, basing on scent of information, 8 162–163
 INDEX 173

E for usability, 76
for you are here cues, 84–85
e-commerce sites. See also transactional intent
Flash sites, appearance as misleading
product pages on, 63
links, 80
searching, 48
Flash videos, problems with, 65
effectiveness, measuring for websites, 6
focus groups
efficiency, measuring for websites, 6
considering for transactional queries, 68
eight-second usability test, performing on
versus usability tests, 131
content pages, 153–154. See also five-second
test formulas
elevators, scent of information on, 8–10 daily profit of e-commerce site, 90
emotional connection, determining for brands, monthly cost of calls, 98
157 monthly intranet use, 101–102
employee intranets, search usability for, 101–102 monthly revenue, 91–92, 94–95
entertainment, transactions related to, 64–65 free exploration test, conducting, 160–161
entry pages, effectiveness of, 77 Frequently Asked Questions (FAQs), keyword
error messages phrases in, 52
checking, 159–160 functional salience test, conducting, 148–149
formatting, 160
writing, 133 G
error prevention and recovery, measuring for
websites, 6 Gestalt Principles of Visual Organization, 10
expectancy tests, performing, 75–76 Google, indicating clickability in, 151–152
expected content test, conducting, 152 Google’s search results, finding site
listings in, 75
exploration tests, results of, 75
graphic designers
eye-tracking studies, results of, 75
decisions made by, 110
techniques used by, 109
F graphic images
facts (quick), finding, 49–51 calls to action as, 74
FAQs (Frequently Asked Questions), keyword changing text to, 111
phrases in, 52 resource for optimization of, 52
features, best practices for, 136 graphics, designing, 111
file extensions, using in transactional queries, 67
file tabs, keeping visible, 78
H
files, placing keywords in, 28
findability experts, goals of, 8 home page expectancy test, conducting, 149–152
five-second test. See also eight-second usability home pages
test; usability testing for navigational intent, 35–37
for orienting behavior, 82 using navigational queries with, 35–37
174 INDEX

how-to keyword phrases, presenting, 52 maintaining for copy, 112–113

HTML headlines, writing with keywords, 79 maintaining on long pages, 79
HTML title tags misleading, 80
format of, 23 on National Library of Medicine’s home
including keywords in, 23–24 page, 11–13
purposes of, 23–24 providing with URLs, 27
using in orienting behavior, 82 on shopping malls, 10–11
hyperlinks validating, 22
clicking, 114 visible signs of, 22
example of, 124 website resource for, 23
figuring out by clicking, 167–168 informational goals, determining, 51
obtaining high quality of, 5 informational intent
underlining, 111 indicators, 51–53
hypertext, applying affordance to, 150 lists, 53–55
overview of, 48–49
questions and answers, 51–53
I quick facts, 49–51
IA (information architecture) informational queries
importance of, 137–138 common use of, 47
versus QA (quality assurance), 139 explained, 29–30, 47
iceberg effect, avoiding, 77–79 important elements of, 57
images. See graphic images indicators, 57
information architects, responsibilities of, maximizing, 120–121
136–137 optimizing for, 58
information pages search engine results page for, 50
category pages as, 48 and search listings, 55–57
types of, 120 versus transactional queries, 122–124
information scent. See also keywords using, 30
above the fold, 53 informational searches, displaying listings for, 52
applying to pictures and captions, 114 informational versus transactional sites, 56
beginning, 22 International Travel Site case study, 128
on elevator and lobby, 8–10 interviews, one-on-one field, 147–148
evaluating, 76–77 intranets, search usability for, 101–102
for file tabs, 78 IT (information technology)
identifying copy areas for, 113–115 avoiding unnecessary features, 136
including in search listings, 26 losing scent of information, 133–136
increasing, 19 iterative design model, use of, 132–133
losing, 133–136, 156
 INDEX 175

J L
jobs language of users
graphic designers, 109–112 reflecting in copy, 115–116
information architects, 136–139 using, 128
IT (information technology), 132–136 learnability, measuring for websites, 6
management, 125–130 link development, defined, 5
market researchers, 130–132 link popularity, defined, 5
marketing communications, 117–119 linking pages, importance of, 120
usability professionals, 119–125 links
web copywriters, 112–117 avoiding misleading types of, 79–80
jump link, defined, 113 clicking, 114
example of, 124
figuring out by clicking, 167–168
K obtaining high quality of, 5
keyword phrases underlining, 111
finding search listings for, 75 listing descriptions, generating, 25
including in informational queries, 57 listings
research by SEM professionals, 38 displaying for informational searches, 52
keyword research, importance of, 125 examples of, 56
keyword search, resource for, 38 using meta-tag descriptions with, 57
keywords. See also scent of information lists, generating, 53–55
avoiding hiding, 77–78 “loaded labor rate,” obtaining, 101
calling attention to, 73 lobby, button associated with, 8–10
including in on-the-page content area, 24 location, determining in websites, 81
including in search listings, 28 login pages, example of, 37–39
including in title tags, 23–24 logos, using in orienting behavior, 82–83
including in URLs, 28
phrasing as questions, 51–52
showing navigational intent of, 45 M
for transactional intent, 62–63 magic bullet, myth about, 161
using with annotated category pages, 49 mall directory, scent of information on, 10
using with category pages, 54 malls, similarity to websites, 11
using with HTML headlines, 79 management, and absence of user feedback,
Kumvar and Baluja, 34 126–127
management decisions, impact on web usability,
129–130
176 INDEX

market researchers, perceptions of search N

usability, 130–132
National Cancer Institute website. See also
marketing communications
websites
avoiding bad search usability, 117
applying reverse card sort test to, 146
improving users’ success rates, 118–119
canonical domain name, 36–37
observing user behaviors, 117–118
category page on, 49, 121, 123
marketing goals, being conservative with, 103
clickable parts of, 151
memorability, measuring for websites, 6
determining location in, 84–5
meta-tag descriptions
home page, 35, 37
keeping short, 45
links to, 5, 121
length for informational queries, 58
meta-tag description for, 45
number of characters in, 24
Pancreatic Cancer page, 26, 121
source code for, 24–25
National Library of Medicine, home pages
using with listings, 57 for, 11–13
monthly revenue, calculating, 91–92, 94–95 navigation buttons, using in orienting behavior, 83
months of data, comparing, 122 navigation pages, website resource for, 118
Morville, Peter, 8 navigational intent
motion and sound, grabbing attention with, category and channel pages, 39–42
111–112
home pages, 35–37
multivariate testing
indicating, 34–36
comparing prior months of data
login pages, 37–39
with, 122
navigational queries
defined, 67
category and channel pages, 39–42
myths
common use of, 33
“best-of-its-kind” sites, 164
explained, 29–30
competition does it right, 163
features of, 33
figuring out links, 167–168
formulation of, 34
magic bullet, 161
goal of, 33
perceived versus actual website speed,
162–163 home pages, 35–37
three-click “rule,” 165 importance of, 33
usability guidelines, 168 importance of URLs to, 45
usability tests are 100% accurate, 167 indicators, 45
user feedback, 163–164 login pages, 37–39
user is always right, 166–167 optimizing for, 45–46
user is the problem, 166 and search listings, 42–44
web is yours, 168 URLs in, 43
use on mobile phones and PDAs, 34
using, 30
 INDEX 177

navigational web searches, observing, 154–155 page views

negative search behavior defined, 94, 165
pogo-sticking as, 86–87 increasing, 96
querying as, 20 increasing for ads, 95
Nielsen, Jakob, 72 increasing with pogo-sticking, 97
Norman, Dan, 150 versus time-on-site metrics, 104–105
nouns, using in transactional queries, 67 pages
duplicates of, 138
navigating to, 39–40
O Palo Alto Research Center, 8
Official Fan Site case study, 69 participants in tests. See also tests
one-on-one field interviews, conducting, for brand perception tests, 158
147–148 choosing, 143
online ads, selling with search usability, 94–97 for free exploration tests, 160–161
on-the-page content for functional salience test, 148
defined, 24 qualities of, 142
including keywords in, 24 for reverse card sort test, 143
optimizing search engines. See SEO (search for usability testing, 167
engine optimization) paths, placing keywords in, 28
organizational roles perceived versus actual website speed, 162
graphic designers, 109–112 performance test, defined, 75. See also tests
information architects, 136–139 personalized searches, using click in, 35
IT (information technology), 132–136 perspective, importance of, 108
management, 125–130 phone calls (incoming), lowering, 99
market researchers, 130–132 phone support, reducing, 99
marketing communications, 117–119 pictures, applying scent of information to, 114
usability professionals, 119–125 Pirolli, Peter, 8
web copywriters, 112–117 plural forms of words, including in informational
orientation, defined, 8 queries, 57–58
orienting behavior, fostering, 81–85 pogo-sticking browsing behavior
explained, 86–87
increasing page views with, 97
P
reducing, 118
page headings, using in orienting poor user experience, example of, 13. See also users
behavior, 83
positions
page view to conversion ratio, example
graphic designers, 109–112
of, 91–92
information architects, 136–139
178 INDEX

positions (continued) querying research, resource for, 34

IT (information technology), 132–136 querying versus searching, 19
management, 125–130 question words, including in informational
market researchers, 130–132 queries, 57
marketing communications, 117–119 questions, phrasing keywords as, 51–52
usability professionals, 119–125 quick facts, finding, 49–51
web copywriters, 112–117 quotation ("") marks, using with keywords in, 75
primary call to action, creating, 74
Principle of Proximity, 10
R
product pages
on e-commerce sites, 63 resources. See also website resources
linking, 54 graphic image optimization, 52
product photos, enlarging in call to action, 74 “How Little Do Users Read?,” 72
product tour, taking, 74 keyword search, 38
products scent of information, 8
cross-selling, 97 site searches, 136
viewing, 74–75 URL length, 43
products and services video optimization, 52
selling, 90–94 web usability guidelines, 168
supporting, 97–101 revenue, calculating, 91–92
reverse card sort test, conducting, 143–147
ROI (return on investment)
Q determining for search usability, 90–91
QA (quality assurance) versus IA (information increasing via queries, 29
architecture), 139 obtaining, 18
queries on selling ads, 95
formulation of, 39 for selling products and services, 90–91
number of words per, 45 for supporting existing products and
types of, 29–30 services, 98
query data, categories for site searches, 155–156
query words. See also words versus actions
S
emphasizing, 22
highlighting, 22–28 satisfaction, measuring for websites, 6
plural form of, 54 scanning behavior, considering, 80, 85
querying behavior. See also usability scent of information. See also keywords
professionals above the fold, 53
defined, 4 applying to pictures and captions, 114
importance of, 120 beginning, 22
perceptions of, 19–21 on elevator and lobby, 8–10
 INDEX 179

evaluating, 76–77 number of characters in, 24

for file tabs, 78 and transactional queries, 67
identifying copy areas for, 113–115 search professionals, cooperation with usability
including in search listings, 26 professionals, 125
increasing, 19 search queries, formulation of, 38
losing, 133–136, 156 search results, flaws in, 20–21
maintaining for copy, 112–113 search usability
maintaining on long pages, 79 addressing problems with, 161
misleading, 80 applied, 4
on National Library of Medicine’s home combining with SEO efforts, 96–97
page, 11–13 creating visual hierarchy for, 110–112
providing with URLs, 27 decreasing, 65
on shopping malls, 10–11 defined, 3, 7
validating, 22 for employee intranets, 101–102
visible signs of, 22 formatting copy for, 116–117
website resource for, 23 and graphic designers, 109–112
Schaffer, Eric, 148 importance of, 13–14
screen reader, defined, 79 increasing, 66
search, definitions of, 1–2 and information architects, 136–139
search engine marketing (SEM), 38–39 and IT (information technology), 132–136
search engine optimization (SEO), overview lowering customer service expenses with,
of, 5 99–101
search engine results page (SERP), highlighting and management, 125–130
elements of, 22 and market researchers, 130–132
search engine traffic, analyzing quality of, and marketing communications, 117–119
121–122 selling ads with, 94–97
search engines. See also site search engines; web selling products and services with, 90–94
search engines
and usability professionals, 119–125
determining rankings of, 35
and web copywriters, 112–117
spam, 21
search usability goals
web versus site, 2–4
bounce rates, 103–104
search listings
page views, 104–105
appearance of, 22
time-on site, 104–105
example of, 24
top SERP rankings, 102–103
finding for keywords phrases, 75
searchers
including keywords in, 28
anticipating needs of, 75–77
including scents of information in, 26
behavior of, 37–38
and informational queries, 55–57
characterizing, 18
and navigational queries, 42–44
180 INDEX

searchers (continued) site search engines. See also search engines; web
goal of, 42 search engines
improving experiences of, 34 assessing user-friendliness of, 7
revealing motivations of, 39 defined, 2
searches site searches
of e-commerce sites, 48 creating, 135–136
intents of, 50 maintaining scent of information in, 135–136
personalized, 35 reducing, 118
secondary call to action, creating, 73–74 resource for, 136
selling versus web searches, 20
ads with search usability, 94–97 site use, myth about, 164
products and services, 90–94 site visitors, behaviors of, 13–14. See also users
SEM (search engine marketing), 38–39 sites. See websites
“sense of place,” establishing, 8 snippets, displaying for meta-tag descriptions,
SEO (search engine optimization), 25–26
overview of, 5 sound and motion, grabbing attention with,
SEO efforts 111–112
applying to selling ads, 95 spam, use as marketing strategy, 21
combining with search usability, 96–97 splash page, defined, 65
combining with web usability, 93 Spool, Jared
SEO improvement, example of, 92 “button of doom,” 117–118
SEO professionals, goals of, 21 eight-second usability test, 153
SEO traffic, impact on customer support calls, five-second test, 76
99–100 losing scent, 135
SEOs, working with marketing, 103 pogo-sticking, 86
SERP (search engine results page), highlighting scent of information report, 23
elements of, 22 stakeholders
SERP rankings, impact on company reputations, graphic designers, 109–112
102–103 information architects, 136–139
services and products, selling, 90–94 IT (information technology), 132–136
shopping malls, similarity to websites, 11 management, 125–130
Sinatra, Frank, 69 market researchers, 130–132
site analytics, importance of, 120 marketing communications, 117–119
site maps, making user-friendly, 118–119 usability professionals, 119–125
site navigation web copywriters, 112–117
evaluating effectiveness of, 75 subroots, placing keywords in, 28
failure of, 20 surfing, defined, 3
site search data, reviewing, 155–156 surveys versus usability tests, 132
SWAG acronym, meanings of, 91
 INDEX 181

T purposes of, 23–24

using in orienting behavior, 82
taglines, using in orienting behavior, 82–83
TLD (top-level domain) extensions, examples of,
term highlighting, defined, 22
36
test participants
traffic. See search engine traffic
for brand perception tests, 158
transactional intent. See also e-commerce sites
choosing, 143
buying products, 63–64
for free exploration tests, 160–161
determining, 62–63
for functional salience test, 148
entertainment, 64–65
qualities of, 142
transactional queries
for reverse card sort test, 143
explained, 29–30
for usability testing, 167
goals of, 61–62
testing, A/B versus multivariate, 67. See also
important items in, 67
usability testing
indicators, 67
tests
versus informational queries, 122–124
brand perception, 157–159
optimizing for, 68–69
click affordance, 150–151
and search listings, 67
eight-second usability on content pages,
153–154 testing, 41–42
expected content, 152 using, 30
free exploration, 160–161 transactional versus informational sites, 56
functional salience, 148–149 travel buyers, behavior of, 43
home page expectancy, 149–152
observing navigational web searches, U
154–155
performance, 75 underlined text, using, 110–111
pre-testing, 149 URL structure, importance of, 134–135
reverse card sort, 143–147 URLs (uniform resource locators)
text importance to navigational queries, 45
changing to graphic images, 111 including keywords in, 28
deciding on type of, 110–111 length of, 43
making clickable, 110–111 in navigational queries, 43
third-level call to action, creating, 73–74 number of characters in, 27
“three-click” rule, myth about, 165 providing scent of information with, 27
Thurow, Shari, 38 structure of, 25–28
time on site, defined, 165 using in orienting behavior, 82
time-on-site metrics versus page views, 104–105 usability
title tags combining with SEO efforts, 93
format of, 23 perceptions of, 6
including keywords in, 23–24 usability guidelines, myth about, 168
182 INDEX

usability professionals. See also querying behavior videos, impact on search usability, 65
cooperation with search professionals, 125 viewing related items, 74–75
goals of, 8, 21 visit to call ratio, determining, 98
neglect of querying behavior, 120 visual hierarchy, creating for search usability,
responsibilities of, 119 110–112
and search usability, 119–125 Visual Organization, Gestalt Principles of, 10
usability testing. See also five-second test; testing Vojnovic, Milan, 34
anticipating searcher needs with, 75–77
conducting, 75
W
versus focus groups, 131
increasing accuracy of, 167 web addresses
myth about accuracy of, 167 importance to navigational queries, 45
participants required for, 143 including keywords in, 28
for pogo-sticking, 86–87 length of, 43
versus surveys, 132 in navigational queries, 43
user behaviors number of characters in, 27
determining reasons for, 130 providing scent of information with, 27
observing, 37–38, 117–118 structure of, 25–28, 134–135
reading web pages, 72 using in orienting behavior, 82
user confidence, encouraging, 58 web analytic terms, 165
user expectations “web is yours” myth, 168
assuming, 66 web pages. See pages
versus business goals, 21 web search engines. See also search engines; site
user feedback, myth about omission of, 163–164 search engines
users. See also poor user experience; site visitors defined, 2
“always right” myth, 166–167 goal of, 42
consulting in usability testing, 7 query types, 29–30
identifying expectations of, 127, 129 use of, 18
knowing before building, 127 web search results, contents of, 56
as “the problem,” 166 web searchers
users’ language anticipating needs of, 75–77
reflecting in copy, 115–116 behavior of, 37–38
using, 128 characterizing, 18
users’ success rates, improving, 118–119 goal of, 42
improving experiences of, 34
revealing motivations of, 39
V web searches
verbs, using in transactional queries, 67 goals and characteristics of, 19
video optimization, resource for, 52 monitoring use of, 34
 INDEX 183

versus site searches, 20 website usability, overview of, 6–8

web usability websites. See also National Cancer Institute
addressing problems with, 161 website
combining with SEO efforts, 93 beginning design of, 137–138
impact of management decisions on, building using iterative design, 132–133
129–130 determining locations in, 81
website resources. See also resources navigating to, 39–40
findability statement, 8 similarity to shopping malls, 11
five-second test, 76, 153 usability testing of, 6–8
keyword research, 125 Weinberger, David, 168
navigation pages, 118 Wikipedia listings, appearance of, 56
pogo-sticking, 86 words versus actions, 51. See also query words
Principle of Proximity, 10
querying research, 34
scent of information, 23 Y
taxonomy of web search, 29 Yi et al, 34
usability, 6 you are here cues, using, 81–84
usability testing, 148
Be inspired—open a New Riders
Voices That Matter book today!
Voices That Matter is a series of books reserved for the voices that are
revolutionary and influential in their field. These books use passionate
storytellers to teach, challenge, and force us to seek new solutions and
view the world from a different perspective.

The books celebrate the knowledge and practical know-how of leading

web design and development experts. These authors have unique
vision—often transforming the digital landscape with their ideas.
Whether you’re looking for foundational information or desire to move
your skills beyond the ordinary, New Riders’ Voices That Matter series
will take you where you want to go.

Neuro Web Design: What Makes Them Click?

Susan M. Weinschenk
ISBN 0-321-60360-5
Information Architecture: Blueprints for the Web,
Second Edition
Christina Wodtke and Austin Govella
ISBN 0-321-60080-0
Fresher Styles for Web Designers: More
Eye Candy from the Underground
Curt Cloninger
ISBN 0-321-56269-0
Designing the Moment: Web Interface Design
Concepts in Action
Robert Hoekman, Jr.
ISBN 0-321-53508-1
Designing for the Social Web
Joshua Porter
ISBN 0-321-53492-1
Building Findable Websites: Web Standards,
SEO, and Beyond
Aarron Walter
ISBN 0-321-52628-7
Bulletproof Web Design: Improving flexibility
and protecting againstworst-case scenarios
with XHTML and CSS, Second Edition
Dan Cederholm
ISBN 0-321-50902-1
Transcending CSS:The Fine Art of Web Design
Andy Clarke, edited by Molly E. Holzschlag
ISBN 0-321-41097-1

Visit www.newriders.com W W W. N E W R I D E R S . C O M
Search Engine Visibility
Second Edition
“Search Engine Visibility is the best introduction to
SEO on the market. It breaks down the walls between
engineering, marketing, and design by illustrating
how structure, content, and code can work together
to deliver optimal findability and a better user
experience.”

PETER MORVILLE, Author of Ambient Findability:

What We Find Changes Who We Become

Search Engine Visibility is a must-read for Web

designers and developers, search engine optimizers,
usability professionals, and marketers who are
seeking to create more effective Web sites that meet
both user and business goals. Besides outlining ways
to obtain top positions in search results, this book
shows you how to design, write, and create user-
friendly Web sites. This new edition of Search Engine
Visibility includes up-to-date coverage of blogs,
podcasts (both video and audio), Web applications,
and more. It also shows you pitfalls to avoid.

Topics covered in this book include: Search Engine Visibility, Second Edition
Shari Thurow
• How to write search engine friendly sales copy
ISBN: 9780321503244
• Search engine friendly design solutions 312 Pages
• Ways to increase your Web site’s popularity

• Solutions for dynamic Web sites

Get free online access
to this book for 45 days!
And get access to thousands more by signing
up for a free trial to Safari Books Online!
With the purchase of this book you have instant online,
searchable access to it for 45 days on Safari Books Online!
And while you’re there, be sure to check out Safari Books
Online’s on-demand digital library and their free trial offer
(a separate sign-up process). Safari Books Online subscribers
have access to thousands of technical, creative and business
books,instructional videos, and articles from the world’s
leading publishers.

Simply visit www.peachpit.com/safarienabled

and enter code PNJQFDB to try it today.