BA 4: WEB AND SOCIAL MEDIA ANALYTICS

MODULE II

New Web Analytics 2.0 Mind set:

 Eight Critical web Metrics.

 Click Stream Analysis.

 Best Web Analytics Reports.

EIGHT CRITICAL WEB METRICS
• A metric is a quantitative measurement of statistics describing events or
trends on a website.
• A key performance indicator (KPI) is a metric that helps you understand
how you are doing against your objectives.
• That last word—objectives—is critical to something being called a KPI,
which is also why KPIs tend to be unique to each company.
Visits and Visitors
• Visits and Visitors form
the basis of nearly every
web metric calculation.
• These are prominently
displayed in web
analytics tool.
• We also find them in
search reports, exit pages,
bounce rate computation,
conversion rates, and so
on.
• So, Visits and Visitors are
very important.
VISITS
• Visits report the fact that someone came to your website and spent some time
browsing before leaving.
• Technically this visitor experience is called a session. Sessions are most commonly
referred to as Visits.
VISIT RECORDING PROCESS:
1. When someone requests the first page or item from your website, then your
analytics tool starts a session for that person from that browser.
2. Each additional request from that person is attached to a unique session ID.
3. When the person leaves your site, that unique session ID is used to combine
together the pages viewed into one complete visit.
4. When you run a report for any given period in your web analytics tools, Total
Visits is the count of all the sessions during a given time period.
UNIQUE VISITORS
Unique Visitors is the number of people who come to your website.
UNIQUE VISITORS RECORDING PROCESS:
1. when someone requests the first page or item from your website, your analytics
tool will set a unique cookie on that person’s browser.
2. This cookie remains on the browser even after the person leaves your website. It
contains a unique anonymous string of numbers and characters.
3. Each time someone visits your website from that browser, this persistent cookie ID
is used to recognize that the same browser has returned.
4. When you run a report for any given time period in your web analytics tool, the
Unique Visitors metric is the count of all the persistent unique cookie IDs
during a given time period.
This metric is not always accurate as it is affected by browsers that don’t accept cookies
or reject third party cookies. It is a good approximation of the number of people
visiting your website.
DIFFERENT MEASURES OF UNIQUE VISITORS:

1. Daily Unique Visitor: Person who visits on a particular day (even multiple times
in that day) is counted as one daily unique visitor.
2. Weekly Unique Visitor: Person who visits in a particular week (even multiple
times in that week) is counted as one weekly unique visitor.
3. Monthly Unique Visitor: Person who visits in a particular month (even multiple
times in that month) is counted as one monthly unique visitor.
4. Absolute Unique Visitor: Person who is a new visitor to the website during the
entire monitoring period.
1. Daily Unique Visitors : 19.
2. Weekly Unique Visitor : 15.
3. Monthly Unique Visitor : 12.
4. Absolute Unique Visitor : 9.
Time on Page and Time on Site
After Visits and Visitors, perhaps the next foundational metric in web analytics is Time.
It measures the time that visitors spend on an individual page and the time spent on
the site during a visit (session).
To understand time, we will use a simple scenario, illustrated in figure below.
Someone surfs over to your website and requests your home page, which starts a visit
(session) on your website. The visitor then requests two more pages from your site
before deciding to leave your website.
Time on Page (Tp) represents the time spent on each page.
Time on Site (Ts) represents the time spent during that session on the website.
PROCESS FOR COMPUTING TIME METRICS:
Time spent on 1st page is calculated as the time difference between request for 2nd page
minus the request for 1st page.

Tp (Home Page) = Request Time (Page2) – Request Time (Home Page)

Tp (Home Page) = 10:01 – 10:00
Tp (Home Page) = 1 minute.

Tp (Page 2) = N/A (not available).

Because we don’t know the request
time for Page 3.
Only way the analytics tool knows how long someone spent on one page is by looking
at the two time stamps: one from the request for the first page and one from the request
for the second page.

Tp (Home Page) = 1 minute.

Tp (Page 2) = 4 minutes
Tp (Page 3) = N / A
The analytics tool has no idea how long the visitor spent on the last page on your site.
This flaw is true for nearly all web analytics programs in terms of default behavior.
Let’s summarize the metrics using Figure below.

Tp (home page) = one minute

Tp (page 2) = four minutes
Tp (page 3) = zero minutes
Ts = five minutes (Time on Site, also known as Session Length)
SPECIAL CASES OF TIME COMPUTATION:

Lesson 1: The Single-Page View Session

Figure illustrates a visit to your website
that had only one page view, and then
the person left your website.
The analytics tool records when the page
was requested (10:00), but it does not
know when the exit happens.

Tp = zero minutes
Ts = zero minutes
Be aware of this important fact as you report your analytics data.
Lesson 2: The Case of Tabbed Browsing
Nearly all browsers now allow you to open different tabs for links on a site. This creates
an interesting scenario from a measurement perspective. Figure below illustrates the
scenario.
A visitor comes to your home page.
From there your visitor opens the first link in a new tab but continues to scan the home
page.
He clicks a link to page 2 from the home page, then on to page 3, and then closes the tab
(or moves away and forgets about it).
The visitor goes to the tab opened from the home page to page 4 of your site, spends
time there, and goes on to page 5 in that tab.
Then the visitor exits.
How is time on site computed?
Some analytics tools will simply create two sessions for this visitor and measure time
separately for both sessions using the method that we have learned in this chapter so
far.
Most web analytics tools will collect all the requests during this session and normalize
the tabbed browsing behavior. Figure below shows how the normalization happens.

Tp (home page) = one minute Tp (page 4) = one minute

Tp (page 2) = three minutes Tp (page 3) = two minutes
Tp (page 5) = zero minutes Ts (session duration) = seven minutes
Bounce Rate
Bounce rate is one of the most insightful web analytics metric. It is defined as the
percentage of sessions on your website with only one page view.

It is a standard metric that is available in all web analytics tools.

It is easy to understand.
It is actionable metric.
It tells you a lot about customer behavior.

A few tools in the market allow you to use Time to measure Bounce Rate. That
is, they measure the percentage of sessions where Time on Site was less than five
seconds.
Figure below shows how Bounce Rate looks in the French tool XiTi.

On the left are all the people who visited the website during this time period, 65,569.
To the right of that, the number on the top represents those who chose to click any link
on the landing page and see more than one page (22%).
On the bottom is the big number, 78% of the people who came to this site and did not
click on any link.
Bounce rate tells you how uninteresting you website or the web page is.
So, you need to measure the Bounce Rate of your site on at least two levels. You need to
measure in aggregate at an entire site level, and you need to measure the Bounce Rate
of your top landing pages.
Exceptions to Bounce Rate:

• If the success of your site depends on users viewing more than one page, then a high
bounce rate is bad. For example, if your home page is the gateway to the rest of your
site (e.g., news articles, product pages, your checkout process) and a high percentage
of users are viewing only your home page, then you don’t want a high bounce rate.

• However, In case of websites such as blogs or single page websites, measuring the
Bounce Rate metric in aggregate might be inappropriate. On such sites people mostly
come only to read latest post or get latest news.
Improving Bounce Rate:
• If your overall bounce rate is high, then you can dig deeper to see whether it’s
uniformly high or whether it’s the result of something like one or two channels,
sources or just a few pages.
• Examine whether the page content correlates well with the marketing you use to
drive users to those pages, and whether those pages offer users easy paths to the next
steps you want them to take.
• You may want to reevaluate your overall site design and examine the language,
graphics, color, calls to action, and visibility of important page elements.
• You can use Optimize to test different versions of your site pages to see which
designs encourage users to engage more.
• Bounce rate for websites such as blogs may be improved by including activities like
subscribing to your feed or adding comments and feedback / rating sections.
Exit Rate
• Exit rate measures how many people leave the website from a certain page on the
website. For example, a website can have 5 pages and the exit rate metric can shown
in analytics report as below: Page Exit Rate
Home Page 10%
Products Page 27%
About Us Page 33%
Careers Page 15%
Contact Us Page 15%

• As an example, let’s say that someone lands on the home page of an electronic store.
They tend to move to a specific product page, and then they leave the website.
• This means the visitor left your website on the product page.
• Therefore, the exit rate would be calculated as the number of total exits from that
page / number of visits to that particular product page.
• Exit rate is supposed to show the leakage from your website.
• In other words, where do people exit after they start their session? It should illustrate
pages that you should fix to prevent leakage and get customers to buy more or sign
up more.
• The problem is that everyone who comes to your website has to leave. They will
browse around your website and leave from any arbitrary page. Their exit from a
page is no indication of the quality of that particular page.
Conversion Rate
• Conversion Rate is defined as Outcomes divided by Unique Visitors (or Visits). It
is measured as a percentage.
• Outcomes are customarily the submission of an order on ecommerce website.
• For example: 5000 customers (unique visitors) visited www.amazing.com on a
particular day and out of them 3000 customers placed orders (objective) for some
products. Then, Conversion Rate = (3000 / 5000) * 100 = 60%.
• Conversion rate is obviously the most important metric in web analytics as it tells us
how successful we are in achieving the objectives of our website.
• There is one question about computing conversion rate: Should you use Unique
Visitors or Visits?
• If you choose Visits as the denominator, you assume that every visit to your website
is a chance to get someone to place an order and get someone converted.
• If you choose Unique Visitors as the denominator, you grant that it is OK for a person
to visit your website multiple times before making a purchase. That behavior is a lot
more common on the Web.
Engagement
• Dictionary defines engaging as “tending to draw favorable attention or interest.”
• We should all try to create website experiences that draw favorable attention or
interest.
• The challenge in the context of measurement is that “favorable attention or interest” is
extremely hard—if not impossible—to measure.
• Many tools measure the time a Visitor spends on the website and call it Engagement.
For example: A tool can set a threshold of 5 minutes of Time Spent on Website to be
considered as Engaged Visitor. So, If 700 unique visitors out of Total 1000 daily
unique Visitors spend more that 5 minutes on a website, the we can say that 70%
engagement has been achieved.
• However, a visitor can spend 4 minutes and place an order on the website. Whereas,
another visitor can spend 10 minutes on the website and be frustrated that he is not
able to find a satisfactory product or solution on the website.
• Same problem exists with tools who define Engagement as the number of repeat
visits by a visitor.
• A visitor may have to repeatedly visit the site due to technical problems faced by him.
This is not taken into account by the analytics tool. How do you distinguish those
visits from someone who visits the website site regularly to learn about the latest
products and updated features?
• Quantitative data (web analytics) is limited in that it can measure the degree of
Engagement (frequency of visit) but not the kind of Engagement (positive or
negative).
Attributes of a Good Metric
A good web metric should be:
1. Uncomplex (Simple): Metric should be easily understood by everyone in the
organization.
2. Relevant: A good metric should be relevant to your business. Every organization /
business is unique and they should use a metric which is suitable to their business.
3. Timely: A good metric should be available in time for the management to make
decisions or take corrective action. If the metric is available in real time, it is great.
4. Instantly Useful: Instantly useful is when you understand quickly what the metric
is and you can get the insights as soon as you look at it.
Example of a Great Web Metric:
Bounce rate is uncomplex and means a visitor just came and left your website. It tells
you about the attractive and unattractive pages on your site which is a very relevant
information for you. It is timely because all web analytics tools provide bounce rate on
a daily basis. It is instantly useful as you can just look at it and know what needs
attention.
Best Web Analytics Report
• If you could pick only one web analytics report to refer for your website performance,
which report would it be?
• It really depends on your business. What are the strategies you are executing? Are
you B2B or B2C?
• There is one report that will work for any type of website, and it qualifies as the best
web analytics report: Outcomes by All Traffic Sources.
• This report fits perfectly with the perspective and objectives of Web Analytics 2.0 and
hence is immensely actionable.
• Figure below shows how an Outcomes by All Traffic Sources report looks, in Google
Analytics.
• In a report full of metrics in our web analytics tools, this report represents two things
you should care about more than anything else: sources of traffic and Outcomes.
Sources of Traffic:
• If you know the source of your traffic, then you can strongly infer what kinds of
people are coming to your website and even a little bit of why they are coming
(intent).
• Direct Traffic represents free traffic because it comes from people who arrive via
using bookmarks, typing in your URL , or other such activities.
• Direct Traffic is also traffic that is familiar with you, so it typically represents
returning Visitors and most likely your existing customers.
• In Figure above Direct Traffic is a big number—and it converts higher, which is very
typical. But if you had spent a lot of money trying to attract new Visitors (prospects)
to your website, a big number from Direct Traffic might not be such a good thing.
• Notice the box drawn around twitter.com in Figure. Suppose, In the prior 30 days, it
was not even in the top 20, and now it’s at 5.
• Immediately you can see how social media efforts might be paying off. You can also
infer that these are much more tech-savvy people. The source helps you better
understand the persona of the audience.
• Row 6, stumbleupon.com, represents new Visitors who might typically be interested
in recent stories, or if this were an ecommerce website, it would represent recent
promotions, product launches, and so on.
• Stumbleupon.com, digg.com, and others also represent a sense of validation that your
content is good and it is being spread by others.
• Row 9, wilsonweb.com, represents very traditional direct response and traditional
marketers.
• Finally, the importance of Google to this website is very clear. The numbers validate
that the work put into search engine optimization (SEO ) is paying rich dividends.
• A specific example of this is row 4, which is traffic from images.google.com; that is a
validation of the time and effort spent tagging each image on the website with
relevant descriptions.
• These examples should show you how much you can learn by understanding the
sources that send you traffic. You can see what’s working and what’s not in terms of
your core acquisition strategy and whether you are attracting the right audience to
your site.
Outcomes:
• Figure shows Conversion Rate for a non-ecommerce website. At a glance, we can see
the overall Conversion Rate, and also go down the list of the websites and very
quickly identify which ones are sending quality traffic, in other words, traffic that
takes action that adds to my bottom line.
• I can also quickly identify my BFF s (twitter.com in this case). I can quickly see
sources that send me lots of traffic yet are not my real BFF s (stumbleupon.com in this
case, or even images.google.com). Notice how in one view, Visits, you come to
different conclusions than looking at another, perhaps more important, view.
• You can dive into many more nuances in this report. Notice there are multiple goals
for you to consider. Each part of your business might have different Outcomes, or
goals, which are shown in the report as Goal2, Goal3. and Goal4.
• The report will tell each business unit or leader how they are performing against their
unique goal. The essence of the best web analytics report is that it highlights two
questions to focus on first: who? and how much?
 T h a n k Yo u ! !
Important Questions
• Explain the meaning and importance of web metrics related to visits and
time in any web analytics report?
• Explain the meaning and importance of Bounce rate, Exit rate, Conversion
rate and Engagement web analytics report?
• What are the characteristics of a good web metric? Explain with an example.
• Which report of metric would you define as the best web analytics report?
Justify your answer with the help of an example.