digital marketing -advertising

Allstate
Changing How the World Sees
Digital Advertising
LINDA ABRAHAM Co-Founder, CMO & EVP of Global Development
ANNE HUNTER SVP, Advertising Effectiveness
ANDREA VOLLMAN Marketing Director
vCE™ CHARTER STUDY REVEALS HOW VALIDATED ADS ARE IMPACTING
THE FUTURE OF DIGITAL AND CROSS-MEDIA MEASUREMENT
MARCH 2012

2
Contents
Introduction
Validated Campaign Measurement
The vCE Charter Study
Executive Summary of Findings
In-View
Audience
Geography
Brand Safety
Fraud
Implications: Putting all of the Pieces Together
Conclusion
About comScore
3
4
5
7
8
16
21
23
25
28
30
31

3
Across the globe, digital media has become an important
component of every advertiser’s marketing mix. According
to the Interactive Advertising Bureau (IAB), display-related
advertising spending in the United States (U.S.) reached $10
billion in 2010 and has grown at 20%+ rates since then, far
exceeding the growth of traditional media. Just as we’ve seen
tremendous growth in terms of the volume of digital advertising,
the landscape itself has also experienced a massive evolution.
From new ad formats and placement strategies to new delivery
systems and ad technology, it has become challenging for
players across the industry to stay up-to-speed.
Until now, digital advertising
measurement has not kept pace with
the complexity of these changes.
The transactional focus has been on
measurement of gross impressions
delivered, as opposed to those that were
actually seen by consumers in a particular
target. As a result, marketers have been
limited in their ability to understand how
online advertising works, especially when
compared to other media channels.
This lack of understanding has resulted
in reluctance by many marketers to
fully embrace digital advertising. From
publishers to ad networks and from
marketers to agencies, key players in the
space are calling for more transparency
and greater accountability as it relates to
online ad delivery.
Addressing this industry-wide call-to-
action, the IAB, the American Association
of Advertising Agencies (4As) and the
Association of National Advertisers (ANA)
– each representing a key constituent
group in the advertising market – jointly
launched an initiative called, Making
Measurement Make Sense (3MS). Simply
put, 3MS’s goal is to improve, standardize
and simplify digital media measurement.
In order to reach this goal, 3MS has
published guidelines and is conducting
research to help address issues
surrounding ad delivery, measurement
and validation.
Introduction
Stay Connected
Follow @comscore
FOR FURTHER INFORMATION,
PLEASE CONTACT:
Andrea Vollman
comScore, Inc.
+1 212 497 1731
press@comscore.com

4
In January 2012, comScore released a breakthrough innovation
to the marketplace that addresses many of the guidelines
outlined in the 3MS initiative as well as some additional
industry issues relating specifically to ad delivery validation.
This solution, validated Campaign Essentials™ (vCE™), provides
an unduplicated accounting of impressions delivered across a
variety of dimensions, helping to significantly improve the value
of online advertising.
Importantly, vCE gleans all this information via a single ad tag, thus enabling a
comprehensive, but holistic, view of digital ad delivery that is unique to the marketplace.
The use of a single ad tag is a critical component of this measurement approach as it
evaluates all impressions consistently and applies validity conditions simultaneously. This
eliminates all issues associated with duplicated measurement and offers a more accurate
view of campaign delivery. Duplication and inconsistency typically arise when disparate
data-collection sources are merged, which can dramatically impact the quality of the
analyzed data.
vCE validates whether or not impressions delivered as
part of a campaign were:
In-view (i.e. viewable by an actual consumer)
Delivered in the right geography
Seen in brand safe environments
Absent of fraud
In addition, vCE evaluates the demographic and behavioral composition of the
campaign audience, enabling the advertiser to assess the degree to which validated
impressions reached the desired campaign targets.
Validated Campaign
Measurement

5
To better understand issues
associated with display ad
delivery and validation, and to
test-drive vCE, twelve leading
marketers participated in
a U.S.-based charter study,
called the vCE Charter Study.
The eye-opening findings help to pave
the way for a more accurate measure of
campaign delivery that relies on validated
impressions, rather than served impressions
(or gross impressions), which are currently
the established currency for online ad
measurement. Validated impressions can
also be used to report validated gross and
target rating points (vGRP/vTRP).
Ideally, this research will help to promote the
broad adoption of new standard measures
that reflect the true delivery of a campaign
(per the 3MS guidelines), and it will also
help to generate greater visibility and
transparency across the industry and across
media. Throughout 2012, similar charter
programs will be rolled out in other global
markets, including Canada, Latin America,
Asia and select European countries.
Parameters and Methodology
Study Participants:
Importantly, 100% of the vCE Charter
Study ad impressions were delivered in
iframes, including a majority of ‘cross-
domain’ iframes. The definition of these
iframes is discussed in the In-View section
of this paper, but it is important to note that
this is the first industry study
to measure and report on
in-view rates for ads delivered
via all iframes, including those
delivered via the notoriously
difficult-to-measure cross-
domain iframes.
For the purposes of this report, all
findings are presented in aggregate, not
by individual campaign, to protect the
confidentiality of client data. Findings are
reported by total campaign as well as by
publisher-level, placement-level and/or
creative-level.
It should also be noted that because vCE
Charter Study participants included major
branded advertisers, who inherently buy
more premium inventory than the average
online marketer, the study findings are not
necessarily representative of the overall
online advertising market. In fact, because
these advertisers generally engage in high-
end, premium campaigns, the findings may
represent ‘best-case scenarios,’ rather than
the norm.
This is the first
study to bring
twelve leading
marketers
together to
holistically
understand how
online advertising
is delivered.
Time Period: December 2011
Total Campaigns: 18
Media Placements: 2,975
Site Domains: 380,898
Ad Impressions: 1.8 billion
Format: All ads were display, delivered via iframes.

6
In-view: In-view is defined as an ad
impression with at least 50% of the ad’s
pixels in the user’s viewport for one second
or more. This definition is consistent with
current working standards outlined as
part of the 3MS initiative. The parameters
for the definition of in-view can be easily
changed to accommodate any change in
industry standards.
Audience: Using the comScore panel of
2 million global consumers, comScore is
uniquely qualified to report on audience
delivery with person-level insights. This
means the study was able to validate
delivery to target audiences based on
traditional demographics as well as more
than 80 behavioral segments.
Geography: Geographic validation is
measured by country on a global basis.
Although vCE is available globally, with
regional data available in some countries,
for the purposes of the vCE Charter Study,
all campaigns were validated based on
delivery in the U.S.
Brand Safety: Ads delivered on sites
deemed not appropriate for brand
advertising due to objectionable content
are considered to be in violation of brand
safety. The definition of objectionable
content is further discussed within the
Brand Safety section of the paper.
Fraud: Fraud was measured by counting
ad impressions served to non-human
agents as per the IAB spiders and bots
list as well as ads that were served to
users via illegitimate methods or content.
Although there are several other types
of fraud detections, these two very basic
types were included in the vCE Charter
Study to establish a baseline.
The goal of the vCE Charter
Study was to quantify the
incidence of sub-optimal ad
delivery across these key
dimensions for the advertised
brands, and in so doing, frame
the relative importance of
each for the industry. Although
vCE offers the ability to
optimize campaigns in-flight in
order to eliminate waste and
generate better advertising
outcomes, this feature was not
deployed for the purposes of
the study, as it would detract
from the study’s objective
of determining a baseline
of delivery prior to in-flight
optimization.
The vCE
Charter Study
Key Metric Definitions

7
1 In-View Rates are Eye-Opening
The study showed that 31% of ads were
not in-view, meaning they never had an
opportunity to be seen. There was also
great variation across sites where the
campaigns ran, with in-view rates ranging
from 7% to 100% on a given site. This
variance illustrates that even for major
advertisers making premium buys, there
is a lot of room for improvement.
2 Targeting Audiences Beyond
Demos Can be Powerful
Generally, campaigns that had very basic
demo targeting objectives performed well
with regard to hitting those targets. For
example, those with an objective of reaching
people in a particular broad age range did so
with 70% of their impressions. Predictably,
as additional demographic variables were
added to the targeting criteria (i.e. income
+ gender), accuracy rates of the ad delivery
declined. However, the results also showed
that, on average, 36% of all impressions in a
campaign were delivered to audiences with
behavioral profiles that were relevant to the
brand (i.e. consumers with demonstrated
interests in categories, such as food, auto
or sports). One campaign had 67% of its
impressions viewed by the target behavioral
segment, demonstrating that targeting to
people based on interests or behaviors holds
strong potential.
3 The Content In Which An Ad
Runs Can Make or Break a Brand
Of the campaigns analyzed, 72% had
at least some impressions that were
delivered adjacent to objectionable
content. While this did not translate to
a large number of impressions on an
absolute basis (141,000 impressions
across 980 domains), it is important to
note that 92,000 people were exposed to
these impressions. This demonstrates that
even with the most premium of executions,
brand safety should be an utmost concern
for advertisers.
4 Fraud is the Elephant
in the Digital Room
Fraud is an undeniably large and growing
problem in digital advertising. The results
showed that an average of 0.16% of
impressions across all campaigns was
delivered to non-human agents from the
IAB spiders & bots list. Although this
percentage might appear negligible, there
are two important considerations to keep
in mind. Only the most basic forms of
inappropriate delivery were addressed
in this study. When additional, more
sophisticated types of fraud are considered,
the problem will only get larger. Like brand
safety, fraud should be an important
concern for all advertisers.
5 Digital Ad Economics: The Good
Guys Aren’t Necessarily Winning
The study showed that there was little
to no correlation between CPM and
value being delivered to the advertiser.
For example, ad placements with strong
in-view rates are not getting higher
CPMs than placements with low in-view
rates. Similarly, ads that are doing well
at delivering to a primary demographic
target are not receiving more value
than those that are not. In other words,
neither ad visibility nor the demographic
target delivery is currently reflected in the
economics of digital advertising.
Executive Summary
of Findings

8
One of the most fundamental aspects
of advertising measurement, particularly
as it relates to cross-media, is the need
for a solid and consistent method of
determining whether a consumer had
an opportunity to see (OTS) an ad.
In television, once an ad is delivered in
a program, it plays, meaning that the
consumer had an opportunity to see it.
While the person might not have been
in the room to see the ad, the industry
accepts the notion that the opportunity
was still there and therefore it gets
counted as such. Alternatively, if the
television is turned off, there isn’t an
opportunity for it to be seen.
The advertising industry has accepted OTS
as a standard metric, which many rely on to
build cross-media campaigns and to assess
the effects of advertising across channels.
This metric is particularly important based
on the very simple fact that:
If an ad does not have
an opportunity to be seen
by a real user, then it
cannot possibly deliver
its intended effect.
When compared to other forms of
media, digital advertising has unique
characteristics relating to an ad’s
opportunity to be seen. To date, the
standard has simply been to measure
whether ads were served to a page.
However, there are many reasons why
a served digital ad might not result in
someone having an opportunity to see
it. For example, consumers often land on
a particular page and then quickly scroll
down to consume content before the
banner ad at the top of the page had a
chance to load. An alternative scenario
is when a user remains at the top of
the page, never scrolling to the bottom
where many ads have loaded. Given
these scenarios, which inherently result
in many ad impressions being delivered
but not seen, the industry has begun to
evaluate ways to accurately measure
viewability and to improve in-view rates to
avoid wasted ad spend. 3MS proposed
a standard definition of in-view, which
states that at least 50% of the pixels of
the ads must be in-view for a minimum
of 1 second.
Aside from adhering to the 3MS proposed
working definition, viewability measurement
must also account for all ad delivery
formats in order for it to be accurate. There
are three distinct ad delivery formats from
which publishers can choose to deliver
ads, and these are:
Format I
Delivery of an ad directly
on a publisher site:
In this instance, the publisher places a
JavaScript ad tag on its page with the
marketer’s ad tag in the same domain
as the site content.
Format II
Same-domain (‘friendly’) iframes:
Many sites choose to use an iframe to
deliver advertising on their site, as it can
help to prevent any unwanted content
associated with the ad from damaging the
main site content. Same-domain iframes,
also known as friendly iframes, typically
refer to instances when a site allows the
iframe to communicate directly with rest
of the page, which in turn, facilitates the
measurement of the iframe location when
the page is rendered on the viewable
screen. This helps to determine whether
the ad is in-view and for how long.
Format III
Cross-domain (‘unfriendly’) iframes:
If a site chooses not to allow the ad to
communicate directly with the page, it
reserves a place for it in an iframe, which
Aside from
adhering to the
3MS proposed
working definition,
viewability
measurement
must also account
for all ad delivery
formats in order
for it to be
accurate.
In-View
defining In-View

9
calls a third-party domain to serve the ad.
This severed communication link presents
a daunting challenge to the measurement
of the iframe’s position on the page, and,
therefore, ad visibility.
The vCE technology is unique
in the marketplace as it is
the first and only that can
see through cross-domain, or
unfriendly, iframes, which means
that vCE’s in-view rate accounts
for all ad delivery formats.
This is particularly important given that
comScore research shows that 61%
of iframed ads are delivered via these
unfriendly iframes. To demonstrate the
value of this patent-pending technology,
100% of the ads served in the vCE Charter
Study were delivered via iframes.
In-View by Campaign & Site
Across all campaigns in the vCE
Charter Study, the average in-view rate
was 69% (See Figure 1). The in-view
rates by campaign, however, showed
significant variation – with a range of
55% to 93%. This indicates that, on
average, 3 out of 10 ads were not
seen and were therefore wasted.
61%
of iframed ads
are delivered
via cross-domain
or unfriendly
iframes.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
93%
85%
84%
80%
76%
75%
74%
73%
71%
69%
68%
66%
65%
59%
58%
58%
57%
55%
Figure 1 Percentage of Ads In-View by Campaign
campaign

10
Figure 2 Percentage of Ads In-View by Site
69%
100%
minimum average maximum
A site-level view across campaigns
revealed even more variation in in-view
rates (See Figure 2). On one site, 100%
of the ads were in-view. For this particular
site, all ads were placed in the center
of the homepage and scrolling was not
required to reach the remainder of the
content. For another site, only 7% of the
delivered ads were in-view, meaning 93%
of all ads served on that site never had the
opportunity to be seen and were therefore
completely wasted.
Since only one out of every 14 ads on
the site had the opportunity to be seen,
if a marketer paid a $1.00 CPM to deliver
advertising on that site, the effective
CPM would have been $14.00. While a
site with a $1.00 CPM may seem like a
bargain, when waste levels on the site are
as high as 93%, it can effectively become
one of the most expensive placements in
a media plan.
7%

11
To better understand in-view rates, the results were analyzed by:
• Placement (premium, standard, etc.)
• Relative Size of Site (overall and within category)
• Content Type (News sites, Sports sites, etc.)
• Ad Size (300x250, 728x90, 160x600)
• Position on the Page (above-the-fold versus below-the-fold)
In-View by Placement
Even within a given site, in-view rates can vary significantly by placement.
A traditional content site, for example, ran several vCE Charter Study campaigns. Across
the various placement locations on this site, the in-view rate varied from 23% to 95%.
The placements appeared to fall into three distinct levels of in-view (See Figure 3).
• The largest number of placements delivered more than 80% of the advertisements
in-view—well above the vCE Charter Study average (69%). Such placements could
be considered high-viewability inventory.
• Approximately one-third of the placements delivered advertisements between 66% to
75% in-view, which indicates they were on-par with the vCE Charter Study average.
• A small number of placements, however, dragged down the site’s average, given
their very low in-view rates. With the use of in-flight optimization (which was not
deployed for the purposes of the vCE Charter Study), these sub-par in-view rates
could have been identified early and removed from the delivery. In addition, these
data suggest an opportunity for this publisher to reconfigure the page layout to
ensure that more advertisements are viewable.
Figure 3 Percent of Ads Delivered In-View for Individual Placements Across a Traditional Content Site
100%
80%
60%
40%
20%
0%
%ofadsin-view
low average high

12
In-View by Relative Size of Site
An important question relating to
viewability is how in-view rates vary based
on the size of a site. To begin to answer
this question, a separate grouping of
average in-view rates was created based
on site size. Using comScore Media
Metrix® rankings within specific content
categories (i.e. Sports sites, News sites,
Food sites, Health sites, etc.) as a proxy
for site size, average in-view rates were
calculated based on Top 50, Top 100, Top
500 and long-tail sites by category, and
the findings were then analyzed. Within
these content categories, in-view levels
decreased as the site rankings decreased.
In fact, the difference in in-view rates
between Top 50 sites versus the long-
tail sites was a full 16-percentage points
(See Figure 4).
This finding suggests that large sites
within a content category do a better job
than smaller sites at ensuring the ads
they deliver to consumers are actually
viewable. Further analysis is needed to
identify exactly why this is the case, but a
few potential options may include the fact
that the quality of the site and the content
within a site is stronger on these more
popular sites.
The difference
in in-view rates
between
Top 50 sites
versus the long-
tail sites in their
category was a
full 16-percentage
points.
In-View by content Type
In-view rates also showed variation by
content type (See Figure 5). For example,
Coupon sites delivered relatively strong
in-view rates (89%), whereas Pet sites
(27%) struggled, delivering slightly
more than a quarter of ads in-view. This
variation across categories might, in part,
reflect the common layouts among sites
in a similar genre.
Figure 4 Percentage of Ads Served In-View within a Given Site Category
80%
75%
70%
65%
60%
%ofadsin-view
77%
74%
70%
61%
top 50
sites
top 100
sites
top 500
sites
long tail
sites
(remaining
501+ sites)

13
% of ads in-view
Figure 5 Percent of Ads Served In-View by Select Content Types
Coupons
directories
entertainment - radio
entertainment - tv
technology - news
weather
teens
politics
department stores
pets
89%
81%
73%
72%
70%
67%
60%
59%
58%
27%
In-View by Ad Size
The most common ad size used in the
vCE Charter Study was the Classic
Leaderboard (728x90), followed by
the Medium Rectangle (300x250),
and then the Wide Skyscraper
(160x600). The Classic Leaderboard
delivered the strongest in-view rates
(74%), but there was significant
variance across all sites with a range
of 7% to 93% using this size. The
Medium Rectangle format (300x250)
delivered 69% of its ads in-view,
and the Wide Skyscraper (160x600)
delivered the lowest portion of ads
in-view (66%).
Although further research is required
to better understand the driving
factors for differing in-view rates
across ad sizes, one potential cause
is the relationship between ad sizes
and their typical placement on a web
page. For example, Wide Skyscraper
ads run vertically along a web page,
making it more difficult for 50% of its
pixels to be in the user’s viewport for
at least one second.
Figure 6 Percent of Ads
Delivered In-View by Ad Size
wide
skyscraper
(160x600)
medium
rectangle
(300x250)
leaderboard
(728x90)
66%
69%
74%

14
In-View by Position on Page
When discussing viewability, there is a
common misperception that ads delivered
‘above-the-fold’ are seen, while ads
delivered ‘below-the-fold’ are not. While
the quality of in-view rates can vary from
‘above-the-fold’ versus ‘below the fold’ ad
delivery, the vCE Charter Study results
help to dispel some of these myths.
Surprisingly, the findings demonstrate that
some ads delivered ‘above-the-fold’ were
not seen because users quickly scrolled
past them before the ad had a chance
to load, and alternatively, many ads
placed ‘below-the-fold’ delivered a high
opportunity to be seen (See Figure 7).
The implications of these findings
are far-reaching, and there are broad
applications for both buyers and sellers
of online media. Publishers, for example,
should monetize all ads on their site
that deliver an opportunity to be seen,
regardless of where the ad is placed on
the site. This might mean that inventory
‘below-the-fold’ can be priced as premium
as long as the publisher can prove it was
viewed. Alternatively, marketers can look
for inventory that is currently identified as
remnant, which still delivers attractive in-
view rates. Much of this inventory resides
in exchanges and can be better optimized
by taking into account its placement-
specific viewability potential.
Figure 7 Percentage of Ads In-View by Location on Page
There’s gold
below the fold.
Marketers and
publishers who
can determine
what is in-view by
page location have
an advantage.

15
In-View & Cost
Finally, comScore explored the relationship
between the cost of the ad and the in-
view rate. Eight of the vCE Charter Study
campaigns provided cost data for use
in the analysis. Some campaigns were
branding-oriented, while others were
direct response. In total, 300 unique ad
placements had accompanying CPM data.
The analysis showed there is virtually no
correlation between the CPM paid for the
ad and whether it was in-view (correlation
coefficient = 0.19). This low correlation
clearly demonstrates that sites with the
ability to garner strong in-view rates are
not being compensated fairly. Without
solid in-view data, current pricing fails to
account for differentials in in-view
rates. Understanding the actual delivery
by both site and placement is critical for
marketers seeking to value media based
on its ability to reach a real user.
Publishers and marketers with
detailed in-view data are better
able to value the placements
that provide true value and
price them accordingly.
Figure 8 Correlation of In-View Rates & CPM
R = 0.03729
-
50
100
150
200
250
300
0% 20% 40% 60% 80% 100% 120%
CPMIndex
% of Ads In-View
300
250
200
150
100
50
0
0% 20% 40% 60% 80% 100%
cpmindex
% of ads in-view
R2
= 0.0373

16
Marketers invest in digital with the goal of
buying ads that are more successful than
traditional media at reaching a desired
audience. Unfortunately, the extent to which
an ad reaches its target can vary greatly
based on many factors. The comScore vCE
Charter Study evaluated audience delivery
in two separate, but important, ways:
Traditional Demographics
Delivery of ad impressions to traditional
demographic targets, including age,
gender, household income and the
presence of children in the home.
Behavioral Segments
Delivery of ad impressions to behavioral
segments based on observed online
behaviors (i.e. food enthusiasts,
sports fans, etc.).
Validating ad delivery based on traditional
demographics is the most common
approach. However, understanding how
well an ad reached a relevant behavioral
target is potentially more valuable, since
it offers perspective on not just who
the person is but on what they are
interested in, especially as it relates
to the advertised product.
To evaluate the accuracy of ad delivery, vCE
Charter Study participants identified their
target audiences for each campaign, which
could include one or any combination of the
traditional demographic attributes as well as
behavioral segments. Behavioral segments
are comprised of the heaviest consumers
(top 50%) of topic-specific Web content
(i.e. sports, food, cars, personal electronics
or travel). vCE Charter Study participants
identified a primary behavioral attribute from
80 different online behavioral profiles.
Figure 9 illustrates the most popular desired
attributes across all campaigns in the vCE
Charter Study. The majority of campaigns
included age in their target set, which
is not surprising given its wide use as a
desired attribute across all forms of media.
Interestingly, however, the ability to reach
a behavioral attribute was the next most
common approach, demonstrating the
growing importance of some marketers’
desire to reach people based on more than
demographics. It should be noted that the
use of behavioral campaign reporting is
relatively unique to digital and certainly a
compelling value proposition for marketers
trying to connect more closely with
consumers who exhibit interests that
are aligned with and/or related to the
advertised brand.
Audience
Defining Target Audience
age
94%
behavioral
segment
50%
gender
44%
household
income
6%
presence
of child
6%
Figure 9 Percent of Charter Campaigns
Using Desired Attribute(s)

17
Audience Targeting by Traditional Demographics
Across all vCE Charter Study campaigns, there was quite a bit of variance in their
ability to reach the desired target audience. As one might imagine, the more complex
the target (i.e. the more demographic targeting variables included in the target set),
the more difficult it was for the campaign to deliver on its promise (See Figure 10).
Campaigns with a target audience that included one demographic variable
(e.g. 25-54 years old) delivered impressions to the target an average of 70% of
the time. In cases where there were two variables (e.g. women + 25-54 years old),
the accuracy of targeting decreased to an average of 48%, and with three variables
(e.g. women + 25-54 years old + with children under 18 in the home) the average
was 11%.
Figure 10 Percent of Ads in Demographic Target Based on Number of Targeting Variables*
*Demographic variables can include: age, gender, household income and/or number of children in the household. Due to
sample size, a meaningful range could not be calculated for campaigns with 3 demographic variables.
100%
80%
60%
40%
20%
0%
%ofAdsDelivered
toDemographicTarget
1 variable
70%
2 variables
48%
3 variables
11%
LOW average high

18
Audience Targeting by
Behavioral Attributes
In addition to looking at the audiences
in terms of their demographics, online
behaviors of people who were exposed
to the campaign were also measured.
The campaigns were measured against
their desired behavioral attribute at the
campaign level. In some cases, specific
cookie-based behavioral targeting
was used in several placements in the
campaign. In other cases, marketers
wanted to reach their desired behavioral
audience through traditional media
placements, such as delivering an ad
alongside content of interest to their
audience. Across all campaigns, the
average campaign reached its behavioral
audience target 36% of the time, with a
wide range from 23% to 67% (See
Figure 11).
One obvious conclusion from this finding
may be that online behavioral targeting
has limitations as an accurate or effective
means of reaching audiences online.
However, if executed correctly, behavioral
targeting can be a very powerful, efficient
and effective means of delivering a
brand message to a valuable audience.
One primary reason for these limitations
includes the cookie-based nature of
behavioral segmentation. For example,
while a user may have visited a travel
site that shared its information with data
providers on the basis of the cookie for
that browser/machine combination, there
is no guarantee that when that cookie is
observed later at some other site, that it
represents the same user. Another reason
relates to the freshness of the information.
Someone may have visited a travel site
six weeks ago, but they are no longer
active in travel research. Finally, one visit
alone may not be sufficient to identify a
serious travel intender. As a result, one
must be careful about the accuracy of the
targets they purchase, which is precisely
why audience validation and in-flight
optimization should be a critical part of
the campaign management process. If
these campaigns were to have leveraged
in-flight optimization (which they explicitly
didn’t for the purposes of the research),
it is likely that these numbers would be
dramatically higher.
Figure 11 Percent of Ads Delivered to the Primary Behavioral Attribute by Campaign
23%
LOW
36%
average
67%
HIGH

19
It is also important to note that, in some
campaigns, the behavioral attribute
target actually did a much better job at
delivering on-target impressions than
the demographic group, suggesting that
using demographics alone to evaluate
the success of campaign delivery is not
sufficient. For example, one campaign for
a CPG-product that had a demographic
target of women between the ages of
25-54 years old, only served 37% of
impressions to that group. However 67%
of the impressions went to people who
were heavy users of food and cooking
content online. With demographic-based
evaluation alone, this campaign delivery
would have appeared unsuccessful.
A separate analysis of an Automotive
campaign in the vCE Charter Study helps
to shed light on the value of behavioral
campaign reporting and its ability to reveal
a deeper portrait of the type of consumer
exposed to the campaign. The analysis
involved creating an index of visitation
to online site categories for consumers
exposed to the ad campaign compared
to the average Internet population. The
findings revealed that the exposed
group over-indexed (158) on automotive
content, meaning the audience reached
by the campaign was 58% more likely
than the average Internet user to be a
significant consumer of online automotive
content (See Figure 12). This is a positive
indication that the campaign reached
the right audience regardless of the
demographic composition.
Another important finding was that the
audience reached in this campaign also
over-indexed significantly in categories
relating to Financial Products (382)
and Family and Parenting (266). This
information can be used to develop
creative messaging that speaks to
the interests of the audience, such as
showcasing a family vehicle or financing
information in ads.
Again, it is important to note that for
the purposes of the vCE Charter Study,
these campaigns were not optimized
in-flight, meaning that no corrective action
was taken throughout the course of the
campaign to improve the extent to which
these ads were able to reach their target
audience. With in-flight optimization,
it is highly likely that all campaigns would
have seen improved on-target delivery
rates for both their demographic and
behavioral targets.
Figure 12 Index of Online Behavioral Activity by Category for Consumers Exposed to an Automotive Campaign
Using
demographics
alone to evaluate
campaign delivery
may not
be sufficient.
biz - financial products
family and parenting
auctions
gaming - online
gambling
automotive
sports and recreation
teens
shopping - fashion
shopping - toys
jobs - seekers
personals
382
266
262
249
227
158
141
109
45
35
7
3
index
0 100 200 300 400

20
Audience Targeting & Cost
Using available CPM data (as outlined
in the prior In-View section), the
correlation between CPMs and the
accuracy of demographic targeting
(primary audience only) was analyzed
as part of this research. The findings
revealed a very small correlation
(correlation coefficient = 0.18),
suggesting that there is little or no
relationship between the amount paid
for an ad and its ability to reach the
desired demographic target audience.
Before drawing macro conclusions about
this finding, it is important to examine
some of the potential reasons for this
lack of correlation between these two
variables. First, some marketers might
simply not be building campaigns with
the core objective of reaching a specific
demographic, and instead they are buying
media based on its ability to hit certain
behavioral segments.
Another very real issue is the accuracy of
cookie-based targeting data. As noted
above, there are a myriad of companies
that provide this data, and there is
very large variation in the quality of the
data. Unless cookie-based audiences
are verified against a credible third-
party source, it is possible that they are
missing the mark. In the vCE Charter
Study, demographically-cookie targeted
ad placements reach their desired
demographic 14% to 96% of the time.
This indicates a wide variation on the
quality of demographic cookie data.
Regardless of the cause, it is clear that,
at present, the market is not rewarding
ads that deliver to the intended audience
compared to those that did not. This
represents an opportunity for both
advertisers and publishers, especially
now that they have transparency into the
accuracy of delivery and the ability to
optimize in-flight to avoid waste.
Figure 13 Correlation of % of Ads Delivered to Primary Demographic Target & CPM
Unless cookie-
based audiences
are verified against
a credible,
third-party source,
it is possible that
they are missing
the mark.
0
50
100
150
200
250
300
0% 20% 40% 60% 80% 100% 120%
IndexofCPM
% of Ads Delivered to the Primary Demographic Target
300
250
200
150
100
50
0
0% 20% 40% 60% 80% 100%
indexofcpm
% of Ads Delivered to the Primary Demographic Target
R2
= 0.0308

21
india philippines australia germany s.korea dom.rep. indonesia brazil franceU.K.
Figure 14 Percent of Ads Delivered to Geographic Market Among All Impressions Delivered Outside of North America
%ofimpressions
1%
0%
2%
3%
4%
5%
6%
7%
8%
9%
Defining Geographic Target
When delivering ads on television, it’s
relatively easy to ensure they run in their
desired geographic market, because
broadcast markets have defined geographic
borders. The Internet, on the other hand, is
borderless and users can access specific
content from anywhere in the world. As a
result, controlling geographic distribution of
advertising can be challenging. For marketers
trying to maximize every dollar of their
advertising budget, it is critical that their ads
are delivered in the desired market where
their products are actually sold. Accordingly,
geographic validation was an important
component of the vCE Charter Study.
Geographic Targeting:
Overall & By Campaign
All campaigns in the vCE Charter Study
had a geographic target of the U.S.,
and in total, about 4% of impressions
were delivered outside of the U.S.
Of impressions delivered outside of
the target, nearly half were served in
Canada, and the remainder spread
across Europe, The Caribbean,
Asia-Pacific and Latin America. This
finding suggests that a good portion of
the wasted impressions were delivered
to people living in countries whose
native language is something other
than English.
Geography

22
When examining the results on a
campaign-by-campaign basis, it is
interesting to note the large range of
impressions delivered outside of the
target geography. While one campaign
performed flawlessly, another wasted
about 15% of its impressions (See Figure
15). Given that the Internet provides
a wealth of geo-location information,
and given the campaigns’ broad target
of ‘inside the U.S.’, this large range is
somewhat surprising.
In such cases, the inability for an ad to
be delivered in its intended geography
is often not a result of poor targeting
capabilities, but rather due to error
in complex ad buying and selling
processes. Delivery of ads outside a given
geographic target often occurs for two
primary reasons:
The first reasonis simple
communication error. In some cases, the site
serving the ad is not aware of the intended
geographic target. This occurs when the
requirement does not appear on the insertion
order (IO), which authorizes the purchase of
impressions from the site and determines the
characteristics of the ads to be served. Such
misfires can be easily remedied by ensuring
geographic requirements are a standard part
of IO contract templates.
The second reason is due to human
error. To target an ad to a given geography,
the requirement must be programmed
in the ad server that is delivering the ad.
Occasionally this step is missed by the
publisher, or in rare cases, the wrong
geography is inadvertently selected.
Fortunately, there are easy ways to combat
these issues. As long as geography is
specified in the IO, performance can be
optimized in-flight in two different ways.
The first is through real-time alerting, which
notifies sites when ads are being served
outside the desired geographic region so
that corrective action can be taken. The
second option is to use an ad blocking
technology, which can be implemented
to prevent ads from being served outside
the geographic target altogether. This is
generally reserved for instances where
serving an ad outside a specified geography
may create privacy or legal concerns, and in
lieu of in-flight course correction, absolute
prevention must be employed. These alert
and blocking features can protect both
marketers and publishers from wasting
inventory and from lowering the overall
effect of a campaign. Although neither
alerting nor blocking was used in the vCE
Charter Study, both of these features are
part of the comScore vCE offering.
Figure 15 Percent of Ads Delivered In Geography by Campaign
The inability for an
ad to be delivered
in its intended
geography is
often not a result
of poor targeting
capabilities, but
rather due to error
in complex ad
buying and selling
processes.
75% 80% 85% 90% 95% 100%
% of Ads in geographic Target
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
campaign

23
A major concern of all marketers is the
relevance of the content in which their
ads are delivered. When brands spend
money on advertising, they need assurance
that their ads will not run next to content
that is at odds with the brand they are
trying to build or the equity they have
already established.
In this context, ‘objectionable content’ can
generally be categorized into two buckets,
the first being rather objective and the
second being much more subjective
and brand-specific:
Type I:
Adult-Content
and/or Hate Sites
Almost all brands want to avoid having their
ads run on Adult-Content or Hate sites.
Although there might be some differences
of opinion on exactly what sites fall into
these categories, there are generally agreed
upon and industry-endorsed lists that
define these, and almost without exception,
reputable marketers want to avoid them
at all costs.
Type II:
Brand-Specific Criteria
There are topics, issues and/or content
that certain brands don’t want to advertise
near because it directly conflicts with
and/or detracts from the advertising’s
objective. For example, consider a major
airline. For obvious reasons, an advertiser
in this space might not want the brand’s
ad to appear next to an article about
significant plane delays. Meanwhile, for
countless other advertisers, delivering
an ad to a consumer in this content
would be completely benign.
Concerns relating to both of these
categories are very legitimate.
Unfortunately, though, due to the complex
chain of online ad delivery through ad
networks and exchanges, it is not always
clear where an ad will appear.
Brand Safety on Adult-Content
and Hate Sites
To begin to understand the extent to
which ads are delivered in content
deemed inappropriate, the vCE Charter
Study quantified the incidence of ad
delivery on Adult-Content and/or Hate
sites (Type I). The study used a standard
definition of ‘objectionable content’, based
on historical data of sites/categories most
commonly identified as being ‘not brand
safe’ by leading advertisers (See Figure
16). The measurement was applied to
all campaigns.
Figure 16 Categories Deemed “Not Brand Safe”
for Purposes of vCE Charter Study
• Piracy and
Copyright Theft
• Anonymizer
• Child Abuse
Images
• Criminal Skills
• Hacking
• Illegal Drugs
• Marijuana
• Spam URLs
• Botnet
• Command
Control Centers
• Comprised and
Links to Malware
• Malware
Call-Home
• Malware
Distribution Point
• Phishing/Fraud
• Spyware and
Questionable
Software
• Peer-to-Peer
• Torrent
Repository
• Hate Speech
• Pay to Surf
• Nudity
• Pornography
• Sex and Erotic
• Content Server
• Private IP
Address
• Redirect
Brand Safety
Defining Brand Safety
Due to the
complex chain
of online ad
delivery through
ad networks and
exchanges, it is
not always clear
where an ad will
appear.

24
To the surprise of many advertisers in the
vCE Charter Study, 72% of the campaigns
had at least some impressions served in
this type of inappropriate content, which
spanned a total of 980 sites (See Figure
17). The good news is that the actual
percentage of impressions involved was
quite small, less than .01%. However, the
study also showed that 92,000 people saw
these ads, meaning that if some of these
people were either loyal or prospective
customers, it could be counter-productive
and/or problematic for the brand.
It should be noted that it is likely that this
number is much higher when evaluating
the broader, online advertising universe
as there are certain factors that may have
positively influenced the low percentage
of inappropriate ad placements in the vCE
Charter Study. These factors include:
• The brands under measurement were
premium national marketers and therefore
more likely to use higher quality content
• Many of the brands were already employing
ad blocking technologies from external
third-parties. Even with these technologies
in place, several instances of inappropriate
placements still appeared.
• In a few instances, select demand-side
platforms chose to obfuscate the URLs
where the ads were run, meaning that
brand safety could not be measured and
clients could not validate where the ads
were run.
Despite the relatively low overall incidence
of ads appearing next to inappropriate
content, these findings still might be
unsettling to advertisers. Even one
ad impression delivered in the wrong
environment can damage a valuable
consumer’s feelings toward a brand.
With the increasing use of social media,
a snapshot of a marketer’s ad in an
inappropriate environment can quickly go
viral, exposing many more people to the
unintended, but negative, association of
a brand and inappropriate content. With
92,000 people being exposed across
all vCE Charter Study campaigns, the
advertisers’ concerns are justified.
The daily alerts and blocking technology
discussed in the geography section of the
report can also be deployed for Type 1 and/
or Type 2 content sites. Real-time alerts can
be set to notify publishers, marketers and/
or agencies if an ad is appearing in content
deemed ‘not brand safe.’ In addition, the
technology can completely block the ad
from being served in certain environments.
The definition of what is brand safe can be
customized by the brand.
Figure 17 Percent of vCE Charter Study Campaigns with Impressions Delivered Next to Content Deemed “Not Brand Safe”
Some Ads in InapproPRiate content
no ads in InapproPRiate content
28%
72%
72%
of the campaigns
had at least some
impressions
served in
inappropriate
content, which
spanned a total of
980 sites.
92,000 people
exposed tO
Adult-Content
and/or Hate Sites

25
Defining Fraud
Today’s world of online advertising
involves many players in the ecosystem,
each with a specific role and goal.
However, the inherent complexity in this
landscape results in a lack of control and
visibility into online ad delivery. While the
vast majority of individuals in the digital
advertising ecosystem operate with the
best of intentions, like any industry, there
are fraudulent players that can disrupt
the value chain. The complicated daisy
chain of ad delivery can involve up to 20
different players, and quite often neither
the buyer nor the seller has insight into
each step in the process.
The term ‘fraud’ as it relates to online
advertising encompasses a variety of
impression-delivery scenarios. In some
cases, there is direct fraud, which is
deliberate and completely illegitimate,
while other types of fraud are an
unintentional by-product of legitimate
business practices. In either case, this
fraudulent activity does not deliver ads
to actual people as intended, so should
therefore be excluded from validated
impression counts.
specifically measured two
aspects of inappropriate
delivery:
1. The incidence of ad delivery
via non-human spiders and
bots identified by the IAB
2. The incidence of ad
delivery on sites with
clear illegitimate and
intentional fraud
Fraud
The complicated
daisy chain of
ad delivery can
involve up to 20
different players,
and quite often
neither the buyer
nor the seller
has insight into
each step in
the process.

26
List of Non-Human Spiders & Bots Identified by the IAB
To help members of the online advertising ecosystem better understand and avoid
issues relating to fraud, the IAB maintains a list of all known non-human spiders and
bots. All IAB-accredited ad servers are required to filter out these known sources
of non-human ad impressions. The use of some of the spiders and bots on this list
is a completely legitimate practice employed by many websites for a variety of uses,
such as to gather data to help index pages for search engines or to determine page
content for the purposes of offering contextual ad placements. Regardless of their use,
however, they do not deliver ads to consumers and can therefore wreak havoc on ad
delivery and validation, causing a lot of wasted ad impressions and skewing the results
of advertising effectiveness measurement. An analysis of vCE Charter Study campaigns
showed that the average campaign in the study had 0.16% of total impressions being
delivered via these spiders and bots, with a range of 0.03% to 0.49% (See Figure 18).
Figure 18 Percent of Total Impressions Delivered Via Non-Human Spiders and Bots as per IAB List
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
0.20%
0.26%
0.27%
0.06%
0.21%
0.09%
0.07%
0.12%
0.17%
0.10%
0.10%
0.09%
0.03%
0.49%
0.04%
0.04%
0.40%
0.10%
campaign

27
Sites with Intentionally Fraudulent & Illegitimate Activity
In addition to known spiders and bots, part of the vCE Charter Study analysis included
an evaluation of fraudulent impressions that were intentionally delivered via illegitimate
online activity. Campaign delivery was manually reviewed for unusual activity indicative of
intentional fraud. Such indicators might be unusually high or unusually low in-view rates
or little or excessive mouse movement on the creative. Upon identifying these outliers,
further human investigation was used to either confirm or negate the hypothesis.
The analysis revealed more than 200 sites that were guilty of this type of fraudulent
delivery. Figure 19 below outlines some of the most common categories of sites with
such activity. Additionally, the investigation uncovered that one of the sites delivered
almost two million ads in the vCE Charter Study, supporting the need for consistent
hygiene on campaigns to accurately measure delivery and ensure only ads that are
delivered to actual humans are counted in validation and effectiveness measurement.
Again these ads were not blocked from serving for the purposes of this study but
instances of delivery were measured.
PhisHing/Fraud
Redirect
miscellaneous
pay to surf
peer-to-peer
criminal skills
hacking
compromised & Links to malware
malware distribution point
no content found
unreachable
spam urls
anonymizer
torrent repository
piracy & copyright theft
Figure 19 Custom Categorization of Sites with Intentionally Fraudulent Activity
0 5 10 15 20 25 30 35 40
Number of Websites WITHIN EACH CATEGORY
While these two categories of fraudulent ad delivery accounted for only a small
percentage of total ad impressions in the vCE Charter Study, there are a variety of
other sources of fraud that consistently result in significant waste. For perspective,
of the approximately 1 trillion URLs that comScore processes each month
(40% more than all the traffic of the entire U.S. Internet population),
the application of comScore’s full suite of fraud detection
technologies identified levels of fraud ranging from 3%
to 10% for a given campaign. Clearly, no brand is immune from fraud,
and it should be an area of concern for all players in the ecosystem.
No brand is
immune from
fraud, and it
should be an
area of concern
for all players in
the ecosystem.

28
The vCE Charter Study demonstrates that each dimension of ad
delivery – viewability, audience targeting, geographic targeting,
brand safety and fraud – has a significant impact on whether or
not an ad has an opportunity to achieve its intended objective,
and should therefore be a central component of ad delivery
validation measurement.
Advertisers want to understand ad
delivery to each of these core dimensions,
and they also require a holistic, un-
duplicated view of total campaign delivery.
In order to achieve this un-duplicated
accounting of delivered impressions,
advertisers require a simple solution
that eliminates all of the wasted time and
error associated with merging disparate
data sources. Consider, for example,
results from a single campaign in the
vCE Charter Study.
IN ONE CAMPAIGN WHEN
MEASURED INDIVIDUALLY, THE
FINDINGS SHOWED THAT:
38.9% of the ads were
delivered to the right
target audience
58.0% of the ads
were delivered in-view
85.7% of the ads
were delivered in
the right geography
Because of duplication across these
three dimensions, one cannot simply sum
the percentages, as this would suggest
that 155.9% of the ads were delivered
according to plan or that 118.4% of the
ad impressions didn’t deliver well. Instead,
through the use of a single ad tag and a
single measurement solution, vCE is able
to validate that a combined total of
33% of the ads were delivered
according to plan (See Figure 20).
Implications: Putting
all of the Pieces Together
in geography
84.7%
valid
Target
33%in-view
58.0%
on
target
38.9%
Figure 20 Intersection of Percent of In-View,
In Geography and On Target Ad Impressions
Delivered For a Sample Campaign in vCE
Charter Study

29
Prior to the introduction of vCE, the
technology to validate all campaign
impressions against core criteria was not
fully available. The vCE Charter Study
demonstrates that the technology now
exists to identify and correct the source
of sub-optimal performance, and that the
opportunity to do so is substantial. In fact, in a
perfect world, advertisers and publishers can
contract and pay on the basis of impressions
that were served for the campaign, but have
also fully met the validity criteria.
vGRP: A Truly Cross-Media
Comparable Metric
In order for marketers to plan, measure
and evaluate media across channels,
they require digital campaign delivery
measurement that can be translated into
traditional metrics, like reach, frequency
and gross rating points (GRPs). A central
component of vCE is the validated GRP,
or vGRP. The vGRP provides the industry
with a cross-media comparable GRP metric
that is also meaningful in the context of how
online advertising works.
vGRPs are calculated by removing all
ad impressions that did not have the
opportunity to make an impact, including
those that were not in-view, delivered to the
wrong geography, served near brand unsafe
content and subject to fraud. Similarly,
validated target rating points, or vTRPs,
include an overlay of audience-validated
data, providing yet another actionable metric
for marketers seeking to plan campaigns
across channels.
The example below of a CPG brand helps
to illustrate how vGRPs can impact the
true reach and frequency of a campaign
(See Figure 21). In this example, using
non-validated impressions, the campaign
appears to have delivered 46.7 GRPs.
When using validated impressions,
however, the campaign delivered 20.7
vGRPs, yielding a vRatio of 44%. This
delta between GRPs and vGRPs in digital
media demonstrates the volume of waste
occurring, and highlights significant areas
for improvement.
IMPLICATIONS
Putting all of the
Pieces Together
Figure 21 Gross and validated GRP for a Sample Campaign in vCE Charter Study
GROSS VALIDATED RATIO
Reach 8.7 4.9 56%
Frequency 5.4 4.2 79%
GRP 46.7 20.7 44%
TRP 61.4 24.5 40%

30
While the vCE Charter Study sheds light across every aspect
of delivery, three consistent themes emerged in the findings.
1 Marketers are not necessarily getting what they expect
when they buy online ads. From ads delivered next to objectionable content to
ads that never had the opportunity to be seen, there are countless examples where the
digital medium is simply not delivering on its promise.
2 The way online advertising is delivered varies significantly by
site, placement and even creative. Across all dimensions of ad delivery, the
vCE Charter Study demonstrated clear examples of situations where ad impressions
were largely wasted. These findings suggest that measuring all dimensions of ad
delivery for every placement in a holistic fashion is critically important.
3 Regardless of the quality of the buy, there is almost always
room for improvement. Advertisers who understand and leverage the power of
validation stand to gain much more value from the digital channel.
The digital medium has advanced the discipline of advertising in many respects, but it
has also introduced significant complexity to the media equation. To maximize the value
of this important medium, it is important to have the tools to ensure the industry regains
its footing on some of the aforementioned pitfalls and continues to advance forward.
The vCE Charter Study has illuminated many of the ways value is currently being left on
the table. Now is the time for advertisers, publishers and other industry stakeholders to
realize that value.
Conclusion: vCE Charter
Study Key ThemesvCE CHARTER
STUDY KEY
THEMES

31
comScore, Inc. (NASDAQ: SCOR) is a global leader
in measuring the digital world and preferred source of
digital business analytics. comScore helps its clients
better understand, leverage and profit from the rapidly
evolving digital marketing landscape by providing data,
analytics and on-demand software solutions for the
measurement of online ads and audiences, media
planning, website analytics, advertising effectiveness,
copy-testing, social media, search, video, mobile,
cross-media, e-commerce, and a broad variety of
emerging forms of digital consumer behavior.
comScore services, which now include the product suites
of recent acquisitions AdXpose, Nedstat, Nexius XPLORE,
ARSGroup and Certifica, are used by more than 1,800
clients around the world, including global leaders such
as AOL, Baidu, BBC, Best Buy, Carat, Deutsche Bank,
ESPN, France Telecom, Financial Times, Fox, Microsoft,
MediaCorp, Nestle, Starcom, Terra Networks, Universal
McCann, Verizon Services Group, ViaMichelin and Yahoo!.
For more information, please visit www.comscore.com
About comScore
Stay Connected
Follow @comscore
FOR FURTHER INFORMATION,
PLEASE CONTACT:
Andrea Vollman
comScore, Inc.
+1 212 497 1731
press@comscore.com

digital marketing -advertising

More Related Content

What's hot (20)

Viewers also liked (19)

Similar to digital marketing -advertising (20)

More from Indranil Bhattacharjee (20)

Recently uploaded (20)

digital marketing -advertising