Numbers,
Please
In the Digital World, Not Everything
That Can Be Measured Matters
How to Distinguish “Valuable” from “Nice to Know”
Among Measures of Consumer Engagement
Gian M. Fulgoni
INTRODUCTION
… along with harder measures of sales lift—often
comScore, Inc.
In 2006, the Advertising Research Foundation
are used to measure consumers’ engagement with
gfulgoni@comscore.com
(ARF) provided one of the first marketing-focused
digital advertisements. And they all still have
definitions of “engagement” as “turning on a pros-
value.
pect to a brand idea enhanced by the surrounding
But, the computerized nature of digital also
context.” The phrase “turning on” would appear to
means that many other engagement metrics—from
encompass any communication from a brand that
clicks, to viewability, to “Likes” and “Shares”—are
improves consumers’ attitudes and sentiments or
available today. The challenge for marketers is to
that induces a positive change in consumer choice
identify the metrics that matter to their return on
in favor of the brand.
advertising investment (ROI) versus those that
1
Some criticized the ARF interpretation as being
either are nice to know or downright misleading.
too broad for such an important indicator, but it has
since morphed into multiple definitions, depending
DIGITAL-ADVERTISING METRICS
on the medium of context. Many are the result of
Why the “Click” Matters Less
the impact of the Internet, social media, and mobile
One of the first measures of online engagement was
devices on the nature and extent of communica-
the “click” on a display advertisement, with the use
tions between a brand and its consumers.
of a computer mouse. Although click rates in the
2
Engagement no longer can be considered to be
early days of online advertising reached levels of
simply a one-way communication from a brand to
3 percent or higher—and it’s true that clicks remain
the consumer. Instead, it now needs to incorporate
meaningfully high for search advertisements—
consumers’ ability to easily provide digital feed-
average click-through rates for display advertise-
back of their own—at scale and with the communi-
ments today have dropped to an abysmally low
cation being either positive or negative.3
level of 0.1 percent. Or, in other words, only one in
From desktop/laptop computers to smartphones
to tablets to game consoles, there are myriad ways
a thousand advertising impressions in a campaign
generate a click.
for a brand to digitally engage with consumers.
Further complicating the click credibility is that
With each tool comes a different set of means to
research has demonstrated the absence of any rela-
measure that engagement. The traditional “softer”
tionship between clicks and effectiveness (Fulgoni
measures of advertising engagement—attitudinal
and Morn, 2009).
shifts in brand recall, likability, and purchase intent
Despite the greater understanding we now have
of the metric, many in the advertising industry still
“The great brand engagement myth.” (2012, January 30).
Retrieved November 13, 2015, from Marketing website: http://www.
marketingmagazine.co.uk/article/1113464/great-brand-engagementmyth
2
“IAB digital ad engagement whitepaper: An industry overview and
reconceptualization.” (2013, January 22). Retrieved November 13,
2015, from Internet Advertising Bureau website: http://www.iab.net/
adengagement
3
“How Brands Define Engagement.” (2012, August 29). Retrieved
November 13, 2015, from Digiday website: http://digiday.com/brands/
how-brands-define-engagement/
1
DOI: 10.2501/JAR-2016-008
use click-through rates as measures of engagement
and effectiveness. In a 2014 survey4 of publishers,
agencies, and advertisers, approximately four out
of ten professionals said they used the click “always
or most of the time” to measure the effectiveness
Proprietary survey conducted by comScore in September 2014 for
inclusion in a presentation at the IAB U.K. Engage 2014 Conference.
4
March 2016 JOURNAL
OF ADVERTISING RESEARCH 9
In the Digital World, Not Everything That Can Be Measured Matters
Only one in a thousand advertising impressions
in-view to consumers—also has prompted
in a campaign generates a click. Further
advertisements are actually in-view.
complicating the click credibility is that
this concern has led to the suggestion that
research has demonstrated the absence of any
The industry’s accreditation group, the
relationship between clicks and effectiveness.
fied that as long as 50 percent of a display
the measurement of how long in-view
Sometimes referred to as “attention,”
this is another valid engagement metric.
Media Ratings Council (MRC), has speciadvertisement is in-view for at least one
second (two seconds for a video advertisement) then the advertisement can be clas-
of display advertisements. It’s likely that
extremely important metric for driving
simplicity, low cost, and speed are the driv-
consumer response to advertising.
ers of the continued use of clicks—which
sified as being in-view.
Some marketers and their agencies have
criticized this definition as being far too
is unfortunate in light of their lack of rel-
How Viewable Is Your Ad?
lax.7 A Millward Brown study showed that
evance to advertising effectiveness.
From a negative perspective, the very
the impact of digital advertisements climbs
Buyers and sellers of digital advertising
nature of digital technology results in some
sharply after the advertisement has been
also use other engagement metrics. These
unique impediments to any consumer
in-view for some time,8 suggesting that a
include
engagement with digital advertisements,
longer in-view definition might be more
including low viewability and ad-blocking
appropriate.
• website reach
Further complicating the situation is that
software.
Needless to say, if the advertisement
in the case of video advertisements, meas-
• time spent
isn’t even in-view to the consumer, then
urement of consumer engagement encom-
• demographics.
engagement can’t occur. Digital in-view
passes not only whether the advertisement
• number of page views
rates (i.e., the percentage of advertising
was ever seen and how long it was viewed
Although these certainly are relevant to the
impressions in a campaign that are in-
but also whether or not the audio was
size and quality of the audience reached,
view to the consumer) often are found to
switched on.7 Speaking at the Internet
the relationship of these measures to
be lower than 50 percent (mainly because
Advertising Bureau’s (IAB) Annual Lead-
advertising engagement and the effective-
the user doesn’t scroll down the page
ership Meeting in February 2015, CBS
ness of brand advertising carried on the
far enough to see the advertisement or
Interactive Chief Revenue Officer David
site is tenuous at best.
because the “viewer” is a fraudulent com-
Morris commented that the introduction
puter with no human user).
of viewability had been somewhat chaotic
By contrast, GfK researchers have dem-
With that evidence in hand, it’s not
for the industry, both in terms of measur-
surprising to see leading advertisers
ing the MRC’s standard effectively and the
• attitudes toward a website,
such as Unilever and the world’s largest
confusion the concept itself has inflicted on
• motivations for using it, and
media buying agency, GroupM, demand-
the marketplace.7
• overall opinion of the site
ing assurances from publishers that their
onstrated that consumers’
advertisements are actually seen.6
are the most relevant metrics that need to
The realization that the metrics provided
be taken into account in gauging engage-
by ad servers were misleading—and that
ment with brand advertising.5 In par-
many digital advertisements were not
ticular, trust in the site appears to be an
“WPP And Unilever Video Woes Mean Human Viewability Metrics Are A Must.” (2015, September 17). Retrieved
November 13, 2015, from MediaPost website: http://www.
mediapost.com/publications/article/258504/wpp-and-unilever-video-woes-mean-human-viewability.html
6
“Measuring Engagement: A White Paper.” Association of
Online Publishers.
5
10 JOURNAL
OF ADVERTISING RESEARCH March 2016
Marshall, J. “Marketers Push Back on MRC Ad Viewability Standards.” The Wall Street Journal, February
25, 2015. Accessed on November 13, 2015, at http://
blogs.wsj.com/cmo/2015/02/25/are-online-ad-viewabilitystandards-failing/
8
“The Value of a Digital Ad: How Delivery and Brand/
Sales Effectiveness Can Drive Digital Advertising ROI
in a Cross-Media World.” Millward Brown. Accessed on
November 13, 2015, at https://www.millwardbrown.com/
Insights/MBArticles/Value_Digital_Ad/default.aspx
7
In the Digital World, Not Everything That Can Be Measured Matters thearf.org
The Battle Against Ad Blocking
It’s too early to predict how the adver-
the impact of a marketer ’s outbound
As advertisers seek more effective ways to
tisement blocking issue will be resolved,
communications.
engage with consumers and measure that
but because it clearly represents a serious
The other side of the social coin is that
engagement, their efforts are hampered by
obstacle to the ability of advertisers to com-
positive consumer engagement on social
ad-blocking software.9 About one in ten
municate directly with online users—and
media can yield impressive results. Brands
U.S. consumers have ad-blocking software
to be able to measure that engagement—
can use paid or organic communications
installed on their computers; the ratio is
we can expect it to get intense industry
to reach their target segments and then
even higher in some other countries.
attention.
benefit from any engagement (e.g., “shar-
10
Advertisement blocking is especially
ing”) that subsequently occurs and which
popular on mobile devices, where slow
SOCIAL-MEDIA METRICS
load times are particularly annoying to
One of the best examples of how digital
Warc, the London-based publisher of
users and the slow loading of advertise-
technology has changed brand engage-
the Journal of Advertising Research, awards
ments can be considered to increase the
ment from one-way to two-way commu-
annual prizes for the best examples of how
cost of consumers’ data plans.
nications—and the related metrics—is the
brands have successfully incorporated
impact of social networks.
social engagement into their marketing
Advertisement blocking is a problem
extends persuasive reach.
across any digital platform, including the
Approximately 50 percent of consum-
strategies (Fulgoni, 2015). Many of these
viewing of video content. 11 Publishers
ers say they use social media to post posi-
successfully leveraged consumer engage-
view its very nature as threatening the
tive or negative brand comments (Sexton,
ment with a wide variety of branded con-
“unspoken agreement” between content
2015)—although criticisms seem to attract
tent that led to sharing and a dramatic
providers and consumers that advertising
the most attention—on a multitude of
increase in reach.
is necessary because it pays for the content
social sites ranging from Facebook to Twit-
Brands have learned that they can maxi-
consumers enjoy for free. Some have sug-
ter to YouTube to Yelp that are viewed by
mize the impact of their social-marketing
gested such solutions as
hundreds of millions of people.
programs by leveraging a framework that
Measurement of the number of postings
helps them move beyond a measurement
• producing more advertisements that
and whether they are positive or negative,
of the simple number of fans/followers
consumers actually want to view while
how many times shared, and how many
or “Like” metrics. These metrics haven’t
simultaneously reducing ad clutter;
users reached, have become important
yet been proven to drive positive business
metrics of social engagement with brands
results, whereas measures of actual shar-
that are initiated by consumers.
ing of content have been shown to deliver
• insisting that a site’s content not be made
available unless the consumer accepts
the advertisements or that sites install
Whether simply an extension of human
measurable marketing ROI.12
nature (or the nature of those most vocal
That said, it’s also apparent that some
on social), social networks are inun-
sharing metrics for organic content (as
dated with negative comments regarding
opposed to the sharing of paid content)
• using more native advertising that isn’t
brands—and the numbers are increasing
are problematic. This occurs because of
delivered by an ad server and therefore
every day. The challenge for any marketer
the difficulty of computing the true num-
can’t be blocked;
is to respond to negative engagement in a
ber of unduplicated people who ever saw
timely manner. Unfortunately, it appears
the content.
software that recognizes and deactivates
advertisement-blocking software;
• taking legal action against the authors of
the software.
“Ad blocking unleashes anxiety across the ad industry.”
Financial Times, October 5, 2015.
10
“The state of ad blocking.” comScore and Sourcepoint.
Accessed on November 13, 2015, at http://sourcepoint.com/
comscore-and-sourcepoint-the-state-of-ad-blocking/
11 “
TV networks confront ad blockers erasing their commercials online.” (2015, August 31). Retrieved November
13, 2015, from Advertising Age website: http://adage.com/
article/media/tv-networks-confront-ad-blocking-erasingcommercials-online/300143/
9
that as many as 70 percent of companies
Simply adding the number of follow-
fail to do just that, thereby risking damag-
ers of the person posting content assumes
ing consequences (Sexton, 2015).
that all saw the original post. In fact, that’s
Organizations long have monitored
frequently not the case, and such simple
changes in brand equity that are driven
math also fails to de-duplicate the same
by their own marketing efforts. But it has
become apparent that monitoring and
responding in real time to inbound comments expressed on social media should
be every bit as important as monitoring
“The Power of Like 2: How Social Marketing
Works.” (2012, June 12). Retrieved November 13,
2015, from comScore website: https://www.comscore.
com/Insights/Presentations-and-Whitepapers/2012/
The-Power-of-Like-2-How-Social-Marketing-Works
12
March 2016 JOURNAL
OF ADVERTISING RESEARCH 11
In the Digital World, Not Everything That Can Be Measured Matters
people across followers. As a result, it’s
very likely that the simple aggregation of
followers will overstate the true number of
people receiving the shared content by a
The winners will be brands that are able to
derive methods to measure and understand
large amount.
consumer demands for mobile engagement
MOBILE-ENGAGEMENT METRICS
and then deliver against those needs.
Arguably the most significant change
in consumer engagement has been the
result of the use of mobile devices. Digital-
Engagement with Mobile Ads
• I want to know.
media time in the United States has surged
Certainly, advertising remains impor-
• I want to go.
recently—growing nearly 50 percent in
tant in a mobile world, and the IAB has
• I want to do.
the past two years, with more than three-
reported that mobile advertising is the fast-
fourths of that growth directly attributable
est growing format in the U.S. (up 73 per-
These micro-moments can occur any-
to the mobile app.13
cent in the fourth quarter of 2014 versus
where—including in-store. Google has
the prior year), accounting for 28 percent
found that consumers often are more
of all digital advertising dollars.14
attentive to their in-the-moment needs
Mobile has grown so fast that it’s now
the leading digital platform, with total
activity on phones and tablets account-
Research also has shown that the impact
than they are loyal to a particular brand
ing for 62 percent of all digital media time
of advertising is surprisingly higher on
or product. At the same time, they’re
spent. It is important to note that apps—
mobile than on desktop. This suggests
attracted to those brands that best address
which help make it easier for consumers
higher consumer engagement because
those in-the-moment needs.
to visit and buy at websites—represent
there is far less advertising clutter on
Immediacy and relevance in an “I-want”
the vast majority of digital mobile time at
mobile, and advertisements can be deliv-
world, therefore, would appear to trump
87 percent.13
ered closer to the actual point of pur-
loyalty nowadays. The winners will be
chase—further increasing relevance and,
brands that are able to derive methods
therefore, engagement and impact.
to measure and understand consumer
As the use of mobile devices has surged,
so have the ways in which consumers can
8
engage with brands. Because of this, it’s
The larger impact of mobile devices,
vital for marketers to ensure they capture
however, goes well beyond advertising.
their “fair share” of consumers’ mobile
They have allowed consumers to initiate
engagement.
engagement with branded content when-
CONCLUSIONS
ever and wherever they choose. There’s a
Marketers have only scratched the sur-
dizzying array of ways in which this can
face of measuring new levels of consumer
happen, which Google describes as “micro-
engagement, given a digital marketplace
• how much mobile time is spent on a
moments.” These unfold through a variety
that remains in flux. For digital advertisers,
marketer’s own brand website versus
of common “I-want” scenarios that help
it’s clear that some of the metrics produced
competitive sites;
people take steps or make decisions such as
by ad servers are problematic. In particu-
Fundamental to this is the need for metrics that measure
15
• how much of that time is spent via the use
of a mobile browser versus mobile app.
demands for mobile engagement and then
deliver against those needs.
lar, the “click” on a display advertisement
• I want to learn.
has been shown to have no value in terms
• I want to buy.
of predicting effectiveness.
Because of the extreme importance of app
By contrast, the traditional met-
engagement, any shortfall in consumer
rics of attitudinal changes or sales lift
downloading and use of a marketer’s app
could be disastrous.
“The 2015 U.S. Mobile App Report.” (2015, September
22). Retrieved November 13, 2015, from comScore website:
http://www.comscore.com/Insights/Presentations-andWhitepapers/2015/The-2015-US-Mobile-App-Report
13
12 JOURNAL
OF ADVERTISING RESEARCH March 2016
“IAB Internet Advertising Revenue Report, 2014 fullyear results.” (2015, April). Retrieved November 13, 2015,
from Internet Advertising Bureau website: http://www.
iab.net/media/file/IAB_Internet_Advertising_Revenue_
FY_2014.pdf
15
“Best practices: 10 ways marketers can compete for micromoments.” (2015, June 3). Retrieved November 13, 2015,
from AdAge website: http://adage.com/article/digitalnext/
practices-cmos-advantage-micro-moments/298855/
14
remain critically important and useful to
marketers.
Digital advertising also is muddled
because of a lack of consensus between
buyers and sellers regarding the basic
definition of an ad impression in terms
In the Digital World, Not Everything That Can Be Measured Matters thearf.org
Mobile devices also allow consumers to engage
so choose and in a plethora of different
with brand content whenever they so choose
with most of these mobile communica-
and in a plethora of different ways.
engagement opportunities, and because of
ways. There is an immediacy associated
tions that provides marketers with many
this, brands need to be able to measure and
address engagement “in the moment.” But
the very nature of this far-reaching engage-
of time in-view. Consensus on the issue
social media also provides smart market-
ment also means that measurement, for the
might simplify matters. But given the abil-
ers with the ability to engage directly with
moment, is a work in progress. ity to measure optimal time in-view of an
targeted consumers in many creative ways
advertisement—and research showing that
that result in amplification of persuasive
impact increases with time in-view—it
messages.
would not be surprising to see advertisers
But to reap the rewards of positive ROI
demand paying only for those advertis-
from social marketing, it’s clear that adver-
ing impressions that are guaranteed to be
tisers need to think beyond simple metrics
in-view for a specific period of time that
such as “Likes” or “Followers” that have
they define. (For more on the topic of view-
not been shown to be correlated with
ability, please see “The Effectiveness of All
improved sales and embrace more pow-
Advertising Impressions vs. Only View-
erful metrics such as “Shares.” However,
able Impressions,” an award-winning case
in computing the total number of people
study by Adobe, on page 109.)
reached by organic shares, it’s important
Meanwhile, massive use of social media
to de-duplicate the counts of followers and
means that marketers no longer can think
recognize that not all “Shares” are ever
of engagement solely as a one-way com-
seen by all followers.
About the author
Gian M. Fulgoni is cofounder and chairman emeritus
of comScore, Inc. Previously he was president/ceo of
Information Resources, Inc. During a 40-year career at
the c-level of corporate management, he has overseen
the development of many innovative technological
methods of measuring consumer behavior and
advertising effectiveness. Fulgoni is a regular
contributor to the Journal of Advertising Research.
REFERENCES
F ulgoni , G. M. “How Brands Using Social
munication from a brand to consum-
Mobile devices represent the most
Media Ignite Marketing and Drive Growth.”
ers. The reality is that social media has
important dislocation in the historical
Journal of Advertising Research 55, 3 (2015):
provided consumers with the ability to
communication flow from brands to con-
232–236.
impact a brand’s performance by provid-
sumers. Not only does mobile usage now
ing public feedback—positive or nega-
account for the majority of time spent
Fulgoni, G. M., and M. P. Morn.“Whither the
tive—at scale.
online but also apps dominate that mobile
Click? How Online Advertising Works” Journal
of Advertising Research 49, 2 (2009): 134–142.
Because of this, metrics that measure
time. And, under such circumstances,
consumers’ comments on social networks
understanding how consumers are using
are important. Advertisers need to respond
apps to interact with brands is vital.
quickly to negative engagement or risk
Mobile devices also allow consumers to
financial damage. The good news is that
engage with brand content whenever they
S exton , D. E. “Managing Brands in a Prickly
Digital World.” Journal of Advertising Research
55, 3 (2015): 237–241.
March 2016 JOURNAL
OF ADVERTISING RESEARCH 13