!" JOURNAL OF ADVERTISING RESEARCH "#$%& '()' *+,- )(.

'/()01234/'4)4(5(4(/'
#$%&'()*%#'$
The key ndings of the current research include
the following:
Facebook is the dominant social-networking site,
with an audience of approximately 160 million
U.S. visitors each month. It accounts for 90 per-
cent of all time spent on social-networking sites.
Branded content on social media can take many
forms, among them content shared directly from
brands, re-shared content from connections,
and social-marketing tools such as Facebooks
Sponsored Stories advertising unit. Although
there has been considerable focus on counting
the incidence of these brand mentions on social
networks (and, to a lesser degree, categorizing it),
the authors found that the reach and frequency
of social-media brand impressions are far more
important than simple counting statistics.
Facebook users spend more than a fourth of their
time on the site consuming and interacting with
the Facebook Newsfeed; this activity represents
4 percent of all time spent online in the United
States. The Newsfeed also is the primary loca-
tion where branded content is consumed. In fact,
users are 40 to 150 times more likely to consume
branded content in the Newsfeed than to visit
the Fan Page itself.
There typically are two potential audiences for
branded content on Facebook:
Fans of brands on Facebook (those who have
explicitly liked a brand) are the easiest to
reach with social-media brand impressions.
Friends of fans typically represent a much
larger set of consumers (34 times larger, on
average, for the top 100 brand pages) and can
receive social-media brand impressions by
way of their friends. In examples from this
study, the reach of branded content among
friends of fans signicantly exceeded the
reach among fans.
When a brand focuses on acquiring and engag-
ing fans, it can benet from a signicant second-
ary effectexposure among friends of fans that
often surpasses reach among fans.
For some brands, Facebook fans may have dif-
ferent aggregate demographic and behavioral
proles than typical brand purchasers, indicat-
ing that social media may require different
approaches to marketing strategy.
The value of a fan can be assessed in three pri-
mary ways:
increasing the depth of engagement and loy-
alty among fans;
generating incremental purchase behavior;
and
leveraging the ability to inuence friends of
fans.
Collectively, this research leads to two important
conclusions:
By focusing on measuring the reach of their
branded content among various constituencies,
brands can more effectively understand and
optimize the way in which they leverage their
presence on Facebook.
Brands can realize signicant untapped benets
by understanding and focusing on reaching the
friends of their fans.
%+, ./0,1 /2 3456,7
!"# %&'()* +,'-. /'() 0(12,(-,3 4'(*
%+1/89+ :/;5<=>?,@5< ?<16,A5B9
C$(&DE 4#.:?C$
%678%6$9
#:;<=7#>?%67=%6$9.%67
F&CGC? ?)((
%678%6$9
@7ABB?%67=%6$9.%67
?#HD &#*G
%678%6$9
7$;%&?%67=%6$9.%67
:DC$ I&)#*G
C#%9D66E
=9#>D?FD.%67

Editors note: The paper that follows examines the
nature of the reach and frequency of branded con-
tent on Facebook. The research largely is based on
May 2011 ndings from comScore Social Essentials
but includes insights from Facebooks internal
analytics platform.
!"#$% '()' JOURNAL OF ADVERTISING RESEARCH !"
*+, -./,0 .1 2345,6
#$%$&#'( *&'+,#-./0
Over the past several years, social media
have emerged to redene the digital
media landscape and, in the process, have
changed the way we think about the dis-
semination of marketing messages.
In particular, Facebook has facilitated
two unique consumer experiences of inter-
est to brand marketers:
The ability for consumers to identify
brands of interest and connect with
them has enabled sharing between
brands and consumers in new ways.
Brands and their consumers can now
create two-way relationships and share
content, news, and feedback.
Social media also have facilitated inno-
vative ways of sharing information
about brands between friends. Whether
consumers are voicing their afnity for
certain brands or their experiences with
products and services, Facebook not
only encourages this type of sharing but
can accelerate its reach and virility.
Brands increasingly are interested in
establishing a social presence and engag-
ing with their fans, helping shape their
customers experiences, and even lever-
aging their voices for greater marketing
impact. Underscoring the importance of
this channel, 100 percent of the Advertising
Age Top 100 Advertisers have established
Facebook pages for their brands.
As this new marketing paradigm
unfolds, there is palpable excitement
about its potential to drive value for busi-
nesses but also anxiety about the dif-
culty of grasping the full promise of this
emerging medium. According to a study
conducted by the Harvard Business Review,
only 12 percent of companies surveyed
indicated that they were effective users of
social media; just 7 percent said they were
able to integrate social media into their
marketing activities.
These problems are all too common
for businesses attempting to succeed in
todays social-media environment.
Part of the challenge is that as social
media have evolved, the research meth-
odologies and key metrics developed to
describe social activity have served to
isolaterather than integrateinto tradi-
tional marketing terms.
For example, most research on branded
content on social platforms has focused on
describing the incidence of these types of
branded content (and, to a lesser degree,
categorizing it as positive or negative)
without regard to the size or composition
of the audience receiving that content.
Simply knowing how many brand men-
tions exist does little to provide insights
for the potential impact of that content. In
fact, it actually can lead brands to focus on
suboptimal strategies because it highlights
only content creatorsat the cost of not
understanding its distribution.
The current research indicates that an
approach that is instead focused on the
audience of branded content on social
platformsnamely, audience reach and
frequencycan unlock a much better
understanding of the true impact (and,
therefore, value) of those impressions.
This approach is more favorable because it
allows brands to focus on marketing activ-
ities that drive scale within audiences of
interest. Additionally, it provides brands
the opportunity to understand the prole
of those audiences reached with various
types of content and to begin to under-
stand the impact of that content exposure.
Important is that an approach focused
on audience reach and frequency also
would serve to integrate brand activ-
ity on social platforms with other parts
of the marketing mix because reach, fre-
quency, and Gross Ratings Point (GRPs)
are standard units of measurement across
most media channels, including television,
online, radio, and print.
1$2(-0-3-,4 &/0 0$56/626-/%
Data in this study are collected and ana-
lyzed using the comScore Social Essentials
product, which is a measurement service
based on comScores proprietary 2-million
person global panel of Internet users from
a home or work computer. (Note: Mobile-
based Internet browsing is not included in
this analysis.)
Opt-in panelists provided comScore
with explicit permission to passively
observe their online behavior, including
site visitation, search activity, advertising
exposure, and online purchases. By meas-
uring online behavior for these opt-in pan-
elists, comScore could observe both public
and nonpublic pages within particular
social networks such as Facebook.
Measurement of panelists exposure to
social-media marketing within social net-
works can be linked to other behaviors
in which they engage around the Web. In
addition, certain select data points and
illustrations in the current article are based
on aggregate, nonpersonally identiable
data sourced from Facebook.
%789:;<1=>9: *?:@> 6AB?=CC97@C
In this article, the authors describe a unit of
branded content as a social-media brand
impression. Although these impres-
sions occur throughout a social network
and may be delivered through a variety
of dissemination channels, there are four
primary vehicles through which these
impressions are delivered:
Page publishing: These unpaid impres-
sions appear on the fan page wall and
may also appear in the Newsfeed of a
fan or a friend of a fan.
Stories about friends: These unpaid
impressions occur when a friend
actively engages with a brand (e.g.,
Nicole indicates she likes Columbia
Sportswear) and become visible either
on a friends wall or in the Newsfeed.
!" JOURNAL OF ADVERTISING RESEARCH "#$%& '()'
*+, -./,0 .1 2345,6
These stories may appear to fans and
friends of fans.
Sponsored stories: These paid impres-
sions are similar to stories about friends,
but they have been actively distributed
more broadly and appear in the right-
hand column to fans and friends of fans.
Advertisements with social: These
branded messages come directly from
the advertisers with a social context on
the unit that appears to friends of fans
(e.g., Nicole Likes Columbia Sports-
wear, appearing at the bottom of the
advertising unit).
Although comScores Social Essentials
reports on all four impression types, the
current research focuses specically on
unpaidor earnedimpressions (See
Figure 1).
#$%&'(() #$+ $+, #-.&+,/0(10#$+/
2&34&+5/
In the current study, the authors discuss
different types of user segments within
Facebook. Although other analyses of
Facebook fans typically have centered on
counts in the total number of accrued fans,
the current study instead measured audi-
ence segments within Facebook based on
their exposure to brand impressions within
the given reporting period (consistent with
the comScore Social Essentials product).
Moreover, the authors point out, not
only are fans exposed to these impressions
but so are their friends. With this under-
standing in mind, the current research ref-
erences the following audience segments:
Fans: This user segment is based on users
who have afrmatively stated their afn-
ity for a particular brand by liking that
brand. Over a given period, a subset of
the total fan base has been exposed to a
social-media brand impression.
Friends of fans: This user segment (aka
friends) is dened as those who are
not explicitly brand fans but have been
exposed to brand impressions through a
friend who is a fan of a particular brand.
Over a given period, a subset of the total
friends of fans is likely to have been
exposed to brand impressions.
Non-fans: This segment encompasses
all users of the social network who have
neither become a fan of a particular
brand nor are connected to a fan who
has taken such a like action.
An example to illustrate the friends-of-fans
multiplier (See Figure 2): Although the fan
base of Microsoft Bing is approximately
1.7 million, the number of friends of fans
in this network is 232 millionmore than
130 times the size of the fan base.
6789: #892 ;8<= #8>=6??< 8 2?>@8A
;87<=B@9C D=E@>A=
Over the past few years, Facebook has
emerged as a powerful social-marketing
and advertising platform. In the United
States alone, Facebook has approximately
160 million visitors each monthabout
three of every four Internet userswith
the majority of them visiting the site on a
daily basis (See Figure 3).
Compared with other leading social net-
works, Facebook today represents the vast
majority of U.S. social networking activity,
accounting for 90 percent of all time spent on
social networking sites in the United States.
1
1
comScore Media Metrix, US, May 2011
Page published update
in a users News Feed
Ad with Social
Sponsored Stories
Story about a Friend
#.3F-& G ,7#89:;< => ,#$?;@ #?@ -#A@ ";@A# 489$;<<A=?<
!""#$%&'( *' +#,-.##/ *% 0%*12# 0$(#-3
Friends of Fans
Fans
1.7MM
232MM
Non-Fans
Source: Facebook, Worldwide Data
#.3F-& " BA?C 1#%;D==E 1#?
F;C8;?G<H
!"#$% '()' JOURNAL OF ADVERTISING RESEARCH !"
*+, -./,0 .1 2345,6
Within Facebook, the largest portion of
users time is spent on the individuals
homepage that features the Newsfeed.
More specically: In May 2011, 27 percent
of engagement on Facebook.com occurred
on the homepage and Newsfeed, followed
by prole viewing (21 percent), photo
viewing (17 percent) and usage of apps
and tools (10 percent;
2
See Figure 4).
Understanding the importance of the
Newsfeed to the Facebook user experience
helps illustrate how (and where) users are
likely to see and digest information on the
site. This implication especially is impor-
tant for brands with a presence on Face-
book, as the Newsfeed offers the greatest
opportunity to reach and engage with cur-
rent and potential customers.
Facebook users demonstrate their afn-
ity for brands through use of the like
button. By liking a brand (also known as
becoming a fan), followers can express
their interest afrmatively in a particu-
lar brand for reasons that may include
self-expression, communicating positive
2
comScore, see above
associations of that brand to others, stay-
ing in communication with that brand,
or receiving deals and promotions. Many
iconic brands have been able to amass sub-
stantial fan followings, with Coca Cola,
Starbucks, and Disney each boasting more
than 20 million fans as of July 2011.
For the current study, the authors have
examined the social-media impact for
three very different brandsStarbucks,
Southwest Airlines, and Microsoft Bing
to understand how their respective mar-
keting messages have reached fans and
friends of fans throughout the network
and the resulting implications of these
social-media brand exposures.
#$%& ()*+,-, .$+&-+& /01$%2)- 3442)% 5+
&6- 7-8%9--,
Despite an increasing willingness to iden-
tify as fans of a brand, it is important to
understand that the core user experience
of consuming (and engaging with) con-
tent in the Newsfeed extends to branded
content.
In other wordsimportantlysocial-
media brand content is primarily con-
sumed within the Newsfeed, as opposed
to on the brands fan pages.
To grasp the signicant difference in
consumption of branded content in the
Newsfeed as opposed to the brand page,
the authors compared the page views on
the fan page to the total number of brand
exposures on Facebook, with the follow-
ing results (See Figure 5):
In May 2011, Starbucks delivered 156
brand impressions throughout Facebook
for every one page view on its fan page.
Southwest delivered 42 brand impressions
for every fan page view; Bing delivered 45.
In each brands case, it managed to reach
millions of people with brand impressions
throughout the site while only attract-
ing tens of thousands of visitors to its fan
pages.
At rst glance, these outsized ratios
seem surprising, but the data actually
are very consistent with how people use
Facebook. Most of their time is spent on
the Newsfeed over individual prole
(or brand) pages, so it makes sense that
the branded content impressions would
reect this reality.
Despite the dominance of consumption
in Newsfeeds relative to brand pages, a
2008
J
u
n
S
e
p
D
e
c
M
a
r
J
u
n
S
e
p
D
e
c
M
a
r
J
u
n
S
e
p
D
e
c
M
a
r
J
u
n
S
e
p
D
e
c
M
a
r
180,000
160,000
140,000
120,000
100,000
80,000
60,000
40,000
20,000
0
Source: comScore Media Metrix, U.S., June 2007May 2011
Facebook.com
MySpace
Twitter.com
2007 2009 2010 2011
:5;2)- " *#789 :8 ;<=< ;8:>?7 @:A:BC#A D(((E BC 1"$7FCCGH
!I=J"$7 "89 *K:BB7#<
Source: comScore Mediabuilder, U.S., May 2011
21%
Profiles 17%
Photos
25%
All Other
27%
Homepage/
Newsfeed
10%
Apps &
Tools
:5;2)- ! =%"#7 CL *:M7 =J78B
C8 1"$7FCCG<$CM FI NC8B78B
=7$B:C8<
!! JOURNAL OF ADVERTISING RESEARCH "#$%& '()'
*+, -./,0 .1 2345,6
minority of fans see content from a brand
in a given week. A Facebook analysis indi-
cates that, on average, 16 percent of fans
are reached by branded content by a brand
that posts 5 of 7 days.
Why do not all fans see a particular piece
of content from a brand? Some fans may
miss content if they are not logged onto
the site during a period when brands are
active. Additionally, the Facebook News-
feed uses an algorithm to rank content
based upon the likely interest to a user.
So, unlike other social-media settings, in
Facebook, only the content determined to
be most relevant to a user is delivered at a
specic time.
Furthermore, the results of the compari-
son of Newsfeeds to brand pages are simi-
lar to the degree of reach between users
and their friends. A given status update
from a user will result in approximately
12-percent reach among their friends,
according to a Facebook analysis.
3
How can fan pages increase the likeli-
hood of reaching their fan base? Publish-
ing content that is highly engaging can
3
Facebook, internal user analysis
increase the likelihood of content appear-
ing in the Newsfeed. Publishing more
often can also increase the rate at which
fans see content. A Facebook analysis sug-
gests that each incremental day of publish-
ing branded content from the Facebook
page increases the reach among fans by
approximately 2.5 percent (See Figure 6).
Of course, brands also must take care to
ensure that the content and frequency of
publishing aligns with the expectations of
their fan base, and does not annoy or dis-
engage fans.
"#$%&'( *+ ",&(- ./0%&'$&1 02% 3*4$,5
0(12,(-, "5 4'(*
Reaching fans with marketing messages
certainly is of interest to brands, but the
current research found that the friends of
fans represent a substantial potential audi-
ence, often well beyond the scale of fans
alone.
A Facebook analysis
4
of the top 100
brand pages suggests that, for every fan,
there are an additional 34 friends of fans
who can be reached (with signicant vari-
ance among brands; See Figure 7). This
multiplier grows even larger when the
analysis extends beyond the top 100 brand
pages (81 on average among the top
1,000 fan pages).
Looking specically at Starbucks,
Southwest, and Bing, the ratio of fans to
4
See above
Source: comScore Social Essentials, U.S., May 2011
156

42
45
0
20
40
60
80
100
120
140
160
180
Starbucks Southwest Bing
"$16#% 7 78#$9:%;<= 7>:8&?@<8= A BCDEF 0#8C> >G 1#%@9>>;
4HI$@<<C>D< G>$ 1#D< #DJ 1$C@DJ< >G 1#D< 8> -#E@ KC@?< >D
B$#DJ 1#D -#E@<L
%
Source: Facebook, Top 100 Brand Pages, Worldwide Data, May 2011
0
5
10
15
20
25
1 Days 2 Days 3 Days 4 Days 5 Days 6 Days 7 Days
"$16#% 8 1#%@9>>; 1#D -#E@<F M:H:N#8CO@ 0@#%& #< # 1:D%8C>D
>G -><8 1$@P:@D%Q RS>L >G T#Q< I@$ /@@;UL
!"#$% '()' JOURNAL OF ADVERTISING RESEARCH !"
*+, -./,0 .1 2345,6
friends of fans varies considerably across
the three brands (See Figure 8). In all three
cases, however, the friends of fans repre-
sent an audience at least an order of mag-
nitude larger than the fan basehundreds
of millions of users globally.
Friends of fans also are an intriguing
audience for marketers. As these consum-
ers have yet to identify themselves as
fans of a given brand, there may be more
upside in messaging to themeither for
brand building or increasing consump-
tion. Brands also can take advantage of
these users social proximity to fans to
deliver stories about those fans interest
and engagement with their brand.
An analysis of the reach delivered via
social-media brand impressions highlights
this ability to extend reach through friends
of fans. Across all three brands analyzed,
the number of friends of fans reached by
these brand impressions exceeded the
reach by fans alone, and by a signicant
amount in the cases of Starbucks and Bing
(See Figure 9).
#$%&'()* ,-)./0$1. %2 3&%4$ 5(2'
61%)7$7 ,-)2$)2
Social media generally are not regarded
in the same way as other media channels
(online display advertising, for instance)
that deliver branded content at scale, such
as online display advertising. The current
research, however, has indicated that, in
some cases, such analysis could warrant
consideration, primarily because of social
medias demonstrated ability to deliver
impressions comparable with that of other
digital-advertising efforts.
With Southwest and Bing, social-media
impressions were marginally incremental
to their existing online display advertising
campaigns in reaching people around the
Web. In Starbucks case, however, unpaid
F
r
i
e
n
d
s

o
f

F
a
n

M
u
l
t
i
p
l
i
e
r
No. of Fans
Source: Facebook, Worldwide Data, Top 100 brand pages
0 10M 20M 30M 40M 50M
0
10
20
30
40
50
60
8(*/1$ 9 1"$7899: 1"; -"<7=> 1#?7;@= 9A 1"; !BCD?EC?7# ";@ 1";
F"=7 G?H7I
Source: Facebook, Worldwide Data
0
100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
Starbucks Southwest Bing
8(*/1$ : GD"#8B$:=J G9BD%K7=DJ ";@ F?;<> -9D7;D?"C 07"$% 9A
1#?7;@= 9A 1";= LB@?7;$7 M(((NI
!" JOURNAL OF ADVERTISING RESEARCH "#$%& '()'
*+, -./,0 .1 2345,6
social-media impressions increased total
online impression volume by an impres-
sive 64 percent. This amplication in social-
media brand impressions largely was attri-
butable to Starbucks ability to leverage
a larger fan base and its modest (com-
pared to Bing and Southwest) advertising-
impression volume (See Figure 10).
Most marketers would agree that,
although they buy on an impression
basis, their true goal is to reach people
particularly those in their target demo-
graphic. Accordingly, the authors contend
that social-media research also should
focus on measuring the number and
composition of people reached by brand
messages on Facebook and other social
networks.
#$%&'()*+,&' .+'%/0 12+34+5%60 '5, 7.89
Facebook provides the opportunity to
deliver not only brand impressions at scale
but impressions with social context. This
means that it has the potential to generate
lift at each stage in the marketing cycle
brand awareness, favorability, purchase
intent, conversion, or even long-term loy-
alty and lifetime value. The ability to meas-
ure the impact of these brand impressions
in terms of traditional marketing metrics,
in fact, can help brands realize the value of
social media as an integral component of
their overall marketing mix.
To achieve this cross-media comparabil-
ity, social-media brand impressions must
be quantied and translated into tradi-
tional media planning variables such as
reach, frequency, and GRPs.
The authors calculated these metrics
for Starbucks, Southwest, and Bing based
on the number of social-media brand
impressions delivered against fans and
friends of fans. The results showed that,
although fans received a higher num-
ber of impressions on average, friends of
fans represented a wider reach.
5
Due
to Starbucks larger overall fan base,
the brand delivered the highest overall
number of total U.S. GRPs at 17.3 (5.7
among fans and 11.6 among friends).
Southwest delivered a combined 2.0
GRPs whereas Bing delivered 2.7 GRPs
(See Table 1).
5
comScore Social Essentials, U.S., May 2011
%
Source: comScore Social Essentials, U.S., May 2011
2.9%

0.4%
0.6%

4.9%

0.5%

1.0%
0
1
2
3
4
5
6
Starbucks Southwest Bing
Fans

Friends
1&:42+ ; 78#$9:%;<= 7>:8&?@<8 A BCDEF *>8#G HI7I 4D8@$D@8 0@#%&
JKL >M 7>%C#G "@NC# B$#DN 4OP$@<<C>D< #O>DE 1#D< A 1$C@DN< >M
1#D<I
Source: comScore Social Essentials, U.S., May 2011
81,652
355,866
158,230
52,365
5,918

8,111

0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
400,000
Starbucks Southwest Bing
Display Ad Impressions

Social Impressions
1&:42+ <= 78#$9:%;<= 7>:8&?@<8= #DN BCDEF 1#%@9>>; B$#DN
4OP$@<<C>D< J(((L Q@$<:< RC<PG#S TN 4OP$@<<C>D< J(((L #O>DE
1#D< #DN 1$C@DN< >M 1#D<I
!"#$% '()' JOURNAL OF ADVERTISING RESEARCH !"
*+, -./,0 .1 2345,6
#$%&'()*+,&' '.,&+/%+ '/'(0#&#
Understanding the size of the audience
reached by means of social media leads
to a natural follow-on question: Who are
these people?
With traditional campaigns, market-
ers attempt to reach particular audience
segments with their messages andin
the context of social-media campaigns
these marketers want to understand how
the audiences align with their market-
ing objectives. The analysis that follows
illustrates the demographic and behav-
ioral attributes of various Facebook fan
segments.
,123456789: ';6<=>9>
On the demographic side, the brand-
exposed fans of the three brands studied
differ by age composition (See Figure 11).
Starbucks actually had a fairly young
demographic prole for its fans, with
more than 40 percent of the audience
younger than 25, and 18- to 24-year-olds
accounting for more than a fourth of all
fans (26.5 percent).
Approximately 40 percent of Bings
fans also were younger than 25, although
they were more evenly balanced among
the younger-than-18-year-old segment
and 18- to 24-year-olds. Southwest fans,
meanwhile, showed an older age prole
with more than half (52.0 percent) at least
35 years old.
In the case of Starbucks, the authors
observed some differences in the respective
demographic proles of its brand-exposed
Facebook fans in comparison to its typical
in-store customers (See Figure 12).
Specically:
Among people age 18 and older, 20
percent of those who had visited a Star-
bucks premises within the past month
were between the ages of 18 and 24;
31 percent of brand-exposed Starbucks
fans also were in that age segment.
At the same time, 14 percent of in-store
visitors were age 55 and older, but this
same age segment accounted for just
7 percent of brand-exposed fans.
With the knowledge that fans skewed
signicantly younger than the typical
customer, Starbucks likely would benet
by designing its social marketing strate-
gies to bring younger customers in the
door.
Understanding the demographic char-
acteristics of a brands social-media fol-
lowing may reveal interesting audience
dynamics and uncover the potential for
new marketing opportunities.
?'@(+ A
789":; <#"8; 4=9#>??:@8? @8 1"$>A@@B
+B73>1C
D6;>
/6663
&2751>>93;>
/6663
?3E6< .F#F
+,'-. /73
?3E6<
&;E15;1E
+,'-. /73 D51GH1;:=
?3E6< .F#F
IJK>
D6;>
CD"#AE$B? FGH)I )JG(KL 'M) 'MK 'MJ NMJ
C@ED%O>?D K)J 'GK'I (MH (MI HM' )M(
<:8P )G'') HGFJ( (MI (MF HM( )M'
D591;C>
CD"#AE$B? )(GFH( HNG'FJ HMN IMK HMH ))MF
C@ED%O>?D )G)'H 'GKKI (MI (MN 'MJ )M(
<:8P 'G'(L IGII) (MJ )M( 'M( )MN
Source: comScore Social Essentials, U.S., May 2011
Age
Under 18
Age
1824
Age
2534
Age
3544
Age
4554
Age
55+
%
Source: comScore Social Essentials, U.S., May 2011
0
5
10
15
20
25
30
Starbucks
Southwest
Bing
D94H51 AA CD"#AE$B?G C@ED%O>?DG "8; <:8PQ 1"$>A@@B 1"8 RP>
4#5*(%$"6&7 0%*12#3
!" JOURNAL OF ADVERTISING RESEARCH "#$%& '()'
*+, -./,0 .1 2345,6
What is clear in these examples is
that it is difcult to generalize about a
brands social-media audience. They may
in some cases closely mirror ones typi-
cal customer base or, in other cases, look
very different. Because attracting fans
can be a function of both organic
brand afnity and active social-media
marketing efforts, there is no obvious rule
of thumb.
Nonetheless, it is important that brands
understand who these fans are to utilize
the strategies that will be most effective in
activating these customers.
#$%&'()*&+ -.&+/0(0
The current research demonstrates that
fans have different online behavioral pro-
les by brand. In the analysis that follows,
the authors examined the percentage of
time a particular brands fans spent in dif-
ferent site categories relative to the aver-
age U.S. Internet user.
The results indicated several strong
behavioral skews by brand among the top
content categories (See Figure 13):
Brand-exposed fans of all three brands
signicantly over-indexed for time
spent in the social networking category.
This fact should not be surprising, given
that these user segments are more likely
to associate with brands on Facebook.
All three brands over-indexed for the
entertainment category (consistent with
their relative preference for leisure-
oriented online content such as social
networking).
All three brands over-indexed for
search/navigation.
Bing fans exhibited a strong propensity
to visit automotive pages but relatively
low visitation for the nancial news,
newspapers and travel categories.
Southwest fans, unsurprisingly, indexed
high on the travel category and the retail
category.
Starbucks fans also indexed high on
retail.
Brands can leverage the information
on their fans online behavior to better
understand what messaging might work
best with customers, which products to
%
Source: comScore Social Essentials and comScore Plan Metrix, U.S., May 2011
30.8
28.5
18.9
14.7
4.7
2.3
19.7
26.4
20.1
19.5
9.7
4.6
0
5
10
15
20
25
30
35
% of Starbucks Fans
% of Starbucks Store Visitors
Age
1824
Age
2534
Age
3544
Age
4554
Age
5564
Age
65+
1(23*$ 45 4#5*(%$"6&7 0%*12# *. 8%$'/9:;"*-#/ <=$%>?7@-
1#78 9:$8;8 <=#$>;%?8 <=@$: AB8B=@$8C
350 300 250 200 150 100 50 0
Source: comScore Social Essentials, U.S., May 2011
Automotive
Financial News
Social Networking
Entertainment
Games
Newspapers
Retail
Search/Navigation
Travel
Starbucks
Southwest
Bing
1(23*$ 46 <=#$>;%?8D <@;=&E:8=D #7F GB7H G$#7FI,JK@8:F
1#78L M@7=:7= M#=:H@$N AB8B=#=B@7 >N *BO: <K:7= 47F:JC
!"#$% '()' JOURNAL OF ADVERTISING RESEARCH !"
*+, -./,0 .1 2345,6
promote via social media, and what type
of promotions are most likely to gain
traction.
Another way for brands to build a pro-
le of their fans is to understand which
content categories they are most likely to
visit relative to the average Internet user
(See Table 2).
Starbucks fans, for example, can be
described as both entertainment-oriented
and information seekers. Southwest fans
show great interest in the weatherprob-
ably because they are heavier travelers
whereas Bing fans are more likely to be
auto-intenders and consumers of teen
content.
#$%&'(() (##&+, -.('$. /$+)&012-
%3$22&. #(+ '+$24,
Although most campaign measurement
and evaluation occur at a local level, it also
is important for many brands to take a
more global perspective.
Facebooks extensive worldwide reach
makes it one of the few media channels
that can be employed for global brand-
ing efforts. According to comScore, Face-
book reached 714 million worldwide
unique visitors in Maya performance
that ranked it as the third largest world-
wide Web property. It also ranks as the top
social-networking site in the wide major-
ity of countries measured by comScore.
In terms of its ability to reach Fans
across the world with brand impressions
on Facebook (See Figure 14), the cur-
rent study showed that Starbucks had
a diverse base of exposed fans across all
ve global regions, with North America
accounting for less than 30 percent of the
total. Southwest had a more concentrated
fan baseunderstandable because it is a
domestic airline. Binga global brand
enjoyed a fairly diverse fan base, though
not as diverse as Starbucks.
Starbucks, which boasts 5,500 coffee
houses in more than 50 countries, clearly
is capitalizing on Facebook as a global
marketing channel. In May, Starbucks
social-media brand impressions reached
more than 53 million people worldwide,
with 10 individual countries delivering
impressions to at least 1 million fans and
friends. The United States led the way with
14.4 million exposed fans and friends, fol-
lowed by Germany (4 million), Mexico
(2.5 million) and Canada (2.3 million;
6
See
Figure 15).
#$2 5$.6& $24 ,(%1$.7/&41$ +&06+2
(2 125&,0/&20
Perhaps the most common questions that
arise in evaluating social-media efforts are
6
comScore, see above
0$'.& 8
789:;9: 7":;<8#=;> !8>:
3=?;@A :8 B; C=>=:;D BA 1"9>
0;@":=E; :8 :%; FE;#"<;
49:;#9;: G>;#
19:;<
,=>?@ABCD #>9D
H8$="@ I;:J8#?=9< K'L
,9:;#:"=9M;9:N!O@:=M;D=" )PQ
H;"#$%RI"E=<":=89 )SQ
T=#;$:8#=;>N0;U;#;9$; )Q(
,9:;#:"=9M;9:N!O>=$ )Q(
,EA=FG;D= #>9D
/;":%;# L)S
H8$="@ I;:J8#?=9< 'VL
W"AR3;>B="9 )SS
H;"#$%RI"E=<":=89 )SV
W"M=9< 49U8#M":=89 )SL
'H9I #>9D
H8$="@ I;:J8#?=9< KK'
FO:8 0;>8O#$;> K)X
78MMO9=:AN*;;9> 'K(
,9:;#:"=9M;9:N!O@:=M;D=" ''K
FO:8 !"9OU"$:O#;# )PX
100% 80% 60% 40% 20% 0%
Source: comScore Social Essentials, U.S., May 2011
Starbucks
Southwest
Bing
North America Europe
Latin America Asia Pacific Middle East
#HIA?; J! H:"#BO$?>Y H8O:%J;>:Y "9D Z=9<[ -;#$;9:"<; 8U
,\]8>;D 1"9> BA /8#@DJ=D; 0;<=89^
!" JOURNAL OF ADVERTISING RESEARCH "#$%& '()'
*+, -./,0 .1 2345,6
What is a fan worth? and What is the
return on my social-media investment?
Answering such queries must begin
with quantifying the value of a fan, which
might be considered in the following ways:
increasing engagement and loyalty
among fans,
generating incremental purchases
among fans, and
positively inuencing friends of fans.
A retailer may want to understand how
much more time fans spend at its store
than non-fans. A Consumer Packaged
Goods (CPG) brand may be interested in
knowing how much more of its product is
purchased in-store by fans than non-fans.
Similarly, a publisher could be curious
about how much more time is spent on
its Web site than a non-fan. The potential
applications for understanding these dif-
ferences are virtually limitless.
A good starting point for quantifying
the value of social-media audiences is
to analyze basic brand engagement. For
Starbucks, Southwest, and Bing, the cur-
rent study began by evaluating how much
more likely fans and friends of fans were
to visit the Web sites for those brands com-
pared to the average visitor of those sites.
Among the results (See Figure 16):
Starbucks fans and friends of fans
showed the highest average propensity
to visit the brand Web site, with fans 418
percent more likely to visit Starbucks.
com and friends of fans 230 percent
more likely.
Southwest fans were 362 percent more
likely to visit Southwest.com, whereas
friends of fans were 165 percent more
likely to follow a comparable action.
Bing, which showed relatively more
modest lifts at 55 percent for fans and
33 percent for friends of fans, had much
higher rates of visitation overall than the
other two brands Web sites, with more
than 1 of every 3 U.S. Internet users vis-
iting Bing in May.
With these examples, the authors were
able to establish that fans and friends
of fans tended to exhibit signicantly
higher brand engagement than the typical
Source: comScore Social Essentials, U.S., May 2011
14,417
3,972
2,549
2,276
1,854
1,616
1,453
1,315
1,260
1,233
United States
Germany
Mexico
Canada
Italy
United Kingdom
Malaysia
Philippines
India
Taiwan
#$%&'( *! 78#$9:%;<= *>? )( @>:A8$BC< 9D E:F9C$ >G ,H?><CI
1#A< #AI 1$BCAI< >G 1#A< J(((KL
%
Source: comScore Social Essentials, U.S., May 2011
418%
362%
55%
230%
165%
33%
0
50
100
150
200
250
300
350
400
450
Starbucks Southwest Bing
Fans Friends
#$%&'( *+ 78#$9:%;<M 7>:8&NC<8M #AI OBAP= /C9<B8C QB<B8#8B>A
#F>AP ,H?><CI 1#A< #AI 1$BCAI< >G 1#A< RC$<:< SRC$#PC
4A8C$AC8 T<C$L
!"#$% '()' JOURNAL OF ADVERTISING RESEARCH !"
*+, -./,0 .1 2345,6
Internet user and also to quantify the mag-
nitude of the difference. This evaluation,
however, can often be taken to the next
level of analysis:
Starbucks
To understand differences in purchase
patterns among the fans and friends
segments, comScore analyzed Starbucks
customers in-store purchase patterns.
The analysis,
7
which examined in-store
purchase behavior during May 2011,
revealed that Starbucks fans and friends
of fans spent 8 percent more and trans-
acted 11 percent more frequently than
the average Internet user who trans-
acted at Starbucks (See Figure 17).
It is important to note that differences
between fans and non-fans in terms of
purchase quantity or frequency are not
truly reective of the value of a like.
At least a portion of these differences
reect the preexisting preferences of
these consumers: they prefer Starbucks,
make a habit of visiting Starbucks often,
and also are more likely to express this
brand afnity by becoming a fan of Star-
bucks on Facebook.
7
Datapoint?
Thus, differences in spend may reect
the strength of preexisting brand prefer-
ences. That said, it is clear that these are
valuable customers for Starbucks, and
social media represent a strong oppor-
tunity to engage with them.
To understand the incremental value
of the published content, this research
technique can be used to identify Star-
bucks fans and friends of fans who have
received branded content and evaluate
their behavior over time. It can be used
to compare this behavior not only to that
of non-fans but based on the frequency
and diversity of exposure to branded
content and its latent impact on pur-
chasing behavior.
Southwest
For Southwest, the authors conducted
an analysis comparing Southwest.com
site visitation patterns between fans and
friends of fans and the average site visi-
tor. The results indicated substantially
higher visitation rates among fans and
friends of fans. Whereas 2.7 percent of
the total U.S. Internet audience visited
Southwest.com in May 2011, 12.4 per-
cent of exposed Southwest Facebook
fans visited the sitenearly ve times
the total Internet average. Perhaps
even more dramatically, 7.1 percent of
exposed Southwest friends of fans vis-
ited Southwest.commore than 2.5
times the rate among the total Internet
population.
These results suggested that exposure
to socially inuenced brand messages
might be persuasive in driving brand
engagement (See Figure 18).
For a Web site such as Southwest.com,
which sells tickets directly and there-
fore represents the companys primary
revenue-generation channel, differences
in site visitation rates can be tied to con-
version to help quantify the value of
fans and friends of fans.
Bing
For Bing, the authors conducted a similar
analysis comparing Bing.com search pat-
terns between Bing fans and friends of
fans and the average Bing.com searcher.
The results, once again, indicated
higher engagement across the board
among the fans and friends of fans seg-
ments. Overall, Bing fans conducted
68 percent more searches on Bing than
%
Source: comScore Social Essentials, U.S., May 2011
2.7%
12.4%
7.1%
0
2
4
6
8
10
12
14
Total Internet Southwest Fans Southwest Friends
#$%&'( "* 789:%;<=:>$8? @A=A:":A8B 0":< 8C 789:%;<=: D#"BEF
,GH8=<E 1"B= "BE 1#A<BE= 8C 1"B=>
%
Source: comScore Social Essentials, U.S.,
May 2011
8%
11%
0
2
4
6
8
10
12
Spending
per Buyer
Transactions
per Buyer
#$%&'( "+ 7:"#I9$J=K 1"B=
"BE 1#A<BE= 8C 1"B= -9#$%"=<
L$:AMA:N M<#=9= LM<#"O<
4B:<#B<: P=<#>
!" JOURNAL OF ADVERTISING RESEARCH "#$%& '()'
*+, -./,0 .1 2345,6
the average Bing searcher (40.9 versus
24.3) whereas friends of fans conducted
27 percent more searches (30.8 versus
24.3). Fans and friends also exhibited
higher search engagement in terms of
the number of searches per session,
searches per usage day, and number of
search usage days.
In each case, Bing fans demon-
strated signicantly higher engagement
than the friends of fans segment (See
Figure 19).
This dynamic illustrates the importance
of the fan and friend segments to Micro-
soft, as their association with the Bing
brand corresponds directly with Micro-
softs ability to monetize search behavior.
#$%#&'()$%
As marketers increasingly look to Face-
book and other social-media platforms as
new ways of communicating with their
consumers, they seek metrics that can help
them understand the value and create an
effective strategy for reaching key audi-
ence segments.
Typical approaches that focus on raw
fan counts, howeveror the total
number of engagements on a given piece
of contentfail to depict the potential
and realized scope of social-media brand
impressions.
As illustrated through the examples in
this study, a measurement approach that
focuses on reach and frequency within
audience types (for example, fans and
friends of fans) can lead to a dramatically
better understanding of how and where
brand messages are reaching consumers.
This same focus means that marketers
also can better place their social-media
strategy within the broader framework
of their marketing objectives for key con-
sumer segments.
In particular, the ndings in this arti-
cle may be of interest to brand marketers
as they consider their realization of the
value of a fan on Facebook.
In addition to understanding the
short- and long-term reach of social-media
brand impressions to fans, this approachs
ability to determine reach among the
largely untapped friends of fans audi-
ences may be important in helping brands
to achieve their marketing goals on Face-
book and beyond.
!"#$%& ()*+,!" 78 97%: -$:87;:<= >? "#$@:=7<A B 4<;C8=$D
E<#FD878 #= %>GH%>$:I E<;$:J 8K:%7#F7L:8 7< 8:M:$#F
#$:#8 >? $:8:#$%&N 7<%FC;7<A 8>%7#F G:;7#N :O%>GG:$%:N
><F7<: M7;:>N #<; ><F7<: #;M:$=787<A #<; 78 ?$:PC:<=FD
PC>=:; QD F:#;7<A <:J8 >$A#<7L#=7><8N 8C%& #8 =&:
New York TimesN Wall Street JournalN RSSN 1>$Q:8N
BusinessWeek #<; NewsweekI E<;$:J 78 %C$$:<=FD
KC$8C7<A &78 "TE #= S>$=&J:8=:$<U8 5:FF>AA H%&>>F >?
"#<#A:G:<= #<; &>F;8 # TE 7< -CQF7% ->F7%D H=C;7:8
?$>G VC@: W<7M:$87=DI
-$!.!, ,/## 78 H:<7>$ 97%: -$:87;:<= >? H:#$%& #<;
":;7# #= %>GH%>$:N Q#8:; 7< H#< 1$#<%78%>I +:
G#<#A:8 $:8:#$%& K$>X:%=8 #<; %F7:<= $:F#=7><8&7K8
J7=& # <CGQ:$ >? K$>G7<:<= ><F7<: G:;7# %>GK#<7:8N
7<%FC;7<A Y>>AF:N 1#%:Q>>@N Z#&>>[N +CFC #<; V78<:DI +:
78 #< :\K:$7:<%:; G#$@:=:$ #<; 7<;C8=$D #<#FD8=N &#M7<A
8K:<= =:< D:#$8 ;7$:%=FD #<; 7<;7$:%=FD A$>J7<A G#X>$
=:%&<>F>AD Q$#<;8I Y$#&#G &>F;8 #< "TE ?$>G =&:
5:FF>AA H%&>>F >? "#<#A:G:<=N S>$=&J:8=:$< W<7M:$87=DN
#<; # TH J7=& &><>$8 7< -8D%&>F>AD ?$>G H#<=# RF#$#
W<7M:$87=DI
,)0% $)1. 78 # 97%: -$:87;:<= 7< %>GH%>$:U8 %C8=>G
G:;7# $:8:#$%& ;7M787><I 4< &78 :7A&= D:#$8 #= %>GH%>$:N
"7@: &#8 &:FK:; K7><::$ <:J G:=&>;8 ?>$ G:#8C$7<A
><F7<: M7;:>N 8>?=J#$:N &#$;J#$:N #<; 8>%7#F G:;7#I -$7>$
=> %>GH%>$:N "7@: J>$@:; 7< G:;7# #<; =:%&<>F>AD
%><8CF=7<A $>F:8 #= E88>%7#=:; -$:88N V>%:C8N #<; 3:#K
/7$:F:88I "7@: &#8 # TH 7< ]>C$<#F78G ?$>G T>8=><
W<7M:$87=DI +: %C$$:<=FD F7M:8 7< H:#==F: J7=& &78 J7?: #<;
=J> %&7F;$:<I
+%!" 2$/)1. F:#;8 1#%:Q>>@U8 G:#8C$:G:<= $:8:#$%&
A$>CKI H:#< #<; &78 =:#G #$: $:8K><87QF: ?>$ G:#8C$7<A
&>J #;M:$=787<A #<; Q$#<;:; %><=:<= 7GK#%= C8:$8U
Q:&#M7>$8 #<; K:$%:K=7><8N #<; ?>$ ;:M:F>K7<A #<;
8=#<;#$;7L7<A G:=&>;>F>A7:8 ?>$ :M#FC#=7<A =&:
:??:%=7M:<:88 >? G#$@:=7<A K$>A$#G8I T:?>$: X>7<7<A
1#%:Q>>@N H:#< J#8 J7=& Y>>AF:U8 RC8=>G:$ 3#Q8 #<;
^C#<=7=#=7M: "#$@:=7<A A$>CK8N J&:$: &: G#<#A:;
Y>>AF:U8 PC#<=7=#=7M: #;M:$=78:$ $:8:#$%& #<; ;:M:F>K:;
#;M:$=787<A :??:%=7M:<:88 G:#8C$:G:<= =>>F8 ?>$
#;M:$=78:$8I H:#< 8=C;7:; %>A<7=7M: <:C$>8%7:<%: #=
H=#<?>$; W<7M:$87=DI
Searches
per Searcher
Searches
per Session
Searches
per Usage Day
Searches
Usage Day
Source: comScore Social Essentials, U.S., May 2011
24.3
2.6
4.6
5.3
40.9
2.8
5.8
7.1
30.8
2.6
5.1
6.1
0
10
20
30
40
50
Total Internet Fan Friend of Fan
*+,-./ 12 T7<A_ T$#<;O,\K>8:; 1#<8 #<; 1$7:<;8 >? 1#<8
H:#$%& E%=7M7=D M:$8C8 EM:$#A: T7<A H:#$%&:$I
Copyright of Journal of Advertising Research is the property of Warc LTD and its content may not be copied or
emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.