ArticlePDF Available

What's Not to "Like?" Can a Facebook Fan Base Give a Brand The Advertising Reach It Needs?


Abstract and Figures

A marketer with a Facebook Fan base has at least some ability to advertise to that audience. What quality of reach, however, does this sort of "earned media" deliver? The landmark discovery by Andrew Ehrenberg of the negative binomial distribution (NBD) implies that the most effective advertising requires media that reach across both heavy and light buyers of the brand. This article investigates the buying concentration of the Facebook Fan base of two different brands (both Fast Moving Consumer Goods (FMCG) categories) and compares it to the brands' actual buying bases. The buyer base of each of the brands is distributed in the typical NBD, whereas the Fan base delivered by Facebook is skewed in an opposite pattern-skewed toward the heaviest of the brands' buyers-making the quality of Facebook's reach appear rather unappealing.
Content may be subject to copyright.
DOI: 10.2501/JAR-52-2-000-000 June 2012 JOURNAL OF ADVERTISING RESEARCH 1
No other single media platform can boast the speed
of user uptake as Facebook. It is estimated that,
by 2015, social media will become a mainstream
mass-media platform that, in one form or another,
will engage one-third of the world’s population.
This penetration would offer advertisers access
to 80 percent of global consumer expenditures—
a potential $29 trillion market (Nuttney, 2010). In
light of such predictions, it is understandable that
many marketers are including social media in their
media mix. They do so, however, with limited
understanding of whether social media are more
effective than other platforms or of how they can
used most effectively (Nelson-Field and Klose, 2010).
Facebook is the dominant—and fastest-
growing—social medium, with more than 850
million active users. For marketers, the Facebook
platform offers a different kind of mechanism for
communicating with their potential audiences.
When it is compared to offline media, Facebook
often is reported as a cost-effective way of develop-
ing and communicating with actual (and/or poten-
tial) customers (comScore, 2011; Millward_Brown,
2011; Gibs and Bruich, 2010; Syncapse, 2010).
Advertisers create Facebook Fan pages for
their brands and then encourage Facebook users
to become “Fans” of these pages by clicking the
“like” button on the page. Once a user has “liked”
a brand’s page in this manner, they may receive
brand updates—and the observations of other
brand enthusiasts—in their personal newsfeeds.
The number of attracted and maintained Fans
typically are key metrics for evaluating the success
of Facebook marketing efforts (Millward_Brown,
2011; Sterne, 2010). How many Fans a brand has
obviously affects the breadth of a brand message,
but who these people are—not just how many
of them there may be—also is important. Some
industry studies have proposed that brand Fans
may spend more than non-Fans (Millward_Brown,
2011; Syncapse, 2010)—a finding that intuitively
would suggest that the Facebook enthusiasts are
heavy brand buyers. The concentration of these
“valuable” Fans across the entire Fan base, how-
ever, is unknown. Furthermore, whether these
valuable brand Fans have changed their buying
behavior after becoming a Fan also is unknown.
As such, the following three questions were the
focus of this research:
What’s Not to “Like?”
Can a Facebook Fan Base Give a Brand
the Advertising Reach It Needs?
Ehrenberg-Bass Institute
Ehrenberg-Bass Institute
Ehrenberg-Bass Institute
A marketer with a Facebook Fan base has at least some ability to advertise to that
audience. What quality of reach, however, does this sort of “earned media” deliver? The
landmark discovery by Andrew Ehrenberg of the negative binomial distribution (NBD)
implies that the most effective advertising requires media that reach across both heavy
and light buyers of the brand. This article investigates the buying concentration of the
Facebook Fan base of two different brands (both Fast Moving Consumer Goods (FMCG)
categories) and compares it to the brands’ actual buying bases. The buyer base of each of
the brands is distributed in the typical NBD, whereas the Fan base delivered by Facebook
is skewed in an opposite pattern—skewed toward the heaviest of the brands’ buyers—
making the quality of Facebook’s reach appear rather unappealing.
• Are Fans of brands on Facebook heavy
buyers of the brand?
• What is the concentration of these buy-
ers across the brand Fan base?
• Has the Fan recruitment profile
changed? Is the reach broadening?
Targeting strategies that focus on reaching
(and rewarding) heavy brand buyers long
have been popular in advertising practice.
For example, an analysis of advertising
effectiveness entries showed that this was
by far the most popular strategy, although
the opposite strategy (targeting light and
non-customers) was associated with far
higher sales and profit results (Binet and
Field, 2009).
The thinking behind such a strategy
seems, at least superficially, to be logical:
It would seem to make sense to spend
more money on customers who are worth
more to the brand. That logic is flawed,
however, because what matters is not
how much customers buy but rather how
they respond to advertising (Wright and
Esslemont , 1994; Watts, 2011). Heavy-loyal
customers are, of course, by definition
worth to the brand per customer, but it is
unlikely that any advertising could stimu-
late them to buy more than they already
do: Customer who already are 100-percent
loyal to a brand are unable to give
the brand any more of their category
Andrew Ehrenberg’s discovery that
a brand’s customer base is described by
a negative binomial distribution of buy-
ing rates (Ehrenberg, 1959) shows that
reaching all types of customers is para-
mount. The NBD accurately describes
the frequency distributions of purchase
rates across a population of consumers
for a single brand or category (Ehrenberg,
1988; Goodhardt, Ehrenberg, and Chat-
field, 1984; Morrison and Schmittlein,
1988; Uncles, Ehrenberg, and Hammond,
The distribution has been validated
in many repeat-purchase markets across
hundreds of categories and brands and
has become a well-established benchmark
of the buying concentration of a brand’s
customer base (Ehrenberg, 1988; Good-
hardt et al., 1984; Morrison and Schmit-
tlein, 1988; Uncles et al., 1995). The NBD
denotes that the heterogeneity in purchas-
ing rates (λ) follows a gamma distribution
in that under most conditions, it reflects a
high incidence of light buyers (shoppers
who have a low to close-to-zero purchas-
ing rate), fewer medium buyers, and very
few heavy buyers (Ehrenberg, 1988).
As a brand grows, it moves from one
NBD to another. As it moves, however, it
achieves higher penetration and higher
average purchase rates. Because of the
shape of the distribution, the bulk of
change is seen among the brand’s very
many light (and non-) buyers (McDonald
and Ehrenberg, 2003; Stern and Ehren-
berg, 2003). Growth-oriented advertising,
therefore, needs to reach light (and non-)
It may be argued that the bulk of work
that advertising accomplishes is not to
increase market share but to maintain it
(Ehrenberg, Barnard, and Scriven, 1998).
In this case, the sales effects of advertis-
ing are not to increase purchase rates but
to prevent sales erosion that would oth-
erwise occur. It, therefore, would seem
highly important to reach the heavy cus-
tomers who do the most buying of the
brand; after all, this audience has the high-
est potential for erosion.
The sales importance of heavy buyers,
however, often is overstated. Recent stud-
ies have hypothesized that, for grocery
brands, the 80/20 adage (the top 20 per-
cent of a brand’s customers are respon-
sible for 80 percent of sales) does not apply
and that the top 20 percent of customers,
in fact, contribute less than 60 percent of
sales (Sharp and Romaniuk, 2007; Sharp,
2010). The brand’s lightest 80 percent of
customers, therefore, are important for
maintenance and growth; they contribute
half of today’s sales—an advantage brand
manager would prefer to retain. Further-
more, these buyers rarely think of the
brand and rarely buy it. They are likely to
be lured away by competitors’ advertis-
ing, particularly if the brand’s own adver-
tising efforts fail to reach them.
The work by Ehrenberg and others in doc-
umenting the sources of a brand’s sales
leads to a succinct rejection of the target-
ing strategy. As Ehrenberg wrote, his dis-
covery ended “marketing’s pipedream of
just recruiting heavy buyers” (Ehrenberg,
2005). Instead of targeting heavy brand
buyers, communication strategies, it was
argued, should seek to communicate with
potential buyers from across the entire
customer base—including not just light
buyers of the brand but non-buyers (so
long as they are already engaged in the
category), many of whom are not non-
buyers at all but rather simply light buy-
ers who did not make a purchase this year
but may return next year (Goodhardt and
Ehrenberg, 1967; Anschuetz, 2002; Morri-
son and Schmittlein, 1988).
At this point, it would seem that a
case could be made for the conclusion
that the ideal medium (or at least mix of
media) should reach all category buy-
ers and, hence, find the brand’s heaviest
and very lightest of buyers. The heavier
buyers, however—the more loyal cus-
tomers—are easier to reach than light buy-
ers because they more easily notice and
mentally process the brand’s advertising
messages (Sharp, 2010). These heavier
buyers also receive more reinforcement
from the brand’s other marketing efforts,
including packaging. The implication is
that advertising media that skew toward
light buyers are particularly valuable. By
contrast, media that skews towards heav-
ier buyers offers something that is easily
achieved and therefore less valuable to an
And, it is from this heavy-user/light-
user perspective that the authors have
examined the influence of marketing on a
brand’s Facebook Fans.
To analyze the difference between a typi-
cal population of shoppers—a group that
typically would be NBD-distributed—and
the population of a Facebook Fan base, the
authors used two data sources:
• Self-reported purchase data from Face-
book Fans of two (unnamed) brands
from two different repeat-purchase cat-
egories (chocolate and soft drinks)
• These data was collected with a self-
completion Web-based survey link. One
of the data sets was collected via a link
on the brand’s Fan page, where only
Fans of the brand could respond; the
other was sourced from a probability-
based online panel—where respondents
could be Fans of any brand in the cat-
egory—with analysis restricted to Fans
of the one focal brand.
• Actual consumer panel data for the
same brands for direct comparison
To ensure consistency between the com-
parisons, the authors
• mined a 12-week rolling average and
• converted this continuous data into the
same categorical measures (i.e., grouped
into “never”; “once”; “two-three times”;
“4 or more times”).
For both data sets, the authors classified
these purchase categories:
• “non-buyers” (“never”);
• “light” (“once”);
• “moderate” (“two–three times”); and
• “heavy buyers” (“four or more times in
3 months”).
The use of retrospective self-reported
purchasing as the basis of frequency dis-
tributions introduced the possibility of
respondent error. For example, one 1979
study concluded that the average correct
classification rate (across seven brands
and three different purchase-related ques-
tions) was only 49 percent (Wind and
Learner, 1979).
More recent studies have considered
purchase intention (Romaniuk, 2004;
Wright and MacRae, 2007); product con-
sumption (Stanton and Tucci, 1982); and
product usage (Hu, Toh, and Lee, 2000;
Ram and Hyung-Shik, 1990). The com-
parisons between recall and panel data
were consistent, with respondents tend-
ing to give responses for a typical period
rather than the specific period they were
questioned about. Those results mean
that self-report data potentially can out-
perform panel data in correctly classify-
ing consumers into light or heavy buyers.
Moreover, because of stochastic variation
in purchasing, panel data always will mis-
classify some normally heavy consumers
as light because they bought at less than
their usual rate during the particular anal-
ysis period (Schmittlein, Cooper, and Mor-
rison, 1993). And some “light” buyers also
will be misclassified as “heavy.”
Some recent research in the area is more
directly relevant to this research because
it investigated distributions of heavy to
light behavior, whereas most previous
research considered only overall averages,
ignoring the heterogeneity in respond-
ents’ behavior (Nenycz-Thiel, Roma-
niuk, Ludwichowski, and Beal, 2012). In
that instance, the authors conducted two
studies—one in the chocolate category
(using purchase recall) and the other at
brand level (using television-program
viewing). Their results suggested that
the main source of error occurred at the
light buying/using end, where respond-
ents slightly underestimated infrequent
events; there was less classification error
with heavy buyers/users. These findings
are consistent with respondents report-
ing on their typical behavior rather than
specific behavior during the questioned
Such biases in self-report data, unsur-
prisingly, did not prevent respondents
from correctly identifying whether they
were heavy or lighter buyers. Indeed, self-
report data may outperform panel data in
making correct individual classifications.
Both approaches should generate correct
distributions of purchase weight—which
is the focus of this article.
In recognition of recommendations on
how to minimize errors to increase overall
survey accuracy (Nenycz-Thiel et al, 2012),
the authors applied the following stand-
ards in the current research:
• Reduced memory decay through
brand/category choice
Memory decay is a key factor hinder-
ing accuracy and can result in under-
reporting. Fewer reporting errors
occur when the (buying) event is more
frequent and, thereby, less reliant on
long-term memory (Hu et al., 2000; Lee,
Hu, and Toh, 2000; Sudman, 1964). In
this research, the authors considered
two leading brands from two repeat-
purchase categories (chocolate and soft
• Decreased the timeframe to improve
To improve recollection and increase the
vividness of the event, the authors lim-
ited the reporting timeframe to 3 months
(typical panel data for the NBD are
12 months; Allison, 1985; Tourangeau,
Rips, and Rasinski, 2000).
• Provided context cues to reduce
Respondents tend to generalize the
responses to reflect an overall behav-
ioral pattern. Such shortcuts can result
in over- and underestimation of fre-
quency (Hu et al., 2000; Tourangeau
et al., 2000). Context cues can improve
accuracy. In the current research, the
authors provided respondents with a
full list of potential responses. Applying
this context cue disallowed the respond-
ents to write in their own number
and potentially overestimate frequency
(Nenycz-Thiel et al., 2012). Further-
more, the authors asked respondents
to report purchase behavior in a “typi-
cal” 3 months (before and after becom-
ing a Fan). Ideally, adding this context
should have assisted the mental averag-
ing process and reduced exaggeration
(Parfitt, 1967).
• Reduced the complexity of the recall
One goal of the current study was to
simplify the recall task (Parfitt, 1967).
The authors simply asked, “In a “typi-
cal” three months, how often would you
buy this brand? Never? Once? Two or
three times? Four or more times?”
These frequency classifications were
appropriate for these categories, given
the expectation that the distribution of the
buying rates in the wider customer base
would be NBD-distributed. The authors
acknowledge that these classifications may
need to be altered for different categories,
with longer inter-purchase intervals.
The current study produced consistent
findings for both categories examined,
and the differences in distributions, in
fact, were so stark that any small biases
in either the panel or the self-report were
The buying concentration of a brand’s
Facebook Fan base was extremely differ-
ent from that of a “typical” (i.e., not NBD-
distributed) population of shoppers. In
fact, the authors found two generalizable
patterns in describing the average brand’s
Facebook Fan base:
• Using the classification of buying rates
as outlined above to study the distribu-
tion of buying rates for a typical choco-
late brand (based on 2011 consumer
panel data with a 12-week rolling aver-
age), the results revealed a typical NBD-
distributed customer base, with high
numbers of light (and zero) buyers and
fewer medium and heavy buyers (See
Figure 1).
• By comparison, the distribution of
buying rates for Facebook Fans of the
same brand in the chocolate category
showed a strikingly different shape
(See Figure 2): The Fan base produced
a particularly high incidence of heavy
buyers (57 percent), and virtually no
non-buyers (1 percent).
The Facebook Fan base of this chocolate
brand was very skewed to heavy buyers—
the opposite of a typically distributed cus-
tomer base.
The same results appeared in the authors’
analysis of the soft-drink category. Figure 3
is an example of a typical soft-drink cus-
tomer base. Using a 2007 consumer panel
data with a 12-week rolling average, the
typical soft-drink customer base showed
a typical NBD-distributed pattern. Con-
sistent with the findings for the chocolate
brand, the buying distribution among the
Buyer Groups
Proportion of shoppers
Figure 1 Buying
Concentration across the
Entire Customer Base for a
Chocolate Brand
Buyer Groups
Proportion of shoppers
Figure 3 Buying
Concentration across the
Entire Customer Base for a
Soft-Drink Brand
Buyer Groups
Proportion of fans
Figure 2 Buying
Concentration across the
Facebook Brand Fan Base for
the Same Chocolate Brand
soft-drink Facebook Fans was very differ-
ent and was skewed toward heavier buy-
ers (See Figure 4).
The findings were consistent between
two different brands in two different
categories: The Facebook Fan base was
strongly skewed to heavy buyers.
The analyses of both products dem-
onstrate that Facebook is a platform that
delivers an audience for advertising that is
skewed toward heavier brand buyers (See
Table 1). In more detail: The Facebook fan
base does not give the marketer access to
sufficient numbers of light buyers to main-
tain communication with a substantial
proportion of the customer base, particu-
larly if the desired outcome of communi-
cation efforts is to grow the brand (Sharp,
2010). Marketers who want to focus on
spending their resources on recruiting and
maintaining a Fan base, in essence, are
limiting their efforts to the small propor-
tion of the customer base who do not have
sufficient capacity to increase their buying
of the brand.
Given the use of self-report data in the
current study, the authors sought vali-
dation by conducting a third study that
examined self-report purchasing profiles
for a different media. Again, the authors
sourced the data from a probability-based
online panel (n = 397). In place of Face-
book Fans, however, in the third study,
the authors interviewed recent television
viewers—specifically, people who had
watched the 2012 Super Bowl (U.S. sample
weighted to the population).
The Super Bowl was chosen because it
delivers the sort of audience the brands
under study normally would like to
reach. (Had the authors chosen all tel-
evision viewers, the results might have
been skewed unduly toward light users
of the brands—an unfair comparison for
The authors asked the same purchase
frequency questions of the same two
brands in the original Facebook ques-
tionnaire. Distribution in the third study
clearly skewed to non- and light buyers
of the brand (See Figure 5)—consistent
with a typically distributed customer
base and diametrically opposite to the
same brands’ Facebook Fan bases (See
Figure 6).
The findings of buying concentration
from 2012 Super Bowl television audi-
ences were more consistent with a typical
population of a brand’s buyers. This dem-
onstrates that the Facebook Fan base gen-
erates a non NBD-distributed population,
not the nature of the claimed data.
In a side-by-side examination of audi-
ence concentration for both brands,
although the chocolate brand showed
a smaller skew to non-buyers than the
Buyer Groups
Proportion of fans
Figure 4 Buying
Concentration across the
Facebook Brand Fan Base for
the Same Soft-Drink Brand
Buyer Groups
Proportion of fans
Figure 6 Buying
Concentration across the
Facebook Brand Fan Base for
the Same Soft-Drink Brand
Buyer Groups
Proportion of viewers
Figure 5 Buying
Concentration of the 2012
Super Bowl Viewing Audience,
for the Same Soft-Drink Brand
Buying Concentration (%) across the Facebook Fan Base by
Light Buyers
Moderate Buyers
Heavy Buyers
Chocolate 1000 1 10 33 56
Soft Drink 520 4 9 24 63
Average 3 10 28 60
soft-drink brand, the direction of the dis-
tribution still followed an NBD and was
starkly different from the Facebook results
(See Table 2).
The act of “liking” brands on Facebook
has dramatically increased in popular-
ity in recent years (Cashmore, 2010). In
part, this is due to the natural growth of
the Facebook platform but also as a result
of its increasing cross-functionality with
other online media and portability with
other digital devices, making access to the
Facebook like button easier. That in turn
raises the question of whether brands’
Facebook brand Fan bases are becoming
more representative of their brands’ actual
buyer bases. If so, lighter buyers may have
been expected to have been Fans of brands
for shorter periods (i.e., they signed on as
Fans more recently). Similarly, heavy buy-
ers of a brand may have been Fans for a
longer period.
There is slight trend toward this split,
with “older” Fans (12–24 months) being
more likely to be heavier buyers than
“newer” Fans (See Table 3). Even the
newer Fan base, however, still was sig-
nificantly different from the brands’ actual
buyer base, demonstrating that Facebook
continues to attract heavy buyers of the
Newer Facebook Fans were highly simi-
lar to older Facebook Fans in that they
both skewed to heavy brand buyers.
This current study drew heavily on the
extensive work of Andrew Ehrenberg and
his colleagues. By discovering the statisti-
cal regularities in the buyer behavior of
individuals, they have highlighted the
important role that light buyers play for all
brands (Romaniuk, 2011). And, the authors
have relied on this insight to clarify one of
the disadvantages of “earned” media.
Specifically, the authors found that the
buying distribution of a brand’s Facebook
Fan base is opposite that of a typical popu-
lation of category buyers, with a signifi-
cantly higher incidence of heavy buyers
reached with this social-media vehicle.
This finding identifies a clear deficiency
of earned media and raises questions about
the value of the Facebook platform as a
stand-alone earned advertising medium.
One could argue that earned such as Face-
book are only parts of a multi-media mix
and that other media with better reach
profiles will reach the missing light buyers.
The question remains, however: In a mix
of media, is it ever a good idea to include a
medium that skews so strongly away from
light users? Almost any medium can reach
some part of the brand-loyal audience.
Consequently, advertisers should identify
media that can reach beyond their most
loyal customers. Media that skew as does
the Facebook Fan base are lower-quality
media and certainly should command low
Cost Per Thousand (CPMs).
There are a number of potential ben-
efits of a Facebook Fan base. Having direct
access to heavy buyers may provide a
useful research/insight opportunity by
providing a forum to listen to customers
(and competitor’s customers). Further-
more, Facebook may offer the potential for
Fans to become active brand advocates,
creating new networks that, in fact, include
light buyers. The ability of Facebook to lev-
erage such networks of brand Fans (and its
relative efficiency) is outside the scope of
this research and demands future research.
This current study should caution
marketers against using Facebook as
Buying Concentration (%) of the 2012 Super Bowl Viewing
Audience by Category
Soft Drink Chocolate
(n = 520)
Super Bowl
(n = 397)
(n = 1000)
Super Bowl
(n = 397)
Non-Buyers 4 47 1 28
Light 9 29 10 34
Moderate 24 11 33 23
Heavy 63 13 56 15
Average Buying Concentration (%)—All Categories
Purchase Classication
Length of time subscribed
<6 months 6–12 months 12–24 months Total
Non-buyers 4 2 0 2
Light 10 7 11 9
Moderate 27 31 26 29
Heavy 61 60 64 61
Total 100
a stand-alone medium to drive brand
growth. Furthermore, marketers should
be wary of over-investing in small rela-
tive pockets of heavy buyers if it comes
at the expense of overlooking light buyers
who may be the primary source of brand
Karen nelson-Field
is a post-doctoral research Fellow at
the Ehrenberg-Bass Institute for Marketing Science at
the University of South Australia. Her current research is
in the social media space, in particular whether existing
empirical generalizations in advertising, buyer behavior
and media hold in the social media context and how this
impacts on the ability of social media to assist brand
growth. Her ndings have been presented internationally
at the European Communications Symposium
(Barcelona), the London Business School, ESOMAR
World Media 3 (Berlin), ARF Rethink (New York), and
the prestigious Wharton Business School. Her industry
experience includes senior marketing roles in FMCG,
media, tourism and major retail over 16 years.
erica riebe
heads the Media Research Group at the
Ehrenberg-Bass Institute in the School of Marketing,
University of SA. Erica Riebe’s research focuses on the
following areas:
1) Determining effective and efcient media placement
2) Measuring the impact of changes in the media
environment on audiences;
3) Predicting the uptake of new products using
probabilistic scales;
4) Determining the impact of customer loss and gain on
the likely growth or decline of a brand.
byron sharp
is a professor of marketing science
and director of the Ehrenberg-Bass Institute at the
University of South Australia. His research is funded
by corporations around the world including Coca-Cola,
Mars, P&G, Kraft, Turner Broadcasting, CBS, and the
Australian Research Council. His book How Brands
Grow, Oxford University Press, 2010, presents a wide
variety of scientic laws and what they mean for
marketing strategy (see www.MarketingLawsofGrowth.
Allison, D.
“Survival Analysis of Backwards
Recurrence Times.” Journal of the American
Statistical Association 80, (1985): 315–322.
Anschuetz, n.
“Why a Brand’s Most Valuable
Customer Is the Next One it Adds.” Journal of
Advertising Research 42, 1 (2002): 15–21.
Binet, l.,
P. FielD .
“Empirical Gener-
alisations about Advertising Campaign Suc-
cess.” Journal of Advertising Research 49, 2 (2009):
cAshmore, P.
“Facebook ‘Likes’ World Domi-
nation [Online].” Mashable Social Media,
like-launch/%5D Retrieved on 3 November,
2011. “The Power of Like. How
Brands Reach and Influence Fans Through
Social Media Marketing.” comScore Inc.
ehrenBerg, A. s. c.
“The Pattern of Consumer
Purchases.” Applied Statistics 8, 1 (1959): 26–41.
ehrenBerg, A. s. c.
Repeat-buying: facts, theory
and applications, London, UK: Oxford University
Press, 1988.
ehrenBerg, A. s. c.
“My Research in Market-
ing.” Admap, 461 (2005): 6.
ehrenBerg, A. s. c., n. BArnArD,
and J.
“Justifying our Advertising Budgets.”
In Warc Conference paper, 1998, pp. 1–13.
giBs, J.,
s. Bruich.
2010. “Advertising
Effectiveness: Understanding the Value of a
social Media Impression. Nielsen and Facebook.
gooDhArDt, g. J.,
A. s. c. ehrenBerg.
“Conditional Trend Analysis: A Breakdown by
Initial Purchasing Level.” Journal of Marketing
Research 4, May (1967): 155–161.
gooDhArDt, g. J., A. s. c. ehrenBerg,
c. chAtFielD.
“The Dirichlet: A Com-
prehensive Model of Buying Behaviour.” Jour-
nal of the Royal Statistical Society 147, 5 (1984):
hu, m. Y., r. s. toh,
e. lee.
“Survey accu-
racy as a Function of Usage Rate.” Marketing
Letters 11, 4 (2000): 335–348.
lee, e., m. Y. hu,
r. s. toh.
“Are Con-
sumer Survey Results Distorted? Systematic
Impact of Behavioral Frequency and Duration
on Survey Response Errors.” Journal of Market-
ing Research 37, 1 (2000): 125–133.
mcDonAlD, c.,
A. s. c. ehrenBerg
“What Happens when Brands Gain or Lose
Share? Customer Acquisition or Increased Loy-
alty? Report 31 for Corporate Members. Adelaide,
Australia: Ehrenberg-Bass Institute for Market-
ing Science, 2003.
2011. “The Value of a Fan.”
morrison, D. g.,
D. c . schmittlein.
“Generalizing the NBD Model for Customer
Purchases: What Are the Implications and Is It
Worth the Effort?” Journal of Business and Eco-
nomic Statistics 6, 2 (1988): 145–159.
nelson-Fie lD, K.,
g. Klose.
“The Social
Media Leap: Integrating Social Media into
Marketing Strategy.” In WM3 Your Audience =
Media Consumer + Generator, Berlin, 2010.
nenYcz-thiel, m., J. romAniuK, g.
v. BeAl.
“Improving the
Accuracy of Consumer’s Self-Reported Brand
Usage Behaviour.” Journal of Business Research,
forthcoming (2012).
nuttneY, A.
The Social Networking Opportunity.
Business Insights, 2010.
PArFitt, J.
“A Comparison of Purchase Recall
with Diary Panel Records.” Journal of Advertis-
ing Research 7, 3 (1967): 16–31.
rAm, s.,
J. hYung-shiK.
“The Conceptu-
alization and Measurement of Product Usage.”
Journal of the Academy of Marketing Science 18, 1
(1990): 67–76.
romAniuK, J.
“Testing the Accuracy of Ver-
bal Probability Scale for Predicting Short-Term
Brand Choice.” Marketing Bulletin 15, (2004):
romAniuK, J.
“Are You Blinded by the Heavy
(Buyer)…Or Are You Seeing the Light?” Journal
of Advertising Research 51, 4 (2011): 561–563.
schmittlein, D. c., l. g. cooPer,
and D. g.
“Truth in Concentration in the Land
of (80/20) Laws.” Marketing Science 12, 2 (1993):
shArP, B.
How Brands Grow. Melbourne, Aus-
tralia: Oxford University Press, 2010.
shArP, B.,
J. romAniuK.
There is a Pareto
Law—but Not As You Know It. Report 42 for
Corporate Sponsors. Adelaide, Australia:
Ehrenberg-Bass Institute for Marketing Science,
stAnton, J. l.,
l. A.tucci.
“The Measure-
ment of Consumption: A Comparison of Sur-
veys and Diaries.” Journal of Marketing Research
19, May (1982): 274–277.
stern, P.,
A. s. c. ehrenBerg.
tions vs. Reality.” Marketing Insights, Marketing
Research Spring, (2003): 40–43.
sterne, J.
Social Media Metrics: How to Measure
and Optimize Your Marketing Investment. New
York, NY: John Wiley & Sons, 2010.
suDmAn, s.
“On the Accuracy of Recording
of Consumer Panels: II.” Journal of Marketing
Research 1, August (1964): 69–83.
The Value of a Facebook Fan: An Empir-
ical Review. London, UK: Hotspex, 2010.
tourAngeAu, r., l. J. riPs,
K. rAsinsKi.
The Psychology of Survey Response, 10th ed. Cam-
bridge, UK: Cambridge University Press, 2000.
uncles, m., A. s. c. ehrenBerg,
and K.
hAmmonD, K.
“Patterns of Buyer Behavior:
Regularities, Models, and Extensions.” Market-
ing Science 14, 3-2 (1995): G61–G70.
wAtts, D. J.
Everything is Obvious: Once You
Know the Answer. New York, NY: Crown Busi-
ness, 2011.
winD, Y.,
D. lerner.
“On the Measure-
ment of Purchase Data: Surveys Versus Pur-
chase Diaries.” Journal of Marketing Research 16,
February (1979): 39–47.
wright, m.,
D. esslemont.
“The Logi-
cal Limitations of Target Marketing.” Marketing
Bulletin 5, (1994): 133–120.
wright, m.,
m. mAcrAe.
”Bias and Vari-
ability in Purchase Intention Scales.” Journal
of the Academy of Marketing Science 35, 4 (2007):
wright, m.,
l. stocchi.
Temporal Stability of a Stochastic Model.” In
ANZMAC Conference Proceedings (2010),
... Overall, the better targeted and more attractive ads help online retailers reach consumers optimally and ultimately gain higher revenues, and the higher relevance and frictionless customer journeys are beneficial to customers as well (Malthouse et al., 2019). However, there are several critical voices that warn that targeting in general may lead to suboptimal spending, where customers who are already loyal are targeted (Nelson-Field et al., 2012;Sharp et al., 2009). This implies that retailers need to go beyond current fan-based targeting methods (e.g. ...
... Although targeting customers with the help of PA and AI is attractive due to cost-effectiveness, it may lead to negative outcomes. On the one hand, precision may entail a drastic drop in reach (Fulgoni, 2018;Nelson-Field et al., 2012). If only current customers receive the brand messages, the targeting may lead to stagnation of the customer base and reduce sales (Nelson-Field et al., 2012). ...
... On the one hand, precision may entail a drastic drop in reach (Fulgoni, 2018;Nelson-Field et al., 2012). If only current customers receive the brand messages, the targeting may lead to stagnation of the customer base and reduce sales (Nelson-Field et al., 2012). On the other hand, microtargeting that increasingly predicts personal needs and wants entails ethical challenges, both in restricting customer choice and in terms of (over-)collection, use and potential sharing of data. ...
Full-text available
Purpose Digital advertising enables retailers to rely on large volumes of data on consumers and even leverage artificial intelligence (AI) to target consumers online with personalised and context-aware advertisements. One recent example of such advertisements is programmatic advertising (PA), which is facilitated by automatic bidding systems. Given that retailers are expected to increase their use of PA in the future, further insights on the pros and cons of PA are required. This paper aims to enhance the understanding of the implications of PA use for retailers. Design/methodology/approach A theoretical overview is conducted that compares PA to traditional advertising, with an empirical investigation into consumer attitudes towards PA (an online survey of 189 consumers using an experimental design) and a research agenda. Findings Consumer attitudes towards PA are positively related to attitudes towards the retailer. Further, perceived ad relevance is positively related to attitudes towards PA, which is moderated by (1) consumer perceptions of risks related to sharing their data with retailers online and (2) consumer perceptions of AI's positive potential. Surprisingly, the disclosed use of AI for PA does not significantly influence consumer attitudes towards PA. Originality/value This paper contributes to the literature on technology-enabled services by empirically demonstrating that ad relevance drives consumer attitudes towards PA. This paper further examines two contingencies: risk beliefs related to data (i.e. the source of PA) and perceptions of AI (i.e. the somewhat nebulous technology associated with PA) as beneficial. A research agenda illuminates central topics to guide future research on PA in retailing.
... Facebook is recognized as a preferred platform to connect people across the globe and a widely used application that is identified as the most influential and commonly used medium for social purposes. Facebook has become an important tool for marketers to market and promote their products (Nelson-Field et al., 2012). People prefer to use Facebook because of its several useful features e.g., socializing with people, obtaining and sharing information, advertisement purpose, and comment posting (Mas'od et al., 2019). ...
... According to the above perspective, several studies have investigated the adoption of social media in general (e.g., Belk, 2013;Chen, 2015;Muntinga et al., 2011), whereas others have focused on the use of a specific platform, such as Facebook (e.g., Alhabash et al., 2014;Bellucci & Manetti, 2017;Giacomini et al., 2022;Hollenbaugh & Ferris, 2014;Nelson-Field et al., 2012;Rocca et al., 2021), YouTube (e.g., Liu-Thompkins, 2012;Pourazad et al., 2023), Twitter (e.g., Ceron et al., 2014;Colleoni, 2013;Etter et al., 2018;Liu et al., 2017) or Pinterest (e.g., Phillips et al., 2014). In this regard, it is noteworthy that social media have contributed to fuel greater environmental consciousness among consumers (Kleinrichert et al., 2012;Lu & Miller, 2019;Reilly & Hynan, 2014;Zahid et al., 2018). ...
Social media have surged prominently as communication channels for corporate social responsibility. However, little is still known about the performance of green versus non‐green communication across different social media. We contribute by examining whether the presence of green features in social media communication exerts a beneficial effect on consumer response in terms of likes, comments, and shares. We also investigate how this effect hinges upon the social media platform where the content is posted as well as the richness of the format (text, photos, videos) utilized for the diffusion. To our scopes, we use an ad hoc dataset of posts of two major large‐scale retailers in Italy across three major social media, namely Facebook, Instagram, and Twitter. Our results show that, while green content generally stimulates larger response than non‐green content, its effect varies across social media, with the highest effect being observed on Instagram (at least for likes) and the lowest on Twitter (at least for comments). Moreover, the extent to which the positive effect of green content increases as media richness increases (i.e., moving from only text to text plus photo, and then to text plus video) is also contingent upon the social media platform. On Facebook, the moderation of media richness is positive and significant, while being insignificant on Instagram. On Twitter, the moderation is even nonmonotonic in the sense that the highest (positive) effect of green content tends to be obtained for either low or high media richness. Our findings offer remarkable implications for firms engaging in environmental sustainability.
... Based on the rationale of the Meaning Transfer Model (McCracken, 1989), the actions of consumers on SNSs are associated with their endorsement of some particular content. For information related to the CSR activity of a company, clicking like, sharing the information or making a comment are behaviors that help consumers to communicate who they are and to establish relationships with similar peers (Nelson-Field et al., 2012). Therefore, this study proposes that when consumers are motivated to share CSR messages for self-enhancement, identity signaling and social bonding, their engagement with CSR communication in social media will be higher. ...
Full-text available
Purpose This study aims to analyze consumer motivations to share information about corporate social responsibility (CSR) activities through electronic word of mouth. It examines the roles of self-enhancement, identity signaling and social bonding as antecedents of consumers’ CSR engagement on social media. Design/methodology/approach A quantitative approach is used with a single-factor between-subjects experimental design in which the presence vs absence of CSR information on a company website is manipulated. The hypotheses are tested through structural equation modeling. Findings Results show that after viewing the company’s CSR message on its website, consumers who generated more CSR associations were more motivated to engage with the CSR information to satisfy fundamental personality traits (need for self-enhancement) and social relationship motivations (social bonding), which increased their intention to share the information. Research limitations/implications This study is restricted to CSR information on websites. Further research should consider what happens if such information is shared on social media, as consumers are more likely to spread CSR messages when they are shared by other public social network sites. Practical implications The study highlights the relevance of including CSR information on websites and offers insights into the importance of considering consumers in disseminating CSR information. Consumers share information when they have personal motivation for doing so. Social implications This study put the focus on the role of consumers in the diffusion of corporate information. Originality/value The results show the importance of personal motivations such as self-enhancement and social bonding in sharing CSR information on social media.
... By contrast, prior studies have found that buyers can be incentivized to continue shopping with offers of early and large discounts ( Creed et al. 2021 ;Dhar et al. 2007 ;Huber et al. 2008 ), thereby sustaining shopping momentum. Moser, Schoenebeck, and Resnick (2019) also note that marketing communications such as sales promotions in the form of price discounts or loyalty points can encourage browsing and shopping behavior. ...
... In a servicedominant (S-D) marketplace, consumers play a more active role in the consumer-brand interactions as value cocreators Lusch, 2004, 2008). By clicking the like button, users can become fans of a brand (Nelson-Field et al., 2012) and express their preference and love to that brand (Wallace et al., 2022). Feedback in the form of like-clicking could be incorporated into brands' product development projects or customer service processes and thus enhances brands' value. ...
Purpose From the perspectives of service-dominant logic and social identity theory, this study aims to assess social networking site (SNS) users’ likes as a form of social endorsement as well as its effects on like-clicking behavior, perceived brand value, customer-brand identification and purchase intention. Furthermore, the different effects of social endorsement on the perceived functional, hedonic, social and monetary brand value were investigated so as to support SNS users’ role as value cocreators. Design/methodology/approach An online survey was administered as a pretest of customer perceptions regarding brands that are liked on SNSs. Next, an experiment was conducted to verify the effects of social endorsement. A mixed-method approach including partial least squares (PLS) and fuzzy set qualitative comparative analysis (fsQCA) was adopted for the data analysis. Findings The results revealed that like-clicking behavior could be contagious because SNS users exposed to others’ likes were more likely to click the like button themselves. Like-clicking behavior positively influenced the perceived functional, hedonic, social and monetary value of the liked brand. Perceived brand value strengthened customer-brand identification, thereby increasing purchase intention. Originality/value Like-based social endorsements were confirmed as a type of value cocreation behavior that benefits the endorsed brand by spreading brand awareness, and increasing customer acquisition and retention. An fsQCA approach was developed to measure the moderating effect of users’ propensity to click the like button on perceived brand value, thus contributing to the advancement of fsQCA.
Purpose Social media is replete with malicious and unempathetic rhetoric yet few studies explain why these emotions are publicly dispersed. The purpose of the study is to investigate how the intergroup counter-empathic response called schadenfreude originates and how it prompts media consumption and engagement. Design/methodology/approach The study consists of two field surveys of 635 in-group members of two professional sports teams and 300 residents of California and Texas with political party affiliations. The analysis uses SEM quantitative methods. Findings Domain passion and group identification together determine the harmonious/obsessive tendencies of passion for an activity and explain the schadenfreude response toward the rival out-group. Group identification is a stronger driver of obsessive passion compared to harmonious passion. Schadenfreude directly influences the use of traditional media (TV, radio, domain websites), it triggers social media engagement (posting), and it accelerates harmonious passion's effects on social media posting. Research limitations/implications The study is limited by the groups used to evaluate the research model, sports, and politics. Social implications The more highly identified and passionate group members experience greater counter-empathy toward a rival. At extreme levels of group identification, obsessive passion increases at an increasing rate and may characterize extremism. Harboring feelings of schadenfreude toward the out-group prompts those with harmonious passion for an activity to more frequently engage on social media in unempathetic ways. Originality/value This study links the unempathetic, yet common emotion of schadenfreude with passion, intergroup dynamics, and media behavior.
Full-text available
Facebook is a social media network site that was introduced by a student at Harvard University in 2004, Mark Zuckerberg. In 2020, Facebook is still at the top of the list of the most popular social media networks in Malaysia, which is 91.7% of all Internet users in Malaysia. Facebook actually experienced a decrease in users this year from 2018. Despite experiencing a slight decrease in Internet users of Facebook since previous years, Facebook is still seen as an active social media platform in Malaysia. Facebook is the best social media for friends and acquaintances no matter old or new acquaintances. In addition, Facebook can deliver a wide area reach and a high frequency of sending messages to the target segment at a low cost. Because of these characteristics, that advertisers expand their social networks and are also the choice to market and deliver messages about their products. This paper will discuss the research conducted on the advertising of Sobella Cosmetics products on Facebook by looking in terms of information search, entertainment, social interaction and brand awareness. The research method used is quantitative, which is by distributing questionnaires online to respondents among Sobella product users to obtain information. The results of the study found that the majority of respondents agreed that Sobella ads on Facebook can provide information that helps them, in addition the ads also contain entertainment features, allowing them to socialize and interact with each other also with the founder of Sobella and last but not least increase their awareness of the Sobella brand.
While previous researchers have addressed motivations to join and continue using social media, this paper focuses on why users quit certain social media and change their favorite platforms, such as the current shift from Facebook to Twitter to Instagram and Snapchat. Furthermore, this exploratory study seeks to build an understanding of social media usage and motivations for switching from a cross-cultural perspective by comparing findings from Korean and U.S. users. Findings from 19 focus group sessions (n = 118) highlight influences regarding modes of usage, user control, commitment, addiction, privacy, perceived relationships, self-construals, and social/cultural trends. Findings are further analyzed and compared in light of relevant theoretical frameworks and cultural differences.
Full-text available
The authors link consumers’ consumption behavior as reported by a 24-hour recall method and two-day diaries for 39 different food groups. Both frequency of ingestion and average serving size are compared across both methods. The study is based on the current U.S. Department of Agriculture Nationwide Food Consumption Survey, spring quarter, including 7945 individuals. The data suggest that the 24-hour recall measurements did not differ significantly from the two-day diaries.
The first large-scale marketing study in which individual consumers’ survey responses could be linked to their panel diary recordings is reported. The results, for the margarine category only, indicate correspondence between the two data sets at the aggregate brand share level but great discrepancies at the individual consumer level. Analysis of this discrepancy calls into question the use of survey reports as an indicator of individual purchase in product positioning, segmentation, advertising media and copy research, and concept/product testing.
The authors link consumers' consumption behavior as reported by a 24-hour recall method and two-day diaries for 39 different food groups. Both frequency of ingestion and average serving size are compared across both methods. The study is based on the current U.S. Department of Agriculture Nationwide Food Consumption Survey, spring quarter, including 7945 individuals. The data suggest that the 24-hour recall measurements did not differ significantly from the two-day diaries.
Some marketers spend a great deal of time, money, and effort trying to define and target a brand's most valuable consumers among current buyers. "Less valuable" consumers, however, are essential to brand health and growth. A larger percent of profitable consumers among all brand buyers is characteristic of smaller rather than larger brands. Because the average consumer buys more as a brand's franchise grows, the next consumer the brand adds will be its most valuable. Brands need to target inclusively and stand for a vivid, clear but broadly appealing benefit. A narrow, exclusive focus on the "most profitable" households is a recipe for stagnation and decline, not for brand health.
In this article Mr Ehrenberg shows how data on purchases of non-durable consumer goods can be fitted by the negative binomial distribution, and discusses applications of this finding. He also considers a simple model for purchases made in different periods of time and some quick and easy methods for calculating standard errors.
The purpose of this research is to identify the key conceptual dimensions of product usage, and to develop reliable and valid measures of product usage. Two different methods (a self-report questionnaire and a diary study), two samples, and four consumer durables have been used to develop the measures of usage. The results suggest that usage frequency and usage variety are two critical dimensions of product usage, and that the measures developed in this study for each dimension have high convergent and discriminant validity. The study highlights the importance of investigating usage in the post-purchase context, and helps to identify issues for future research.
A method of analyzing the components of a trend in consumer purchasing is described. An empirically-based mathematical model is first used to predict the purchasing pattern in the absence of a trend. Comparison between the observed data and these predicted norms permits a detailed evaluation of the trend. Two examples of practical applications of the technique are presented.