Please write annotation as simple as you can You are doing three annotations and very easy to understand, use simple words please, because English is not my first language. , each of which should be no longer than 250 words MUST BE 250 WORS OR MAYBE LESS. and Each annotation should include What the purpose of the article? and What the issue being addressed, or alternatively ? What are the authors findings/ conclusions. Must be double-spaced using 12-pt Times New Roman font. The reference information should appear on top of each annotation. The annotations should include a summary, evaluation, a critique, and a statement of how the articles are related to one another

Thank you

Journal of Marketing Research
Vol. XLIX (February 2012), 83 –91
*Femke van Horen is a postdoctoral researcher, Social Psychology
Department, University of Cologne (e-mail: [email protected]). Rik
Pieters is Professor of Marketing, Marketing Department, and Fellow at
the Tilburg Institute of Behavioral and Economics Research (TIBER),
Tilburg University (e-mail: [email protected]). This article is based on
the first author’s dissertation, defended at Tilburg University. The authors
thank Claessens Product Consultants (Chris Huynen) in Hilversum, the
Netherlands, for developing the stimuli used in Study 2 and Luk Warlop
for helpful comments on a previous version of this article. Ziv Carmon
served as associate editor for this article.
FEMKE VAN HOREN and RIK PIETERS*
Copycats imitate features of leading brands to free ride on their equity.
The prevailing belief is that the more similar copycats are to the leader
brand, the more positive their evaluation is, and thus the more they free
ride. Three studies demonstrate when the reverse holds true: Moderatesimilarity
copycats are actually evaluated more positively than highsimilarity
copycats when evaluation takes place comparatively, such as
when the leader brand is present rather than absent. The results
demonstrate that blatant copycats can be less and subtle copycats can
be more perilous than is commonly believed. This finding has implications
for marketing theory and practice and trademark law.
Keywords: copycats, similarity, comparative evaluation, trademark
infringement
When High-Similarity Copycats Lose and
Moderate-Similarity Copycats Gain: The
Impact of Comparative Evaluation
© 2012, American Marketing Association
ISSN: 0022-2437 (print), 1547-7193 (electronic) 83
Imagine the following scenario: You are in a supermarket,
facing a shelf stacked with jars of peanut butter, and
you are ready to make a choice. Your eyes catch the leading
Skippy brand and the highly similar brand next to it: The jar
has a similar material and size, the same light-blue lid and
an identically colored blue label, and the same red lettering
and a similar name. How would you evaluate this “copycat”
brand?
Copycats imitate the name, logo, and/or package design
of a leading national brand to take advantage of the latter’s
positive associations and marketing efforts. When the copycat
is too similar to the leading brand, it is liable to trademark
infringement. A trademark is infringed when there is a
likelihood of confusion or when the copycat takes unfair
advantage of or is detrimental to the distinctive character or
reputation of the trademark (Trademarks Directive, Article
5). Copycatting practices are widespread. For example, a
survey of national U.S. supermarkets found that half the
store brands imitated a leader brand package at least in
color, size, and shape (Scott-Morton and Zettelmeyer 2004),
and trade loss due to trademark infringement was estimated
to be $512 billion, just in 2004 (Zaichkowsky 2006).
Marketing research and the legislative literature on trademark
infringement have emphasized the threats that highsimilarity
copycats pose to leader brands (Morrin and
Jacoby 2000; Zaichkowsky 2006). These threats are based
on the idea that the greater the similarity is between copycat
and leader, the higher is the likelihood of brand confusion,
and the more positive is consumers’ evaluation of the copycat
(Loken, Ross, and Hinkle 1986; Warlop and Alba 2004).
Thus, copycatting research has focused on demonstrating
potential brand confusion caused by high-similarity copycats
(Foxman, Muehling, and Berger 1990; Howard, Kerin,
and Gengler 2000; Kapferer 1995; Miaoulis and d’Amato
1978; Simonson 1994), and these are typically the cases
brought to court (Collins-Dodd and Zaichkowsky 1999;
Mitchell and Kearney 2002).
Gaining more insight into the determinants and effects of
brand confusion is important because researchers are only
just beginning to understand this phenomenon. Yet brand
imitation strategies are often subtler than the blatant, highsimilarity
cases that may lead to brand confusion and get
most attention (Planet Retail 2007, p. 52). Often, copycatting
brands have their own name, logo, and/or packaging
style but are still similar to the leader brand. Consumers
generally do not confuse such subtle, moderately similar
copies with the original (“This is not the leader brand X”)
and are aware of their true source (“It is a store brand,” or
“The brand is from manufacturer Y”; Wilke and
Zaichkowsky 1999).
The current research aims to demonstrate that even such
subtle copycats can free ride on the equity invested in the
leader brand. We also demonstrate that subtle copycats may
often be more effective than blatant copycats in their
attempt to leverage the positive associations surrounding
the leader brand. We test the idea that subtle copycats can
take (unfair) advantage of the leader brand, without brand
confusion taking place. In addition, we posit that blatant,
high-similarity copycats may actually be liked less than
subtle, moderately similar copycats, exactly because of the
former’s unashamed imitation strategy. Specifically, we
hypothesize that when the shopping situation enables consumers
to explicitly compare copycat brands with leader
brands, blatant copycats will lose and subtle copycats will
gain. Support for this idea would imply that copycat evaluation
critically depends on the degree of similarity between
copycat and leader brand (moderate vs. high) and on the
evaluation mode of consumers (comparative vs. noncomparative).
This would mean that high-similarity copycats
could be less perilous to leader brands than commonly
believed. High-similarity copycats are evaluated less positively
when direct comparisons with leader brands are
made, and the situations to do so are abound in daily shopping
situations. Moreover, high-similarity copycats are clear
litigation candidates. It would also mean that moderatesimilarity
copycats could be more menacing than generally
believed. They cue consumers to positive and familiar
attributes of the leader brand through similar, but not confusing,
package design, thus flying more easily under the
radar of trademark law.
Three controlled studies that systematically varied the
degree of similarity between copycat and leader brand and
the evaluation mode of consumers test our predictions. The
studies employed student and nonstudent samples, brand
name and brand package similarities, and a range of product
categories to ensure generalizability of the findings. The
next sections describe the conceptual framework on which
our predictions rest.
COPYCAT EVALUATION
Copycat brands try to free ride on the positive associations
that consumers have with a leader brand by having a
similar trade dress. Through their similarity in trade dress,
copycats try to access information that consumers have
stored in memory about another brand and to transfer it to
them. Research in social cognition has shown more generally
that the effect of such accessible information on the
evaluation of target (here, the copycat) can be assimilative
or contrastive.
Assimilation occurs when accessible information guides
the interpretation of target stimuli, causing a shift toward
the activated information (Stapel 2007). Thus, when the
positive leader brand information is interpreted and
included into the representation of the copycat, the lookalike
peanut butter brand from our opening example will be
evaluated more positively and closer to the leading Skippy
brand. Contrast occurs when information is used as a comparison
standard in evaluation, causing a shift away from
the accessible information (Herr 1989; Stapel, Koomen, and
Velthuijsen 1998). Thus, when the positively evaluated
Skippy brand is used as a comparison standard, the lookalike
peanut butter will be contrasted from the Skippy brand
and will be evaluated negatively. Then, the copycat pales in
comparison with the leader brand (Carpenter and Nakamato
1989).
What determines whether assimilation or contrast
occurs? That is, what determines whether references to a
leader brand help copycats and lead to more positive
evaluations or hurt them and lead to more negative evaluations?
We propose that both similarity and the way consumers
make their evaluation play a pivotal role. Evaluations
may take place in a noncomparative or a comparative
way (Oakley et al. 2008; Olsen 2002). Noncomparative
evaluations are evaluations made in the absence of an
explicit comparison. In that case, references to other brands
or products are lacking, and evaluation occurs in isolation
(“How do you like product X?”). In contrast, comparative
evaluations are made in the presence of a comparison standard:
The product is evaluated against a comparable other
product, and evaluation is guided by direct comparisons
(“How do you like product X compared with product Y?”).
In turn, these different evaluation modes influence attitudes,
purchase intentions, and behavior (Hsee 1996; Hsee and
Zhang 2010; Nowlis and Simonson 1997).
We posit that the effect of higher and lower degrees of
similarity on copycat evaluation is contingent on consumers’
evaluation mode. When copycat evaluation takes
place noncomparatively and no explicit references are made
to the imitated leader brand, evaluation is likely to be
guided by the activation of positive feelings, associations,
and attitudes surrounding the leader brand. As a consequence,
higher degrees of similarity should be evaluated
more positively because a higher resemblance of the copycat
brand with the leader brand will activate more positive
associations. These associations will readily “spill over”
(Murphy and Zajonc 1993) and be “included” (Schwarz and
Bless 1992) into the representation of the copycat. Therefore,
noncomparative evaluation is likely to be more beneficial
to high-similarity copycats than to moderate- and lowsimilarity
copycats.
High-similarity copycats may, however, be evaluated
quite differently when comparative evaluation takes place.
When the copycat is explicitly evaluated against the imitated
leader brand and similarity is high, consumers are
especially likely to become more aware of the high resemblance
with the leader brand and will sense that their judgments
might be biased by positive feelings derived from the
leader brand. In that case, they are likely to adjust their
response to the effect of this influence by consulting their
naive theories of persuasion. Such theories include beliefs
about marketers’ influence strategies and tactics (Boush,
Friestad, and Wright 2009), including how they entice consumers
to buy a product through similarity.
Persuasion knowledge is particularly likely to be used
when the situation makes a marketer’s potential ulterior
motives accessible (Campbell and Kirmani 2000), which we
believe is the case when the situation facilitates comparative
evaluation. Thus, when similarity is high and evalua-
84 JOURNAL OF MARKETING RESEARCH, FEBRUARY 2012
Impact of Comparative Evaluation 85
tion is comparative, the awareness of possible insincere persuasion
tactics negates or prevents positive associations
because of image transfer from the leader brand. Consumers
realize that the copycat is trying to leverage the reputation
of the leader brand through imitation, which results in contrast.
On the basis of this reasoning, we propose that comparative
evaluations are likely to hurt rather than help highsimilarity
copycats.
Conversely, we predict that the evaluation of moderatesimilarity
copycats is unaffected by comparative evaluation.
Even when moderate-similarity copycats are directly compared
with the leader brand, little awareness of ulterior
motives will exist because of their moderate degree of similarity
(Warlop and Alba 2004). However, because moderately
similar copycats do show resemblance, they should
still profit from the positive associations with the leader
brand. This suggests that when evaluation is comparative,
moderate-similarity copycats still benefit from being similar
but high-similarity copycats do not.
Our prediction that moderate similarity is evaluated more
positively than high and low similarity is comparable to
Meyers-Levy and Tybout’s (1989) findings in the domain of
new product evaluation. Their results indicate that new
products that are moderately congruent with the existing
product schema are more positively evaluated than either
completely congruent or extremely incongruent products.
Yet the process that we believe accounts for copycat evaluation
also leads to different predictions. That is, under comparative
evaluation, moderate copycats should be evaluated
more positively than both high- and low-similarity copycats.
Then, leader brands act as comparison standards, and
high-similarity copycats are contrasted away from them.
However, under noncomparative evaluation, high-similarity
copycats should be evaluated most positively because their
evaluation is assimilated toward the leader brand. The question
then becomes when consumers adopt a comparative
evaluation mode.
COMPARATIVE EVALUATION MODE
A comparative evaluation mode may be activated by various
shopping situation characteristics. For example, it might
be induced when descriptions of products in advertisements
or in-store displays are alignable and described on the same
dimension (e.g., the preparation times for popcorn A and B),
rather than when they are nonalignable and described on
different dimensions (e.g., preparation time for popcorn A
and sweetness for popcorn B; Markman and Loewenstein
2010). The categorization of product assortments might also
determine the evaluation mode. Comparisons may be
encouraged when consumers are exposed to narrow rather
than broad categories (Ülkümen, Chakravarti, and Morwitz
2010) or to organized rather than disorganized categories
(Kahn and Wansink 2004; Mogilner, Rudnick, and Iyengar
2008) or when the layout of the store is mostly feature based
rather than benefit based (Poynor and Diehl 2010).
Here, we investigate a determinant of the evaluation
mode that is under direct managerial control: product display
at the point of purchase. Prior research has shown that
product display influences consumers’ preferences and purchase
likelihoods (Buchanan, Simmons, and Bickart 1999;
Simonson 1999; Simonson, Nowlis, and Lemon 1993).
Whether products are displayed side-by-side or farther apart
at the point of purchase is particularly likely to influence the
evaluation mode. Two products that are presented side-byside
are more readily and more spontaneously compared
with each other than when they are presented in isolation
(Hsee 1996; Muthukrishnan and Ramaswami 1999). Thus,
when the copycat and leader brand are placed next to each
other on the store shelf and are in the same visual field during
copycat evaluation, comparative evaluation is more
likely to occur. When the copycat and leader brand are
placed on different shelves, in different parts of the store, or
even in different stores, direct references to the leader brand
are absent, and noncomparative evaluation is more likely to
take place.
THE CURRENT RESEARCH
Three studies test the idea that the appraisal of copycats
critically depends on brand similarity and on consumers’
evaluation mode. Study 1 establishes the critical role of
evaluation mode in copycat evaluation. It tests whether
high-similarity copycats are indeed evaluated more positively
than low-similarity copycats, when evaluation is noncomparative
but not when it is comparative. We test this
prediction by manipulating the specific characteristics of
the shopping environment (absence vs. presence of the
leader brand, Study 1a) and by more directly inducing a
noncomparative versus comparative evaluation mode
(Study 1b).
Studies 2 and 3 examine more closely whether moderatesimilarity
copycats actually gain when evaluation is comparative.
Specifically, using a student sample (Study 2) and
a representative consumer sample (Study 3), we test
whether high-similarity copycats are liked more than lowand
moderate-similarity copycats when the leader brand is
absent during evaluation but whether moderate-similarity
copycats are liked more than low- and high-similarity copycats
when the leader brand is present during evaluation. In
all studies, leader brands and copycat brands differ only in
the similarity of their trade dress, that is, brand name (Studies
1a, 1b, and 3) and package design (Study 2).
STUDY 1
Study 1 investigates high versus low degrees of similarity
between a copycat brand and the leader brand. It tests
whether high degrees of similarity are evaluated more positively
when evaluation is noncomparative but less positively
when evaluation is comparative. Study 1a tests
whether the physical arrangement of brands (absence or
presence of the leader brand) influences copycat evaluation.
Study 1b induces comparative and noncomparative evaluation
modes directly in consumers, instead of manipulating
these modes indirectly through the characteristics of the
shopping situation. Both studies use brand names because
cases of trademark infringement often deal with these (see,
e.g., www.darts-ip.com).
Study 1a
Method. Fifty paid students (30 men and 20 women;
Mage = 20.30, SD = 2.22) participated in a 2 (similarity: low,
high) ¥ 2 (brand presence: absent, present) mixed design,
with similarity as a within-subject factor and brand presence
as a between-subjects factor. The target product category
was “olive oil,” with Bertolli as the leader brand. This brand
has the highest market share, a strong reputation, and a distinctive
name. Each participant evaluated six brand names
that were either low or high in similarity to the leader brand.
We used the following procedure to select the copycat
brand names: First, we created a pool of 30 brand names
that systematically differed in similarity to the Bertolli
brand name. Because brand name similarity in court cases
is typically assessed by comparing the number of identical
letters and their placement in the name (e.g., Bastolli is
more similar to Bertolli than Lucini), we used that as a basis
for name creation. From a pretest (N = 45), 6 of the 30
brand names (3 names per similarity condition) were
selected that did not differ in attractiveness (seven-point
scales, F < 1). A second pretest (N = 40) established that the
degree of similarity was successfully manipulated: Lowsimilarity
brand names were rated as less similar to the
leader brand Bertolli (seven-point scales) than the highsimilarity
brand names (t(39) = –12.21, p < .001). Lowsimilarity
brand names were, respectively, “Lucini,” “Santini,”
and “Malzani.” High-similarity brand names were,
respectively, “Vintolli,” “Bastolli,” and “Bertino.”
In the main study, participants were told that a new olive
oil, which did not have a name yet, would enter the market
soon. They were asked to imagine being in a grocery store
and considering buying olive oil with the indicated brand
name. Next, the six brand names appeared on the computer
screen in random order. In the “leader absent” condition,
only the new brand name appeared on the screen. In the
“leader present” condition, the new brand name and the
leader brand name (Bertolli) appeared. Participants evaluated
each name on a nine-point scale (1 = “negative,” and 9 =
“positive”). Afterward, they indicated their familiarity with
the leader brand (1 = “not familiar at all,” and 7 = “highly
familiar”), brand importance (1 = “not important at all,” and
7 = “very important”), and their evaluation of the leader
brand (1 = “negative,” and 9 = “positive”).
Results. A 2 (similarity) ¥ 2 (brand presence) repeated
measures analysis of variance (ANOVA) gave no significant
main effect of brand similarity (F(1, 48) = .04, p = .84, p2 =
.001) or brand presence (F(1, 48) = 2.85, p = .10, p2 = .06)
but revealed the predicted interaction between similarity
and brand presence (F(1, 48) = 29.85, p < .001, p2 = .38;
see Figure 1). Planned contrasts showed that participants
indeed evaluated the high-similarity brand names more
positively (M = 6.13, SD = 1.31) than the low-similarity
brand names (M = 4.80, SD = 1.00) when the leader brand
was absent (F(1, 48) = 13.80, p < .001, p2 = .22). However,
when the leader brand was present, participants evaluated
the high-similarity brand names less positively (M = 4.37,
SD = 1.40) than the low-similarity brand names (M = 5.81,
SD = 1.00; F(1, 48) = 16.10, p < .001, p2 = .25). Within
similarity conditions, planned contrasts also revealed that
participants evaluated the high-similarity brand names more
negatively when the leader brand was present than when it
was absent (F(1, 48) = 21.10, p < .001, p2 = .31), while the
opposite was true for low-similarity brand names (F(1, 48) =
12.83, p = .001, p2 = .21). None of the control variables
affected copycat evaluation (all Fs < 1).
These results provide evidence that evaluation mode
(comparative vs. noncomparative) critically determines the
effect of brand similarity on copycat evaluation. Copycats
benefit from high resemblance when the brand they imitate
is absent but not when it is present. Study 1b tests whether
it is indeed evaluation mode that moderates, as we predict,
the effect of similarity on copycat evaluation, by inducing a
noncomparative versus comparative evaluation mode directly.
Study 1b
Method. Forty-eight paid students (28 men and 20
women; Mage = 21.12, SD = 2.56) were randomly allocated
to a condition of a 2 (similarity: low, high) ¥ 2 (evaluation
mode: noncomparative, comparative) mixed design, with
similarity as a within-subject factor and evaluation mode as
a between-subjects factor. Brand names were the same as in
Study 1a.
Participants were asked to imagine themselves being in a
grocery store in front of the shelf and considering buying a
particular product. In line with Oakley et al.’s (2008) and
Olsen’s (2002) description of evaluation mode, participants
in the “comparative mode” condition were asked to imagine
making a decision based on a comparison of the brand with
another brand that would immediately come to their mind.
To come to a decision, they were asked to consider how
much they liked the new brand as compared with this other
brand. In the “noncomparative mode” condition, participants
were asked to imagine making a decision based on
their general impression about the brand. To come to a decision,
they were asked to consider how much they liked the
brand on the basis of this impression.
After participants visualized the specific decision-making
process, they followed the same procedure as in Study 1a.
After evaluating the brand names, they indicated their willingness
to buy an olive oil with this new brand name (1 =
“definitely not,” and 9 = “definitely yes”). The control
variables were the same as in Study 1a. Finally, as a
manipulation check, participants indicated whether they
were asked to imagine making decisions based on their general
impression of the new brand or by comparing it with
another brand (all were correct). Participants also indicated
how easily they could bring the specific decision strategy to
mind (1 = “not at all,” and 9 = “very well”). Four partici-
86 JOURNAL OF MARKETING RESEARCH, FEBRUARY 2012
Figure 1
STUDY 1A: BRAND SIMILARITY AND LEADER BRAND
PRESENCE INFLUENCE COPYCAT EVALUATION
7
6
5
4
3
2
Copycat Evaluation
Low Similarity High Similarity
! Leader brand absent Leader brand present
!
Notes: Scale ranges from 1 (“low”) to 9 (“high”). Error bars indicate ±1
standard error of the mean.
Impact of Comparative Evaluation 87
pants who could not imagine the specific decision strategy
were dropped, which left 44 participants for the analyses.
Because the measures of evaluation and willingness were
highly correlated (rs > .83), we collapsed them into a single
evaluation measure.
Results. A 2 (similarity) ¥ 2 (mode) repeated measures
ANOVA revealed a significant effect of evaluation mode
(F(1, 42) = 5.68, p = .02, p2 = .12), an effect of similarity
(F(1, 42) = 19.28, p < .001, p2 = .32), and, more important,
the expected interaction between similarity and evaluation
mode (F(1, 42) = 4.15, p = .05, p2 = .09; see Figure 2).
Planned contrasts showed that, consistent with the results of
Study 1a, participants evaluated the high-similarity brand
names more positively (M = 6.14, SD = 1.14) than the lowsimilarity
brand names (M = 4.76, SD = .94) when a noncomparative
evaluation mode was activated (F(1, 42) =
20.67, p < .001, p2 = .33). In addition, participants evaluated
the high-similarity brand names as positively (M =
5.05, SD = 1.30) as the low-similarity brand names (M =
4.55, SD = 1.20; F(1, 42) = 2.77, p = .11, p2 = .05) when a
comparative evaluation mode was activated. Within similarity
conditions, as we predicted, participants evaluated the
high-similarity brand names more negatively when a comparative
evaluation mode was induced than when a noncomparative
mode was induced (F(1, 42) = 8.71, p = .01,
p2 = .17), while there was no difference in evaluation for
the low-similarity brand names (F(1, 42) = .43, p = .52, p2 =
.01). Again, the control variables did not change the results
when entered as covariates (all Fs < 1).
Together, Studies 1a and 1b show that the effect of similarity
on copycat evaluation is context dependent. Thus, it is
not the case that high-similarity copycats are uniformly liked
more than low-similarity copycats. Rather, high-similarity
copycats are liked more when evaluation is noncomparative
and the decision-making process is guided by the overall
impression of the copycat, which occurs, for example, when
the leader brand is absent. Then, positive associations with
the leader brand are transferred to the copycat, resulting in
more positive evaluations of high-similarity copycats. However,
when evaluation is comparative—for example, when
the leader brand is present or when consumers come to the
decision in a comparative evaluation mode—the leader
brand acts as a comparative standard, and high-similarity
copycats are liked less.
STUDY 2
One implication of our theorizing is that high-similarity
copycats are affected more by the evaluation mode than
moderate- and low-similarity copycats. Whereas highsimilarity
copycats gain under noncomparative evaluation
mode because of assimilation, they lose under a comparative
evaluation mode because of contrast. Because moderatesimilarity
copycats are much less affected by the evaluation
mode but still gain from being similar, they are evaluated
more positively than high-similarity copycats under a comparative
evaluation mode. Low-similarity copycats score
the lowest. Study 2 tests this prediction, using package
designs to generalize the findings to important features of
the trade dress of brands beyond the name.
Method
Sixty-five paid students (41 men and 24 women; Mage =
20.1, SD = 2.44) were randomly assigned to a condition of
a 3 (similarity: low, moderate, high) ¥ 2 (brand presence:
absent, present) mixed design, with similarity as a withinsubject
factor and brand presence as a between-subjects factor.
We counterbalanced presentation order to rule out order
effects.
A professional design company created images of three
packages of fictitious brands in the product category
“Spreadable butter with olive oil.” Packages varied in
degree of similarity to the package of Bertolli, the leading
brand within the product category (see Figure 3). A pretest
(N = 45) established that the three packages did not differ in
overall attractiveness (as in Study 1, F < 1). They did differ
in degree of similarity to the leader brand, as measured on a
0–100 scale, with higher numbers reflecting greater similarity.
A repeated measures ANOVA revealed that participants
perceived the high-similarity copycat as more similar to the
leader brand (80.89) than the moderate-similarity (58.30)
and low-similarity (32.07) copycats (F(2, 88) = 163.28, p <
.001, all contrasts ps < .05).
Participants were instructed to evaluate new products in
the category spreadable butter with olive oil. Manipulation
of the “leader brand absent” and “leader brand present” conditions
was the same as in Study 1a. Participants saw the
packages of the three copycats for ten seconds. Then, they
evaluated the three copycats on four semantic differentials
with seven-point response scales (“bad/good,” “unattractive/
attractive,” “uninteresting/interesting,” and “negative/ positive”;
aggregated scale  = .90). Control variables were the
same as previously.
Results
A 3 (similarity) ¥ 2 (brand presence) repeated measures
ANOVA revealed no effect of similarity (F(2, 126) = 1.27, p =
.28, p2 = .02) but an effect of brand presence (F(1, 63) =
24.50, p < .001, p2 = .28). This effect was qualified by the
predicted interaction between brand similarity and brand
presence (F(2, 126) = 4.18, p = .02, p2 = .06; see Figure 4).
For the leader brand absent condition, planned contrasts
showed that participants evaluated the high-similarity copycat
Figure 2
STUDY 1B: BRAND SIMILARITY AND EVALUATION MODE
INFLUENCE COPYCAT EVALUATION
7
6
5
4
3
Copycat Evaluation
Low Similarity High Similarity
! Noncomparative Comparative
!
Notes: Scale ranges from 1 (“low”) to 9 (“high”). Error bars indicate ±1
standard error of the mean.
marginally more positively (M = 4.96, SD = 1.08) than the
moderate-similarity copycat (M= 4.55, SD = 1.06; F(1, 63) =
3.03, p = .09, p2 = .05) and, in turn, evaluated the moderatesimilarity
copycat as positively as the low-similarity copycat
(M = 4.50, SD = 1.09; F(1, 63) = .25, p = .87, p2 = .00). The
results were quite different in the leader brand present condition.
There, participants evaluated the moderate-similarity
copycat more positively (M = 4.30, SD = 1.06) than, respectively,
the low-similarity copycat (M = 3.69, SD = 1.40;
F(1, 63) = 4.36, p = .04, p2 = .07) and the high-similarity
copycat (M= 3.53, SD = 1.28; F(1, 63) = 9.55, p = .001, p2 =
.13). An additional planned contrast showed that in the leader
brand absent condition, participants indeed evaluated the
high-similarity brand more positively (F(1, 33) = 3.96, p =
.05, p2 = .11) than the low- and moderate-similarity brands
taken together. A planned comparison showed that in the
leader brand present condition, participants evaluated the
moderate brand more positively (F(1, 30) = 10.45, p = .003,
p2 = .26) than the low- and high-similarity brands taken
together.
Within conditions, as we expected, planned contrasts
showed that participants evaluated the high-similarity copycat
more negatively in the leader brand present condition
than in the leader brand absent condition (F(1, 63) = 23.81,
p < .001, p2 = .27), whereas the moderate-similarity copycat
was unaffected by brand presence: Participants evaluated
it as positively in the leader brand present condition as
in the leader brand absent condition (F(1, 63) = .93, p = .34,
p2 = .02). None of the control variables had an effect on
evaluation (all Fs < 1).
Taken together, Studies 1 and 2 show that when the leader
brand is absent, copycats with higher degrees of similarity
are evaluated more positively. However, when the leader
brand is present, moderate-similarity copycats are evaluated
more positively than both high- and low-similarity copycats.
When comparative evaluation is activated, for example,
through leader brand presence, the evaluation of highsimilarity
copycats suffers, whereas the evaluation of
moderate-similarity copycats is unaffected. Compared with
low-similarity copycats, moderate-similarity copycats benefit
from assimilation to the leader brands’ positive associations
and, at the same time, do not suffer from the contrast
effect caused by very high degrees of similarity.
STUDY 3
We conducted Study 3 to establish that the findings
obtained thus far are not limited to student samples and a
single product category and leader brand. It aims to generalize
the results to regular (nonstudent) consumers, other
product categories, and other brands.
Method
A nationally representative sample of 542 members of a
household Internet panel participated. Panel members
receive questionnaires electronically, complete them at
home on their personal computers, and then return them.
Participants (281 men and 261 women; Mage = 43.22, SD =
10.44) were randomly assigned to a condition of a 3 (similarity:
low, moderate, high) ¥ 2 (presence of leader brand:
absent, present) ¥ 2 (product category: chocolate spread,
French cream cheese) mixed design, with similarity as a
within-subject factor and product category and brand pres-
88 JOURNAL OF MARKETING RESEARCH, FEBRUARY 2012
Figure 3
STUDY 2: BRAND STIMULI
A: Leader Brand
B: High-Similarity Brand
C: Moderate-Similarity Brand
D: Low-Similarity Brand
Figure 4
STUDY 2: BRAND SIMILARITY AND LEADER BRAND
PRESENCE INFLUENCE COPYCAT EVALUATION
6
5
4
3
2
Copycat Evaluation
Similarity
Low Moderate High
! Leader brand absent Leader brand present
!
Notes: Scale ranges from 1 (“low”) to 7 (“high”). Error bars indicate ±1
standard error of the mean.
Impact of Comparative Evaluation 89
ence as between-subjects factors. We counterbalanced order
of brand names to rule out order effects.
The stimuli were nine brand names in the product category
“chocolate spread” and “French cream cheese,” differing
in degree of similarity with the leader brand names,
respectively, Nutella and Paturain. We used the same procedure
as in Study 1 to create brand names with low, moderate,
and high degrees of similarity. Respective examples of
the brand names were “Valina” (low similarity), “Notina”
(moderate similarity), and “Latella” (high similarity) for the
leader brand Nutella, and “Racorin” (low similarity),
“Romatain” (moderate similarity), and “Pitorain” (high
similarity) for the leader brand Paturain. Separate pretests
revealed that the nine selected brand names within each
product category (three brand names per similarity condition)
did not differ in attractiveness (as in Studies 1 and 2,
all Fs < 1) but did differ in degree of similarity to the leader
brand (chocolate spread: F(2, 106) = 408.24, p < .001;
French cream cheese: F(2, 90) = 196.00, p < .001).
Instructions, manipulations, and measures were the same
as in Study 1a. After participants read the scenario, they
evaluated nine brand names and indicated their willingness
to buy a product with this brand name for either a new
chocolate spread or a new French cream cheese, while only
the new brand name or the new brand name and the leader
brand name appeared on the screen. Finally, we collected
sociodemographic information, such as participants’ age,
income, education level, and household size.
Results
Copycat evaluation. A 3 (similarity) ¥ 2 (presence) ¥ 2
(product) repeated measures ANOVA revealed a main effect
of similarity (F(2, 1076) = 162.64, p < .001, p2 = .23), presence
of the leader brand (F(1, 538) = 7.38, p = .007, p2 =
.01), and product category (F(1, 538) = 6.14, p = .01, p2 =
.01). More important, the interaction between the presence
of the leader brand and brand similarity was significant
(F(2, 1076) = 19.38, p < .001, p2 = .04; see Figure 5). None
of the other interactions were significant (all Fs < 1).
Because there were no significant interactions between
product category and any of the other variables, we further
analyzed the results across the collapsed product categories.
Planned contrasts showed that when the leader brand was
absent, participants evaluated the high-similarity copycat
more positively (M = 4.73, SD = 1.63) than the moderatesimilarity
copycat (M = 4.39, SD = 1.54; F(1, 540) = 15.07,
p < .001, p2 = .03) and, in turn, evaluated the moderatesimilarity
copycat more positively than the low-similarity
copycat (M = 3.31, SD = 1.30; F(1, 540) = 138.45, p < .001,
p2 = .20). When the leader brand was present, the results
showed, consistent with our prediction, that participants
evaluated the moderate-similarity copycat significantly
more positively (M = 4.34, SD = 1.43) than, respectively,
the high-similarity copycat (M= 3.99, SD = 1.63; F(1, 540) =
14.31, p < .001, p2 = .03) and the low-similarity copycat
(M = 3.34, SD = 1.33; F(1, 540) = 104.32, p < .001, p2 =
.16). Additional planned contrasts showed that in the leader
brand absent condition, participants indeed evaluated the
high-similarity brand more positively (F(1, 288) = 137.41, p <
.001, p2 = .32) than the low- and moderate-similarity
brands taken together. In the leader brand present condition,
participants evaluated the moderate brand more positively
(F(1, 252) = 68.24, p < .001, p2 = .21) than the low- and
high-similarity brands taken together.
Within conditions, planned contrasts showed that, as we
expected, participants evaluated the high-similarity copycat
more negatively in the leader brand present condition than
in the leader brand absent condition (F(1, 540) = 28.20, p <
.001, p2 = .05), whereas they evaluated the moderate copycat
equally positively in the leader brand present and leader
brand absent conditions (F(1, 540) = .17, p = .68, p2 = .00).
Willingness to buy. A 2 (product) ¥ 2 (presence) ¥ 3
(similarity) repeated measures ANOVA revealed a similar
pattern of results for the willingness-to-buy measure as for
the evaluation measure. As we hypothesized, there was an
interaction between the presence of the leader brand and
brand similarity (F(2, 1076) = 18.38, p < .001, p2 = .03; for
the leader brand absent condition: MLow = 3.21, SD = 1.27;
MMod = 4.28, SD = 1.49; MHigh = 4.59, SD = 1.67; for the
leader brand present condition: MLow = 3.30, SD = 1.39;
MMod = 4.20, SD = 1.52; MHigh = 3.94, SD = 1.63; all
planned contrast at ps < .001).
Control variables. For both product categories, the level
of education, work/daily activity, family status, household
size, and income did not affect evaluations; however, the
age of the participants did. When we included age as a
covariate, the critical similarity ¥ brand presence interaction
(F(2, 1070) = 3.16, p = .05, p2 = .006) remained significant,
but a significant interaction emerged between similarity
and age (F(2, 1070) = 3.59, p = .03, p2 = .007). Simple
slope analysis revealed that older participants rated the
moderate-similarity copycat more negatively than younger
participants ( = –.018, p = .01), whereas the evaluation of
the low-similarity ( = –.004, p = .51) and high-similarity
( = .00, p = .98) copycats was unaffected by age. The crucial
interaction between similarity and leader brand presence,
however, remained intact even with this age effect,
which is desirable.
These findings provide additional support for the idea
that highly similar copycats may be less hazardous than
moderately similar copycats when sold in the context of the
original brand. However, if the highly similar brands are
Figure 5
STUDY 3: BRAND SIMILARITY AND LEADER BRAND PRESENCE
INFLUENCE COPYCAT EVALUATION (COMBINED DATA)
5
4
3
2
Copycat Evaluation
Similarity
Low Moderate High
! Leader brand absent Leader brand present
!
Notes: Scale ranges from 1 (“low”) to 9 (“high”). Error bars indicate ±1
standard error of the mean.
offered in stores that do not carry the leading brand, they are
evaluated more positively.
GENERAL DISCUSSION
Copycats take unfair advantage of the reputation of
leader brands. As one brand owner formulated, “We invest a
lot and take a lot of risk creating brands and images people
want. Those who coat tail on this unfairly are ripping us
off ” (Intangible Business n.d.). The marketing literature has
typically emphasized the threats of high-similarity copycats,
based on the extent to which copycats are more similar to
the leader brand. The current research shows that the
appraisal of copycats, in addition to the degree of brand
similarity, is critically dependent on consumers’ evaluation
mode. The liking of copycats increased uniformly with
higher degrees of similarity, but only when evaluation was
noncomparative. In contrast, when evaluation was comparative
and the copycat was explicitly compared with the
leader brand, high-similarity copycats were evaluated less
positively than moderate-similarity copycats. A comparative
or noncomparative evaluation mode can be activated
directly in consumers or can be brought about by small
changes in the physical arrangement of products (i.e., by the
absence or presence of the leader brand). Three studies
using samples of students and typical consumers, textual
(brand names) and visual (package designs) brand trademarks,
and various product categories and leader brands
established the robustness of the findings. Together, the findings
provide evidence that high-similarity copycats may be
ripping off the leader brands less and that moderate-similarity
copycats may be more ominous than commonly believed.
This research provides evidence that subtler, moderatesimilarity
copycats can be more threatening than blatant,
high-similarity ones. This is important because the free riding
of subtle copycats remained undetected by consumers,
even under comparative evaluation. Moderate-similarity
copycats may also fly under the radar of trademark legislation
because they have similar but not confusing package
designs, and it is brand confusion on which trademark legislation
focuses. A small survey that we conducted among 28
trademark lawyers at a national conference on trademark
law in the Netherlands supports this idea. We asked the
lawyers to indicate which copycats from Study 2 they would
bring a lawsuit against, and they could select one, two, or
all three copycats. All lawyers chose the high-similarity
copycat, but only 8 of the 28 lawyers would also bring a
lawsuit against the moderate-similarity copycat. This illustrates
that despite a recent court ruling declaring that free
riding on a brand’s reputation is unacceptable even without
a likelihood of confusion (L’Oreal S.A. v. Bellure N.V.
2009), subtle copycats can profit from imitation without
being prosecuted.
Previous research has demonstrated that the more similar
copycats are to the leader brand, the more positively they
are evaluated (Loken, Ross, and Hinkle 1986; Warlop and
Alba 2004). The current findings show, however, the crucial
role of evaluation mode: When evaluation is comparative,
high-similarity copycats are evaluated more negatively,
rather than more positively, than moderate-similarity copycats.
Evaluation mode was manipulated either directly or
indirectly by leader brand presence or absence. Other characteristics
of the shopping environment under managerial
control may also influence the likelihood that consumers
engage in comparison shopping—for example, when assortments
are organized rather than disorganized (Kahn and
Wansink 2004) or when assortment organization is more
attribute based than benefit based (Poyner and Diehl 2010).
In addition, leader brands that are salient in memory
because they are pioneers in the product category or have a
high market share (Carpenter and Nakamato 1989; Robinson
and Fornell 1985) may automatically be brought to
mind and be used as a comparison standard, resulting in
contrast when the copycat is highly similar.
This study indicates how copycats can gain or lose from
their resemblance to the leader brand, but it remains silent
on how copycat similarity affects the reputation and sales of
the leader brand. Trademark dilution and slipping sales of
leader brands due to copycatting practices have been documented
(Morrin and Jacoby 2000; Morrin, Lee, and Allenby
2006; Steenkamp, Van Heerde, and Geyskens 2010), but we
are not aware of research on potential gains in brand equity
and sales for leader brands when copycats are contrasted
away from them. Our findings hint at the possibility that the
evaluation of leader brands might unexpectedly gain from
copycatting practices in particular when copycat imitation is
blatant and occurs in the presence of the leader brand. Then,
blatant copycatting may highlight the qualities of the leader
brand. Follow-up research should test these speculations.
We used established procedures to systematically vary
the degrees of (verbal) brand name copycatting but needed
to resort to pretesting to determine the degrees of (visual)
package design copycatting, which is a limitation. To
advance marketing science and trademark law, it is paramount
to develop strong theories and methodologies to
establish degrees of similarity between package designs and
between other visual marketing stimuli. Such theories and
methodologies would be crucial tools for managers and
lawyers who need to determine the degrees of visual similarity
between the leader brand and copycats unequivocally
and perhaps even a priori. Thus, we call on researchers to
develop these theories and methodologies. Doing so would
extend our finding that subtle copycats can “coattail” more
than blatant copycats on the leader brand’s equity and
would determine more generally which degree of similarity
has the most effect.
REFERENCES
Boush, David M., Marian Friestad, and Peter Wright (2009),
Deception in the Marketplace: The Psychology of Deceptive
Persuasion and Consumer Self-Protection. New York: Routledge/
Taylor & Francis.
Buchanan, Lauranne, Carolyn J. Simmons, and Barbara A. Bickart
(1999), “Brand Equity Dilution: Retailer Display and Context
Brand Effects,” Journal of Marketing Research, 36 (August),
345–55.
Campbell, Margaret C. and Amna Kirmani (2000), “Consumers’
Use of Persuasion Knowledge: The Effects of Accessibility and
Cognitive Capacity on Perceptions of an Influence Agent,”
Journal of Consumer Research, 27 (1), 69–83.
Carpenter, Gregory S. and Kent Nakamoto (1989), “Consumer
Preference Formation and Pioneering Advantage,” Journal of
Marketing Research, 26 (August), 285–98.
Collins-Dodd, Colleen and Judy L. Zaichkowsky (1999),
“National Brand Responses to Brand Imitation: Retailers Versus
Other Manufacturers,” Journal of Product and Brand Management,
8 (2), 96–105.
90 JOURNAL OF MARKETING RESEARCH, FEBRUARY 2012
Impact of Comparative Evaluation 91
Foxman, Ellen R., Darrel D. Muehling, and Phil W. Berger (1990),
“An Investigation of Factors Contributing to Consumer Brand
Confusion,” Journal of Consumer Affairs, 24 (1), 170–89.
Herr, Paul M. (1989), “Priming Price: Prior Knowledge and Context
Effects,” Journal of Consumer Research, 16 (1), 67–75.
Howard, Daniel J., Roger A. Kerin, and Charles Gengler (2000),
“The Effects of Brand Name Similarity on Brand Source Confusion:
Implications for Trademark Infringement,” Journal of
Public Policy & Marketing, 19 (Fall), 250–64.
Hsee, Chrisopher K. (1996), “The Evaluability Hypothesis: An
Explanation for Preference Reversals Between Joint and Separate
Evaluations of Alternatives,” Organizational Behavior and
Human Decision Processes, 67 (3), 247–57.
——— and Jiao Zhang (2010), “General Evaluability Theory,” Perspectives
on Psychological Science, 5 (4), 343–55.
Intangible Business (n.d.), “Copying: Fair or Unfair?” research
report, (accessed January 2011), [available at http://www.intangiblebusiness.
com/Reports/Copying-Fair-or-Unfair~828.html].
Kahn, Barbara E. and Brian Wansink (2004), “The Influence of
Assortment Structure on Perceived Variety and Consumption
Quantities,” Journal of Consumer Research, 30 (4), 519–33.
Kapferer, Jean-Noel (1995), “Brand Confusion: Empirical Study
of a Legal Concept,” Psychology and Marketing, 12 (6),
551–69.
Loken, Barbara, Ivan Ross, and Ronald L. Hinkle (1986), “Consumer
‘Confusion’ of Origin and Brand Similarity Perceptions,”
Journal of Public Policy & Marketing, 5, 195–211.
L’Oreal S.A. v. Bellure N.V. (2009), Court of Appeal, Rule of Professional
Conduct, Case C-487/07, 125, 196–242, (accessed
March 2011), [available at http://eur-lex.europa.eu/LexUriServ/
LexUriServ.do?uri=OJ:C:2009:180:0006:0006:EN:PDF].
Markman, Arthur B. and Jeffrey Loewenstein (2010), “Structural
Comparison and Consumer Choice,” Journal of Consumer Psychology,
20 (2), 126–37.
Meyers-Levy, Joan and Alice M. Tybout (1989), “Schema Congruity
as a Basis for Product Evaluation,” Journal of Consumer
Research, 16 (1), 39–54.
Miaoulis, George and Nancy d’Amato (1978), “Consumer Confusion:
Trademark Infringement,” Journal of Marketing, 42
(April), 48–55.
Mitchell, Vincent-Wayne and Ide Kearney (2002), “A Critique of
Legal Measures of Brand Confusion,” Journal of Product and
Brand Management, 11 (6), 357–79.
Mogilner, Cassie, Tamar Rudnick, and Sheena S. Iyengar (2008),
“The Mere Categorization Effect: How the Presence of Categories
Increases Choosers’ Perceptions of Assortment Variety
and Outcome Satisfaction,” Journal of Consumer Research, 35
(2), 202–215.
Morrin, Maureen and Jacob Jacoby (2000), “Trademark Dilution:
Empirical Measures for an Elusive Concept,” Journal of Public
Policy & Marketing, 19 (Fall), 265–76.
———, Jonathan Lee, and Greg M. Allenby (2006), “Determinants
of Trademark Dilution,” Journal of Consumer Research, 33 (2),
248–57.
Murphy, Sheila T. and Robert B. Zajonc (1993), “Affect, Cognition,
and Awareness: Affective Priming with Optimal and Suboptimal
Stimulus Exposures,” Journal of Personality and Social
Psychology, 64 (5), 723–39.
Muthukrishnan, A.V. and Seshan Ramaswami (1999), “Contextual
Effects on the Revision of Evaluative Judgments: An Extension
of the Omission-Detection Framework,” Journal of Consumer
Research, 26 (1), 70–84.
Nowlis, Stephen M. and Itamar Simonson (1997), “Attribute–Task
Compatibility as a Determinant of Consumer Preference Reversals,”
Journal of Marketing Research, 34 (May), 205–218.
Oakley, James L., Adam Duhachek, Subramanian Balachander,
and S. Sriram (2008), “Order of Entry and the Moderating Role
of Comparison Brands in Brand Extension Evaluation,” Journal
of Consumer Research, 34 (5), 706–712.
Olsen, Sven Ottar (2002), “Comparative Evaluation and the Relationship
Between Quality, Satisfaction, and Repurchase Loyalty,”
Journal of the Academy of Marketing Science, 30 (3),
240–49.
Planet Retail (2007), Private Label Trends Worldwide. London:
Planet Retail.
Poynor, Cait and Kristin Diehl (2010), “The Effects of Benefit and
Feature-Based Assortment Structures on Similarity Perceptions
and Strength of Preference,” working paper, Katz Graduate
School of Business, University of Pittsburgh.
Robinson, William T. and Claes Fornell (1985), “Sources of Market
Pioneer Advantages in Consumer Goods Industries,” Journal
of Marketing Research, 22 (August), 305–317.
Schwarz, Norbert and Herbert Bless (1992), “Constructing Reality
and Its Alternatives: An Inclusion/Exclusion Model of Assimilation
and Contrast Effects in Social Judgment,” in The Construction
of Social Judgment, Leonard L. Martin and Abraham
Tesser, eds. Hillsdale, NJ: Lawrence Erlbaum Associates,
217–45.
Scott-Morton, Fiona and Florian Zettelmeyer (2004), “The Strategic
Positioning of Store Brands in Retailer–Manufacturer Negotiations,”
Review of Industrial Organization, 24 (2), 161–94.
Simonson, Itamar (1994), “Trademark Infringement from the
Buyer Perspective: Conceptual Analysis and Measurement
Implications,” Journal of Public Policy & Marketing, 13 (Fall),
181–99.
——— (1999), “The Effect of Product Assortment on Buyer Preferences,”
Journal of Retailing, 75 (3), 347–70.
———, Stephen M. Nowlis, and Katherine Lemon (1993), “The
Effect of Local Consideration Sets on Global Choice Between
Lower Price and Higher Quality,” Marketing Science, 12 (4),
357–77.
Stapel, Diederik A. (2007), “In the Mind of the Beholder: The
Interpretation Comparison Model of Accessibility Effects,” in
Assimilation and Contrast in Social Psychology, Diederik A.
Stapel and Jerry Suls, eds. New York: Psychology Press,
143–64.
———, Willem Koomen, and Aart S. Velthuijsen (1998), “Assimilation
or Contrast? Comparison Relevance, Distinctness, and the
Impact of Accessible Information on Consumer Judgments,”
Journal of Consumer Psychology, 7 (1), 1–24.
Steenkamp, Jan-Benedict, Harald van Heerde, and Inge Geyskens
(2010), “What Makes Consumers Willing to Pay a Price Premium
for National Brands over Private Labels?” Journal of
Marketing Research, 47 (December), 1011–1024.
Ülkümen, Gülden, Amitav Chakravarti, and Vicki G. Morwitz
(2010), “Categories Create Mind-Sets: The Effect of Exposure
to Broad Versus Narrow Categorizations on Subsequent, Unrelated
Decisions,” Journal of Marketing Research, 47 (August),
659–71.
Warlop, Luk and Joseph W. Alba (2004), “Sincere Flattery: Trade-
Dress Imitation and Consumer Choice,” Journal of Consumer
Psychology, 14 (1), 21–27.
Wilke, Ricky and Judith Lynne Zaichkowsky (1999), “Brand Imitation
and Its Effects on Innovation, Competition, and Brand
Equity,” Business Horizons, 42 (6), 9–18.
Zaichkowsky, Judith L. (2006), The Psychology Behind Trademark
Infringement and Counterfeiting. Hillsdale, NJ: Lawrence
Erlbaum Associates.
Copyright of Journal of Marketing Research (JMR) is the property of American Marketing Association 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.