Original Article

From endorsement to celebrity

co-branding: Personality transfer
Received (in revised form): 21st February 2014

Laure Ambroise
is an assistant professor at University of Saint Etienne (France). Her studies focus on consumers and brands relationship. She
has published articles in international refereed marketing journals and conferences, especially on brand management
(Recherche et Applications en Marketing, Advances in Consumer Research). Her recent research projects cover brand personality
transfer and consumer empowerment.

Gaëlle Pantin-Sohier
is an assistant professor at University of Angers (France). Her research explores the relationship between human and brands.
Current projects cover brand personality and brand image creation through personality transfer and product package. She has
published in international and French journals as Innovative marketing and Recherche et Applications en Marketing.

Pierre Valette-Florence
is full professor of Marketing and Quantitative Methods at IAE de Grenoble, Grenoble University, France. He has many years
of experience as a brand management and marketing research consultant for leading international companies and has
published numerous articles in international refereed marketing journals and conferences. He has recently co-published a
special issue of the Journal of Brand Management on luxury and counterfeiting in May 2012.

Noel Albert
is an assistant professor at Kedge Business School (France). His studies especially focus on consumers’ love for brands: What
is love? Can we feel love for a brand? Is this feeling the same as love for a person? How can we measure a consumer’s love for a
brand? His works have appeared in various international and French journals (for example, Journal of Business Research, Journal
of Consumer Marketing, Innovations, Marché & Organisation).

ABSTRACT A brand’s personality is rooted in different sources, including its spokes-

persons, who can express individual values and visions of the self, in association with a
certain type of personality. This study seeks to measure the extent to which a celebrity
spokesperson’s personality transfers to the brand, as well as how this transfer influences
consumers, using a co-branding framework. An empirical study of two ready-to-wear
fashion brands specifies the conditions of a successful endorsement: The influence of
personality transfers on consumer behavior varies with the brand’s reputation and the
celebrity’s profile. Accordingly, the authors propose a new approach for selecting brand
ambassadors on the basis of their personality and in the context of a form of co-
branding.
Journal of Brand Management (2014) 21, 273–285. doi:10.1057/bm.2014.7;
published online 21 March 2014
Correspondence:
Laure Ambroise,
COACTIS, Université de Saint
Etienne, 6 rue Basse des Rives,
Keywords: brand personality; brand ambassador; co-branding; endorsement
42023 Saint-Etienne Cedex 2,
France.
E-mail: laure.ambroise@univ-st-
etienne.fr

© 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285

www.palgrave-journals.com/bm/
 Ambroise et al
INTRODUCTION
The increasingly important role of brands in
the daily lives of consumers has prompted
some creative terminology, referring to
the ‘brandization of society’ (Kornberger,
2010). Furthermore, people’s search for
emotions coincides with and often stimu-
lates their purchases, so brands use various
strategies to trigger emotional links with
their consumers. Among these approaches,
the use of celebrity endorsers provides a
means to humanize the brand and develop a
specific identity for it, through the accent-
uation of certain values and symbols. For
example, Nespresso lures international cli-
ents with a luxurious marketing strategy,
such that in the United States Penelope
Cruz cosies up at home, whereas in Eur-
opean advertising George Clooney creates a
‘clubby’ feeling.
Generally, a celebrity endorser’s char-
acteristics (for example, familiarity, exper-
tise) and the degree of congruence between
the brand and the celebrity represent the
central criteria for selecting a spokesperson
(Erdogan et al, 2001; Rice et al, 2011).
Managers also rely on pragmatic criteria (for
example, cost and loyalty) and other key
indicators, such as associations between the
celebrity and other brands or the celebrity’s
commitment to humanitarian causes. Because
modern celebrities constitute brands in their
own right, endorsements represent a form
of a co-branding strategy (Thomson, 2006;
Seno and Lukas, 2007; Halonen-Knight and
Hurmerinta, 2010). Yet to the best of our
knowledge, no existing method or tool
offers a means to assess a celebrity’s proximity
to a brand, in terms of their relationships
or identity, even as marketers increasingly
demand insights into the personal traits of
brand ambassadors to determine their quali-
fications for serving as brand representatives
(Carroll, 2009). This study proposes the
concept of personality for clarifying the
added value of a celebrity, in the context
of consumer–brand relationships.
274 © 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285
Accordingly, we present an approach that
can help managers assess the personality of
brand ambassadors, before they make cri-
tical endorsement decisions for their brand
(Ilicic and Webster, 2013). To specify the
value of celebrity co-branding, we analyze
the influence of the celebrity’s personality
on the brand’s personality; then we consider
the influence of the brand’s personality on
consumer behavior. In so doing, this study
offers a comparative assessment of the influ-
ence of the celebrity’s personality, accord-
ing to the reputation of the brand.
In the remainder of this article, we begin
with a focus on trends in brand manage-
ment that have promoted relationships
between the brand and consumers, in
which setting celebrity endorsements have a
particular role. After we define our research
hypotheses, we present our proposed
method for assessing the personality profiles
of several brands and celebrity spokes-
persons. The findings reveal a personality
transfer between celebrities and brands and
emphasize the differentiated influences of
brand personalities on attachment and pur-
chase intentions, depending on the brand’s
ambassador. The added value of personality
transfers within a celebrity co-branding
strategy has notable managerial implications;
we also recognize some study limitations
and avenues for research.
CONCEPTUAL FRAMEWORK
Considering the vast development of pri-
vate labels and increased sophistication of
consumers, marketers and their brands must
react carefully to trigger feelings among
consumers. From this perspective, the use of
celebrity endorsements offers a prime
method for advertisers to humanize their
brands. Managers thus devote significant
resources to solicit and reap the benefits of a
celebrity’s image; in 2006, for example, US
retailers spent an estimated US$3 billion on
celebrity advertising (White et al, 2009).
 From endorsement to celebrity co-branding

Table 1: Models on celebrity endorsement strategy

Model Description

Source credibility model
Source attractiveness model
Product match-up hypothesis
Meaning transfer models
Associative learning model
Co-branding
In this model, two variables impact the message effectiveness: the perceived level of
expertise (the extent to which a communicator is perceived to be a source of valid
assertions) and trustworthiness in an endorser (honesty, integrity and believability of an
endorser)
The effectiveness of the message depends on the attractiveness of the celebrity endorser.
The endorser’s attractiveness goes beyond the physical and is made up of three
dimensions: similarity (the supposed resemblance between the source and the receiver
of the message), familiarity (knowledge of the source through exposure) and likeability
(affection for the source as a result of the source’s physical appearance and behavior)
According to this model, messages conveyed by celebrity image and the product message
should be congruent for effective advertising. The determinant of the match between
celebrity and brand depends on the degree of perceived ‘fit’ between brand and celebrity
image
Celebrities represent symbolic significations that are directly transferred to the product/
brand. Then meanings pass from the product to the consumer through the brand’s
consumption
In this model, a cognitive connection/associative link is built between the endorser and the
brand, during the endorsement process. Associative learning is also called classical
learning
In this model, the association between the brand and the endorser is perceived as the
pairing of two (or more) brands. Such an association aims at an image transfer that
becomes mutually beneficial for the partners involved

Yet well-known stars also manage and pro-
tect their images as if they were brands. In
this sense, the notion of simple celebrity
endorsements may be outdated, in that the
situation evolved to something closer to a
brand alliance. Returning to the previous
example, George Clooney takes great care in
managing his image for authenticity, such
that he provides endorsements for Nespresso
in Europe but refuses to enter into a similar
brand alliance in the US market.
Brand ambassadors: Symbols of the
consumer–brand relationship
In their efforts to move consumers from
rational preferences to stronger, more emo-
tional links to the brand, brand managers
rely on psychosocial dimensions. That is,
they seek to move beyond technical or
performance-related considerations to pro-
vide meaning, particularly that derived from
their relationships with consumers. Simi-
larly, an advertising strategy that relies
mainly on a celebrity endorsement seeks to
create emotional connections with con-
sumers (McCracken, 1989; Erdogan, 1999).
Advertisers thus need to ensure that the
celebrity’s image and personality fit the nat-
ure of the product and its target audience.
Selecting a celebrity is a crucial but difficult
exercise (Carroll, 2009; Ilicic and Webster,
2013) that can constitute a substantial risk of
negative impacts on the brand and/or the
firm (Louie and Obermiller, 2002).
Several models offer possible guidelines
for selecting brand ambassadors and mana-
ging their influence on consumers (see
Table 1). These various approaches investi-
gate the credibility, believability and exper-
tise of the spokesperson (Ohanian, 1990);
his or her attractiveness (Kalhe and Homer,
1985); spokesperson–brand congruency
(Kamins, 1990); the likely transfer of sym-
bolic meaning from the celebrity to the
brand (Parulekar and Raheja, 2006; Miller
and Allen, 2012); the potential for co-
branding (Seno and Lukas, 2007); or

© 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285 275
 Ambroise et al
associative learning processes (Till, 1998). In
this view, consuming the brand allows
individual consumers to appropriate its var-
ious meanings. As an extension of these
models, we consider the relevant person-
ality transfer between the celebrity and the
brand from a co-branding perspective.
Personality transfers
Aaker (1997) recommends approaching the
consumer–brand relationship through the
concept of personality, because consumers
humanize brands, which facilitates their
sense of emotions felt toward the brand
and their creation of a relationship with it.
The concept of personality therefore can
help explicate the brand identification
process. However, in interpersonal rela-
tionships, people develop their own per-
sonality even as they incorporate desirable
traits from partners. When a partner pos-
sesses more highly valued traits, he or she
is more attractive and interesting, as an
extension of the focal individual’s self.
Similarly, consumers tend to include well-
known brands as integral parts of their self
(Aaker, 1999). Brand ambassadors or
spokespeople can manifest the human
component of the brand’s value system (de
Chernatony, 2001), as well as offering
their own desirable traits, due to their
celebrity status (Kurzman et al, 2007). In
turn, if we regard the celebrity endorse-
ment as more than a simple advertising
initiative, such that it provides a basis for
an authentic, long-term change in con-
sumers’ relational connection and identi-
fication with the brand then it constitutes
a co-branding process (Leuthesser et al,
2003; Seno and Lukas, 2007).
Co-branding brand personalities
Brands represent a vast field of action
(Kornberger, 2010). Ranging from
a simple distinctive sign to a profitable
276 © 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285
immaterial asset, they exist in an infinite
variety of forms. Their importance is such
that political parties, associations and even
countries have become brands; high-
profile personalities also achieve brand sta-
tus, from political figures (Barak Obama),
to athletes (Usain Bolt), to entertainers
(Lady Gaga). Celebrities rely on branding
strategies to promote themselves, including
merchandizing, licensing their name,
launching their own product lines and
endorsing products to increase their visibi-
lity (Thomson, 2006).
Traditional investigations of the celebrity
endorsement process rely on communica-
tion theory (which emphasizes endorsers’
credibility and attractiveness) or theories of
cultural meaning transfer. These theories
highlight a single pathway for this process:
The celebrity introduces meanings and
associations to the brand and receives
financial compensation for doing so. How-
ever, because celebrities have become their
own brands, existing models for explaining
celebrity endorsement strategies appear
incomplete. Accordingly, some studies pro-
pose that celebrity endorsement is an inter-
active, reciprocal process, such that the
brand also has an impact on the celebrity’s
equity (Till, 1998; Seno and Lukas, 2007;
White et al, 2009; Halonen-Knight and
Hurmerinta, 2010). We advance this line
of reasoning by positing that the celebrity
endorsement process reflects a brand alli-
ance strategy, through co-branding. Such a
strategy is orchestrated by brand managers
and celebrities in an effort to increase their
respective brand equity (Seno and Lukas,
2007). As in a brand alliance, the celebrity
endorsement or co-branding requires a
meaning transfer and associative learning
(Halonen-Knight and Hurmerinta, 2010).
The link between the brand and the celeb-
rity gets created in consumers’ minds
through actions that repeatedly pair these
partners (Till, 1998). Associations then
transfer from one entity to the other,

resulting in associative networks, such that
when consumers think about the brand
they transfer their brand evaluation to the
celebrity, and vice versa (Halonen-Knight
and Hurmerinta, 2010).
Studying celebrity endorsement in more
depth as a form of co-branding offers sev-
eral advantages. In particular, this view
broadens the meaning of celebrity (Seno
and Lukas, 2007; Halonen-Knight and
Hurmerinta, 2010), because the celebrity
moves to a level equivalent to that of the
endorsed brand (Seno and Lukas, 2007),
such that his or her equity can be affected
by the endorsed brand (Halonen-Knight
and Hurmerinta, 2010). This perspective
offers a potentially more comprehen-
sive depiction of celebrity endorsements.
Moreover, by regarding celebrity endor-
sements as a form of co-branding, mar-
keters can go beyond classical notions
of credibility or attractiveness (Halonen-
Knight and Hurmerinta, 2010). In addition
to developing indicators of the commercial
or emotional value contributed by brand
ambassadors, understanding their relational
and identification value is critical. Because
the celebrity’s personality offers an inter-
esting vector of individualization and
identification, this study seeks to assess
personality transfers that occur in a celeb-
rity co-branding strategy.
RESEARCH HYPOTHESES
A widely accepted meaning of brands
regards them as a form of self-expression,
which can be heightened through the use of
celebrity endorsements (Park and John,
2010). Endorsements support the transfer
of meanings, associations and values from
the celebrity to the brands he or she repre-
sents (McCracken, 1989). de Chernatony
(2001) also emphasizes that brand ambassa-
dors represent the human components of a
brand’s value system. Across this set of
symbolic associations, a personality transfer
© 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285
From endorsement to celebrity co-branding
is likely between the celebrity and the
brand, and we hypothesize:
Hypothesis 1: The brand ambassador’s
personality transfers significantly to
the brand personality.
This transfer of brand image varies with the
context though (Dickinson and Heath,
2006). Prior co-branding research reveals
that strong brands influence a new (co-)
brand more (Simonin and Ruth, 1998;
Leuthesser et al, 2003), because they enjoy
powerful advantages, in terms of encoding
and storing associations in consumers’ mem-
ory, over lesser-known brands (Hoeffler and
Keller, 2003). In contrast, the images of
lesser-known brands can be easily influenced
by emotional reactions triggered by advertis-
ing (Ilicic and Webster, 2013), and they are
more open to emotional transfers (Miller and
Allen, 2012). Transferring this reasoning to a
brand–celebrity co-branding context, we
predict:
Hypothesis 2: A brand ambassador’s per-
sonality exerts a greater influence on
the personality of relatively lesser-
known brands than on those of well-
known brands.
In general, the link between personality and
behavior constitutes a founding principle of
personality theories (Pervin and John, 2001).
The structure of an individual’s personality
dynamically organizes his or her perceptions,
beliefs, attitudes and behaviors (Ozer and
Reise, 1994). Furthermore, the act of con-
sumption always transmits symbolic mean-
ings, whether consciously or unconsciously
(Elliott and Wattanasuwan, 1998). Brands
therefore can be understood as instrumental
consumer behavior variables (Fournier,
1998), and brand personality can explain, at
least partly, consumer behavior, which itself
serves to express a certain personality (Batra
and Homer, 2004; Mulyanegara et al, 2009).
If a personality transfer differs with various
brand ambassadors, the influence of the
277
 Ambroise et al
Celebrity
personality
transfer
Figure 1: Research model.
brand’s personality on consumer behavior
should differ as well, as our conceptual fra-
mework in Figure 1 suggests.
Hypothesis 3a: Brand personality has a
significant effect on attachment to the
brand.
Hypothesis 3b: The effect of brand per-
sonality on attachment differs signifi-
cantly according to the celebrity who
represents it.
Hypothesis 3c: Brand personality has a
significant effect on purchase inten-
tions toward the brand.
Hypothesis 3d: The effect of brand per-
sonality on purchase intentions differs
significantly according to the celebrity
who represents it.
EMPIRICAL STUDY
To test for the existence of personality
transfers between celebrities and brands, this
study involved three phases.
Selecting measurement scales
Personality barometer
Considering the study aim, we sought
to use the same scale to measure the
278 © 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285
Endorser
Brand
attachment
Brand
personality
Purchase
intention
personalities of both the brand and the
celebrity. The brand personality barometer
(Ambroise and Valette-Florence, 2010)
corresponds to a specific scale for brands and
is well adapted to a French context. This
inventory of 24 trait adjectives is organized
into 9 facets and 5 dimensions. The five-
dimensional structure reflects most research
on personality; four of these dimensions
(introversion, agreeableness, conscientious-
ness, sophistication) are similar to the brand
personality scale developed by Aaker
(1997), and a fifth dimension (disin-
genuousness) refers to a less desirable per-
sonality trait, which can reveal possible
brand distrust among consumers. The relia-
bility, validity and stability of this scale were
tested previously with 39 brands in 12 dif-
ferent product categories (Ambroise and
Valette-Florence, 2010).
Brand personality scales usually are not
used to measure human personality, and it is
reasonable to question their legitimacy for
celebrities. However, the brand personality
scale development process entailed selecting
from inventories of human personality traits
the main elements that consumers applied
to brands. Thus, the brand personality bar-
ometer includes items that measure per-
sonality traits, using adjectives that are
exclusively applicable to brands. The use of
such a barometer to measure celebrity

Originality
Introversion
-0.5
Conscientiousness
Deceitfulness
Dominance
Figure 2: Celebrities personality profiles.
personalities is therefore justified and can
effectively, if partially, depict the perceived
personality of the celebrity that consumers
transfer to brands and/or identify.
Behavior toward the brand
Prior literature devoted to the consumer–
brand relationship cites three constructs that
determine attitudinal loyalty and behaviors
toward the brand: trust, attachment and
commitment (Chaudhuri and Holbrook,
2001; Thomson et al, 2005). Because our
research goal is not to contribute to the
debate about the relative influence of these
concepts but rather to offer insights into the
consumer–brand relationship that reflect
the concept of personality, we focus mainly
on attachment and purchase intentions.
With an adaptation of the scale proposed by
Thomson et al (2005), we measured con-
sumer attachment to the brands studied
from an emotional perspective. To estimate
the behavioral dimension, we also included
an item to measure purchase intentions
toward the brand.
Selecting brands and celebrities for
the study
Ready-to-wear fashion provided the
context for our data collection; in this
product category, consumer identification
tends to be strong (Caroll, 2009) and
© 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285
From endorsement to celebrity co-branding
1
0.5 Precociousness
0
Creativity
-1
Seduction
Vanessa
Congeniality
Madonna
celebrity endorsements are relatively
common, so it should offer clear insights.
To compare influence across different
levels of brand recognition, we selected
two brands in this category, one with
strong brand equity and another that was
not well known to the study sample. A
pre-test indicated two brands of jeans:
Levi’s with its strong reputation and
Abercrombie & Fitch, which was gen-
erally unknown in France at the time of
the data collection. (We confirmed these
levels of brand recognition among study
respondents in advance.)
We also needed to select very different
brand ambassadors, because the aim of the
study was to distinguish different profiles.
An initial sample of 60 students indicated
which female celebrities they thought
would best represent a brand of jeans.1 Four
names emerged: Laetitia Casta, Jennifer
Aniston, Madonna and Vanessa Paradis (a
French singer and actress and former
romantic partner of Johnny Depp). Next,
we questioned an ad hoc sample of
76 students about their perceptions of
the personalities of these four celebrities.
These profiles revealed that Madonna and
Vanessa Paradis achieved the greatest dis-
tance (Figure 2). Moreover, these two stars
had not been featured recently in advertis-
ing campaigns in France at that time, so
their images would not be influenced
by associations with multiple brands.
279
 Ambroise et al
Table 2: Differentiating power of the brand personality barometer
F Significance
Conscientiousness 23.35
Introversion 1.31
Sophistication 27.87
Agreeableness 6.36
Disingenuousness 3.96
Accordingly, we used Vanessa Paradis
and Madonna as spokespersons, to assess
the influence of celebrity personalities on
brand personalities.
Collecting data and confirming
measurement tools
We administered six questionnaires, such
that for each brand we provided three
variations of the same questionnaire, fea-
turing an advertisement similar to those
published in fashion magazines. The over-
all advertisement remained the same across
versions, presenting the endorser in a sen-
sual pose, wearing a pair of jeans with her
arms crossed in front of her body. Only the
face was modified, using a photomontage
technique, such that it showed Madonna,
Vanessa Paradis or an anonymous model.
Each questionnaire also asked respondents
to assess the personality of the star who did
not appear in the ad they saw. Of the 309
French students who answered the self-
administered questionnaires, approxi-
mately 50 people considered each of the
six versions.
To confirm the measurement tools, we
ran exploratory and confirmatory factor
analyses, which affirmed the structure
of the scales. Regarding the brand person-
ality barometer, the satisfactory indicators
of reliability and validity indicated good trait
validity for the personality dimensions (root
mean square error of approximation
= 0.085; t-tests statistically significant;
Jöreskog’s ρ > 0.64).
280 © 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285
F Significance
0.00 0.03 0.85
0.25 320.81 0.00
0.00 0.38 0.54
0.01 9.09 0.00
0.05 144.05 0.00
RESULTS
Personality: Legitimate differentiation
for brands and celebrity endorsements
To confirm the personality barometer’s
ability to differentiate the personalities of
brands and celebrities, we conducted var-
iance analyses, using the scores for each
dimension. They reveal that the brand per-
sonality barometer can discriminate differ-
ent personality profiles of brands and
celebrities (Table 2).
These preliminary results also confirm the
appeal of using personality to distinguish
between brands (for example, Abercrombie
& Fitch versus Levi’s). The celebrities can be
significantly differentiated. Three personality
dimensions indicate significant statistical dis-
tinctions: introversion (Madonna displays an
extraverted personality, Paradis appears shy),
agreeableness (Paradis is perceived as friend-
lier than Madonna) and disingenuousness
(Paradis appears more deceptive and arrogant
than Madonna). With these results, we high-
light the relevance for marketers of using the
brand personality barometer to characterize
potential celebrity endorsers.
Influences of celebrity personality on
brand personality
To assess the influence of the endorsers’
personalities on the brands we studied, we
conducted variance analyses with the scores
for the different personality dimensions.
The results appear in Table 3, with sig-
nificant findings in bold.
 From endorsement to celebrity co-branding

Significance
These results confirm Hypothesis 1: The

0.08

0.59

0.16
0.86

0.68

0.82
brand’s profile differs statistically across cer-

Conscietiousness
tain dimensions, depending on which brand
ambassador represents it. For example, both

difference

0.35

0.09

−0.26
−0.03

−0.07

−0.04
Mean
Levi’s and Abercrombie & Fitch are per-
ceived as more extraverted when they are
represented by Madonna (positive and sig-

Significance
nificant mean difference). Thus, we identify

0.00

0.00

0.29
0.97

0.98

0.99
a transfer of the star’s perceived personality

Sophistication
to the brand on this dimension. These initial
results reaffirm the need to consider simul-

difference

−0.79

−0.58

0.21
−0.01

0.00

0.00
Mean
taneously the personalities of both compo-
nents: the brand and the celebrity endorser.
Furthermore, our results show that the

Significance
celebrity personality transfer is more sub-

0.04

0.20

0.29
0.79

0.69

0.90
Disingenuousness
stantial for lesser-known than for well-
known brands, in support of Hypothesis 2.
Four dimensions of Abercrombie & Fitch’s

difference

−0.45

−0.23

0.22
−0.05

−0.07

−0.03
Mean
brand personality are significantly influ-
enced by the celebrity, whereas Levi’s
reveals substantially less influence of the

Significance
celebrities.

0.39

0.46

0.79
0.98

0.97

0.99
Influences of brand personality on Agreeableness

difference

−0.20

−0.14

0.06
0.00

−0.01

0.00
consumer behavior, depending on Mean
celebrity endorser
For measuring the influence of brand per-
Significance

sonality on consumer behavior, we estab-

Note: Bold values correspond to those whose significance is lower than 0.1
0.02

0.81

0.07
0.10

0.92

0.17
lished structural equation models, with
Introversion

brand attachment or purchase intentions as

Table 3: Ambassador personality influence on brand personality

the dependent variables and brand person-

difference

0.45

−0.05

−0.39
0.30

0.02

−0.28
Mean

ality as the explanatory variable. We also

conducted multi-group analyses, to com-
pare the influence of brand personality on
attachment and purchase intentions for the
Madonna versus Vanessa

Madonna versus Vanessa

Unknown ambassadress

Unknown ambassadress

Unknown ambassadress

Unknown ambassadress

different celebrity endorsers.

versus Madonna

versus Madonna
versus Vanessa

versus Vanessa

As the results in Table 4 demonstrate,

brand personality affects attachment, in sup-
port of Hypothesis 3a, and purchase inten-
tions, in support of Hypothesis 3c. Moreover,
these influences vary significantly depending
Multiple comparisons

on who provides the endorsement, for both

Abercrombie &

attachment (Hypothesis 3b) and purchase

intentions (Hypothesis 3d). However, these
(LST test)

Fitch

results also offer an insight we did not predict

Levi’s

explicitly; that is, the impact of the celebrity’s

© 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285 281
 Ambroise et al

Table 4: Influence of brand personality on consumer behavior according to brand ambassador

Model with a third-level variable of personality

Results on the complete data set

Purchase intention Brand attachment

R2 R2(Bootstrap) R2 R2(Bootstrap)

0.249 0.257 0.303 0.310

Path coefficient Path coefficient
Personality 0.499 Personality 0.550

Results differenciated by endorser/brand pairs

Levis’

Purchase Intention Brand attachment

R2(Bootstrap) Path coefficient R2(Bootstrap) Path coefficient

Unknown ambassadress 0.156 0.360 0.265 0.492

Madonna 0.374 −0.607 0.318 −0.546
Vanessa 0.249 0.492 0.336 0.567

Abercrombie & Fitch

Purchase Intention Brand attachment

2 2
R (Bootstrap) Path coefficient R (Bootstrap) Path coefficient

Unknown ambassadress 0.171 0.386 0.481 0.684

Madonna 0.259 0.488 0.431 0.625
Vanessa 0.365 0.600 0.338 0.569

personality is not always positive. The brand
personality path coefficients for Madonna
are negative in relation to attachment to and
purchase intentions toward Levi’s, but for
this same brand the path coefficients are
positive for Vanessa Paradis or an unknown
model. These results reinforce the impor-
tance of careful endorsement choices: If a
co-branding is not relevant, the effect may
damage the brand.
The results also reveal that the impact of
brand personality on brand attachment and
purchase intentions varies with the brand’s
reputation. For Levi’s, the impact of brand
personality on attachment is much greater,
whether the brand is represented by
Madonna or Paradis. For Abercrombie &
Fitch, their famous personalities do not have
strong influences on brand attachment,
but the impact on purchase intentions is
much stronger. We posit that sophisticated
consumers, who are generally aware of
marketing techniques, might consider an
endorsement insufficient as a motivation to
commit to a sincere relationship with the
brand. However, the brand ambassador still
can promise quality, which improves con-
sumers’ purchase intentions. These pre-
liminary (and exploratory) results indicate
that the choice of a well-known spokes-
person strongly influences purchase inten-
tions, but a similar brand attachment effect
could be achieved with a relatively
unknown representative, particularly if the
brand is less well known. Personality trans-
fers almost certainly take place, in line with
the theory of morphopsychology (Pantin-
Sohier, 2009). However, for a well-known
brand, a personality transfer from a famous
ambassador appears more beneficial, in
terms of both attachment and purchase
intentions. Because the impact differs with

282 © 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285

the varying personalities of the ambassador,
our results emphasize the need to measure
potential endorsers’ personalities, in con-
junction with the brand personality, to ver-
ify congruence and ensure the relational
proximity for the brand and consumers.
These results also might be interpreted in
relation to the learning associative approach.
For an unknown brand, no network of
cognitive associations has been built. The
impact of the personality of a specific
celebrity thus is not particularly convincing,
because the relevant links and transfers can-
not operate effectively yet. This finding
suggests a necessary path for further
research, which can clarify the specific
effects that result from varying degrees of
fame of the brand.
DISCUSSION AND MANAGERIAL
IMPLICATIONS
This study highlights the theoretical and
managerial interest of using personality to
measure and ensure consistency across a set
of communication vectors to reinforce the
desired personality they seek to project. In
this sense, brand personality has a strategic
role in brand management (Keller and
Lehmann, 2006). This study also empha-
sizes the effectiveness of tactics that use
celebrity personality transfers to promote
new brands or reinforce the images of
familiar ones. The transfer of meaning can
help modify or determine the personality of
a brand represented by a certain celebrity. It
also tends to increase attachment in some
situations and exerts a positive effect on
purchase intentions in most cases.
Marketers thus have a means to measure
brand ambassador personalities to make
better choices of apt representatives for their
communication campaigns. In particular,
brand personality scales can serve to com-
pare the personalities of possible spokes-
persons and the brand. For brands that are
already well known, the choice of a
© 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285
From endorsement to celebrity co-branding
celebrity endorsement might have specific
purposes, such as revamping the brand by
calling on celebrities with a strong media
presence or those who are popular with
younger audiences (for example, Natalia
Vodianova for the legendary fragrance Sha-
limar by Guerlain). Conversely, the use of
emblematic celebrities can be used to
associate the brand with a period for which
consumers feel deeply nostalgic. French
luxury brands often use such strategies: Dior
employs footage of Alain Delon from the
1969 film The Swimming Pool to advertise
Eau Sauvage, and it mixes images of
Charlize Theron with nostalgic shots of
Grace Kelly and Marilyn Monroe in a US
advertisement for its namesake fragrance.
The brand ambassador also can reinforce the
identity that the brand seeks to establish in
the public’s mind. For example, Schweppes
chose Uma Thurman to tout its drinks in
a humorous campaign that featured the
slogan ‘What did you expect?’ Her image
helped the brand emphasize the ambiguous
nature of the sensual situations depicted in
the ads, adding a greater sense of mischief,
creativity and seduction, because of the pre-
sence of the heroine of Pulp Fiction and the
Kill Bill films. Finally, the choice of associa-
tions with certain celebrities can allow the
brand to communicate its commitment to a
cause or create a more tolerant, humanist or
ethical brand image. L’Oréal and Red Bull
are represented, respectively, by Aimée
Mullins, a bilateral amputee whose athletic
accomplishments led to her successful mod-
eling career, and Ashley Fioleck, a two-time
Women Motocross Association champion
who was born deaf. It would be interesting
to measure a posteriori the impact of such
choices on separate dimensions of the brand’s
personality.
In summary, the results of this study
point to a substantial influence of the per-
sonalities of well-known brand ambassadors
on purchase intentions, regardless of the
status of the brand (well known or
283
 Ambroise et al
unfamiliar). It also would be interesting to
compare more brands with varying levels of
recognition or awareness, beyond the two
extremes we tested in this study. Further-
more, we hope further research tests celeb-
rity personality transfers in more detail by
explicitly varying the type of celebrity (for
example, singer, actor, model), which
might clarify relevant endorsements in light
of Ilicic and Webster’s (2013) results. Other
studies could analyze the repercussions of
multiple brand endorsements too, as Rice
et al (2011) have done.
Moving beyond our focal investigation,
we also note that consumers tend to be
sensitive to how brands react when their
ambassadors behave badly. Any incident
with negative effects is likely to be judged in
various ways by consumers, who distinguish
the nature of the act in question, the celeb-
rity’s degree of responsibility and his or her
reactions to media coverage of the incident
(Louie and Obermiller, 2002), as well as the
extent of actions that associated brands take
in response. When Gatorade, AT&T and
Accenture ended their contracts with Tiger
Woods after an infidelity scandal in 2009,
Knittel and Stango (2010) estimated that it
produced total shareholder losses of $5–12
billion. Yet his actions appeared clearly less
serious than the allegations surrounding
Olympian Oscar Pistorius, whose murder
charges prompted quick disavowals from
co-brands such as Nike and Oakley. It
would be interesting to contextualize the
protocol used in this study by analyzing the
personalities of various celebrities at differ-
ent stages in their careers.
Similarly, though this study justifies the
use of celebrities to reinforce certain
dimensions of brand personality, the specific
product we study (jeans) typically entails
individual purchases, such that attitudes
toward these items might result from ela-
borate processes, due to the involvement
triggered. Further research should replicate
this experiment with other product catego-
284 © 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285
ries to confirm the impacts of celebrities for
products of varying types and in different
situational purchase contexts – just as Batra
and Homer (2004) tested McCracken’s
(1989) transfer of meaning model and
examined situational influences on the
construction and reinforcement of brand
imagery beliefs.
Finally, opportunities for further research
result from the need for a closer considera-
tion of the actual level of fit between the
brand and the celebrity personality, the sta-
bility of the results over time and the
potential for cultural effects. Testing the
proposed model into more countries, espe-
cially through cross-cultural comparisons,
would be insightful. From a methodological
perspective, this research also suffers limita-
tions related to the samples of brands and
celebrities, as well as the student respon-
dents. It would be interesting to determine
if other age groups and consumer profiles
process the personality transfers differently.
NOTE
1 To avoid introducing a potential source of gender bias, we
used female celebrities, because women are more often
associated with brands that offer psycho-sociological
benefits. For the purposes of this study, it was more
appropriate to study female brand ambassadors.
REFERENCES
Aaker, J.L. (1997) Dimensions of brand personality.
Journal of Marketing Research 34(3): 347–356.
Aaker, J.L. (1999) The malleable self: The role of self-
expression in persuasion. Journal of Marketing Research
36(1): 45–57.
Ambroise, L. and Valette-Florence, P. (2010) The brand
personality metaphor and inter-product stability of a
specific barometer. Recherche et Applications Marketing.
(English edition) 25(2): 3–29.
Batra, R. and Homer, P.M. (2004) The situational
impact of brand imagery beliefs. Journal of Consumer
Psychology 14(3): 318–330.
Carroll, A. (2009) Brand communications in fashion
categories using celebrity endorsement. Journal of
Brand Management 17(October/November): 146–158.
Chaudhuri, A. and Holbrook, M.B. (2001) The chain of
effects from brand trust and brand affect to brand
performance: The role of brand loyalty. Journal of
Marketing 65(April): 81–93.

de Chernatony, L. (2001) A model for strategically
building brands. Brand Management 9(1): 32–44.
Dickinson, S. and Heath, T. (2006) A comparison of
qualitative and quantitative results concerning
evaluations of co-branded offerings. Journal of Brand
Management 13(6): 393–406.
Elliott, R. and Wattanasuwan, K. (1998) Brands as
symbolic resources for construction of identity.
International Journal of Advertising 17(2): 131–144.
Erdogan, B. (1999) Celebrity endorsement: A literature
review. Journal of Marketing Management 15(4): 291–314.
Erdogan, B.Z., Baker, M.J. and Tagg, S. (2001) Selecting
celebrity endorsers: The practitioner’s perspective.
Journal of Advertising Research 3(41): 39–48.
Fournier, S. (1998) Consumers and their brands:
Developing relationship theory in consumer
research. Journal of Consumer Research 4(24): 343–373.
Halonen-Knight, E. and Hurmerinta, L. (2010) Who
endorses whom? Meaning transfer in celebrity
endorsement. Journal of Product and Brand
Management 19(6): 452–460.
Hoeffler, S. and Keller, K.L. (2003) The marketing
advantages of strong brands. Brand Management
10(6): 421–445.
Ilicic, J. and Webster, M. (2013) Celebrity co-branding
partners as irrelevant brand information in adver-
tisements. Journal of Business Research 66(7): 941–947.
Kalhe, L.R. and Homer, P.M. (1985) Physical
attractiveness of the celebrity endorser: A social
adaptation perspective. Journal of Consumer Research
11(4): 954–961.
Kamins, M.A. (1990) An investigation into the ‘match
up’ hypothesis in celebrity advertising: When beauty
may be only skin deep. Journal of Advertising 19(1):
4–13.
Keller, K.L. and Lehmann, D.R. (2006) Brands and
branding: Research findings and future priorities.
Marketing Science 25(6): 740–759.
Knittel, C.R. and Stango, V. (2010) Shareholder Value
Destruction Following the Tiger Woods Scandal.
Davis: Graduate School of Management, University
of California.Working Paper, http://faculty.gsm
.ucdavis.edu/~vstango/tiger004.pdf.
Kornberger, M. (2010) Brand Society: How Brands
Transform Management and Lifestyle. Cambridge, US:
Cambridge University Press.
Kurzman, C. et al. (2007) Celebrity status. Sociological
Theory 25(4): 347–367.
Leuthesser, L., Kohli, C. and Suri, R. (2003) 2+2 = 5?
A framework for using co-branding to leverage a
brand. Journal of Brand Management 11(1): 35–47.
Louie, T.A. and Obermiller, C. (2002) Consumer
response to a firm’s endorser: (Dis)association
decisions. Journal of Advertising 31(4): 41–52.
McCracken, G. (1989) Who is the celebrity endorser?
Cultural foundations of celebrity endorsement pro-
cess. Journal of Consumer Research 16(3): 310–321.
© 2014 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 21, 4, 273–285
From endorsement to celebrity co-branding
Miller, F.M. and Allen, C.T. (2012) How does celebrity
meaning transfer? Investigating the process of meaning
transfer with celebrity affiliates and mature brands.
Journal of Consumer Psychology 22(3): 443–452.
Mulyanegara, R.C., Tsarenko, Y. and Anderson, A.
(2009) The big five and brand personality:
Investigating the impact of consumer personality on
preferences towards particular brand personality.
Journal of Brand Management 16(4): 234–247.
Ohanian, R. (1990) Construction and validation of scale
to measure celebrity endorsers’ perceived expertise,
trustworthiness and attractiveness. Journal of Advertisi-
ng 19(3): 39–52.
Ozer, D.J. and Reise, S.P. (1994) Personality assessment.
Annual Review of Psychology 45(1): 357–388.
Pantin-Sohier, G. (2009) The influence of the product
package on functional and symbolic associations of
brand image. Recherche et Applications en Marketing.
(English edition) 24(2): 53–72.
Park, J.K. and John, D.R. (2010) Got to get you into my
life: Do brand personalities rub off on consumers?
Journal of Consumer Research 37(4): 655–669.
Parulekar, A.A. and Raheja, P. (2006) Managing celebrities
as brands: Impact of endorsements on celebrity image.
In: L.R. Kahle and C. Kim (eds.) Creating Images and the
Psychology of Marketing Communication. Mahwah, NJ:
Lawrence Erlbaum, pp. 161–169.
Pervin, L.A. and John, O.P. (2001) Personality Theory and
Research, 8th edn. New York: Wiley.
Rice, D.H., Kelting, K. and Lutz, R.J. (2011) Multiple
endorsers and multiple endorsements: The influence
of message repetition, source congruence and
involvement on brand attitudes. Journal of Consumer
Psychology 22(2): 249–259.
Seno, D. and Lukas, B.A. (2007) The equity effect of
product endorsement by celebrities: A conceptual
framework from a co-branding perspective. European
Journal of Marketing 41(1/2): 121–134.
Simonin, B.L. and Ruth, J.A. (1998) Is a company
known by the company it keeps? Assessing the
spillover effects of brand alliances on consumer
brand attitudes. Journal of Marketing Research
35(February): 30–42.
Thomson, M. (2006) Human brands: Investigating
antecedents to consumers’ strong attachments to
celebrities. Journal of Marketing 70(3): 104–119.
Thomson, M., MacInnis, D.J. and Whan, P.C. (2005)
The ties that bind: Measuring the strength of
consumers’ emotional attachment to brands. Journal
of Consumer Psychology 15(1): 77–91.
Till, B.D. (1998) Using celebrity endorsers effectively:
Lessons from associative learning. Journal of Product &
Brand Management 7(5): 400–409.
White, D.W., Lucretia, G. and Wilbur, N. (2009) The
effects of negative information transference in the
celebrity endorsement relationship. International Journal
of Retail & Distribution Management 37(4): 322–335.
285