Call For Autenticity
Call For Autenticity
fi
Pöyry , E I , Pelkonen , M , Naumanen , E & Laaksonen , S-M 2019 , ' A Call for Authenticity
: Audience Responses to Social Media Influencer Endorsements in Strategic Communication
' , International Journal of Strategic Communication , vol. 13 , no. 4 , pp. 336-351 . https://doi.org/10.1080/1553118X.
http://hdl.handle.net/10138/327468
https://doi.org/10.1080/1553118X.2019.1609965
acceptedVersion
Pöyry, E. I., Pelkonen, M., Naumanen, E. & Laaksonen, S-M., (2019). A Call for
Authenticity: Audience Responses to Social Media Influencer Endorsements in Strategic
Communication. International Journal of Strategic Communication, 13(4), 336–351.
Abstract
As social media have matured, the relative influence of one’s peers is decreasing,
while celebrities are gaining ground, particularly in certain social media channels such as
Instagram and Snapchat (Chaffey, 2017). Indeed, both general celebrities (athletes, actors,
etc.) gone digital and so-called social media influencers (people who have become famous
via social media) attract an increasing number of followers (Abidin, 2015; Giles, 2017;
Kapitan & Silvera, 2016; Marwick, 2015), thus increasing their share of voice.
Communication professionals and marketers have realized the potential of collaborating with
these celebrities, and consequently the terms influencer communication and influencer
marketing have been widely embraced to refer to the appearance of branded goods in the
Not only information about products and services, but also ideas, ideologies and
public information can be communicated through the channels of social media celebrities to
very specific, and often devoted, audiences (Khamis, Ang & Welling, 2017; Senft, 2008).
Thus, for organizations, social media celebrities provide a new channel through which they
can purposefully communicate to reach their goals (Hallahan, Holtzhausen, van Ruler,
Verčič, and Sriramesh, 2007) and to to frame topics identified as strategic and substantial
(Zerfass, Verčič, Nothhaft & Werder, 2018). The potentially great reach and impact of social
media celebrities, as well as their lending of symbolic meaning and reputation (McCracken,
1989; Aula & Heinonen, 2016), can be crucial to successful participation in the ongoing
nothing novel. Celebrity endorsements have typically been studied in branding and
advertising research (e.g., Erdogan, 1999; Kamins, 1990; Keel & Nataraajan, 2012; Ohanian,
A CALL FOR AUTHENTICITY 3
1990). However, the current practice of influencer communication challenges many of the
basic tenets of the traditional celebrity endorsement model. Social media celebrities come
from a much wider spectrum of people than the celebrities who have traditionally been
chosen for celebrity endorsement campaigns. Many of the most influential social media
celebrities are known by a very small share of people in a given region while general
celebrities are usually known by relatively many (Abidin, 2015). Past research has studied
celebrity endorsements extensively as a part of the marketing process (e.g., Hsu &
McDonald, 2002; Ohanian, 1990; Silvera & Austad, 2004), but research in the area of
strategic communication remains scarce, and thus we lack understanding of the ways in
which celebrity endorsement can be used to advance the survival and sustain the success of
This study investigates the effectiveness of the content that social media celebrities
often linked to attributes such as genuinity, truthfulness and originality (Molleda, 2010), all
authenticity as a perceived feature of a celebrity but also through attributes of their social
media content that attempt to manifest authenticity. Empirically, we focus on photos that
celebrities have posted on Instagram, a popular social media channel. We use a survey
method to study the effect of celebrity characteristics, type of celebrity, congruence between
a photo and the celebrity’s style, as well as photo sponsorship and sponsor disclosure on the
followers’ attitudes towards the photo and purchase intentions. The respondents are sampled
among the followers of a number of different celebrities, which increases the generalizability
In the following sections, we will first introduce and discuss celebrity endorsement as
a strategic communication practice and then move on to formulate our hypotheses based on
A CALL FOR AUTHENTICITY 4
existing research. Then, we will explain our methodological approach and the regression
models, and present the findings of the statistical analyses. Finally, we will discuss the
findings in connection with strategic communication literature and provide future research
allows marketers to take advantage of celebrities promoting their products (Erdogan, 1999).
A celebrity endorser can be an artist, a movie star, or any other public figure who uses their
recognition and image to make paid product endorsements (McCracken, 1989). Research
widely suggests that celebrity endorsements have positive effects on brand recall, brand
attitudes, purchase decisions, product sales, and stock value (Djafarova & Rushworth, 2017;
Elberse & Verleun, 2012; Kamen, Azhari & Kragh, 1975; Kamins, 1990; Mathur, Mathur &
Rangan, 1997). According to McCracken (1989) and Roy (2016), the effectiveness of a
symbolic meanings first reside in the celebrity, and, through the endorsement, they transfer to
a product, and, from the product to the consumer. Recently, similar approaches have been
adopted beyond marketing in areas such as CSR communication, lobbying, and strategic
traditional media. First, social media blends the content of one’s peers and the content of
celebrities, which can increase the attention that the celebrities receive as opposed to
traditional advertisements (Giles, 2017; Kapitan & Silvera, 2016). Moreover, in social media,
the user has explicitly decided to follow the celebrity and is able to interact with him or her
(Khamis et al., 2017; Senft, 2008), and the content is distributed and usually also created by
the celebrity (Abidin, 2015; Kapitan & Silvera, 2016). A celebrity endorser traditionally
A CALL FOR AUTHENTICITY 5
works only with a few select brands (Hsu & McDonald, 2002), but many social media
influencers make their living from sponsored endorsements with various partners (Colliander
& Erlandsson, 2013). These factors have arguably increased the rate and changed the nature
the social media platform (Hou, 2018; Thomas, 2014); they must constantly negotiate
between the pressures of being interesting and credible, as well as economically profitable,
2016; Audrezet, de Kerviler & Moulard, 2018), a property of communication that has gained
interest in areas such as advertising, communication studies, marketing, and public relations
(e.g., Audrezet et al., 2018; Brown, Kozinets & Sherry, 2003; Molleda & Roberts, 2008).
Authenticity has been identified as an important but also an elusive concept: the term is used
to describe a variety of objects and behaviors, and is often perceived differently by different
parties. In sociology and social psychology, the concept relates to self-presentation and
conceptions of the self, meaning “the feeling and practice of being true to one’s self or
others” (Vannini & Franzese, 2008, p. 1621). Authenticity can thus be understood through
different definitions, see Molleda, 2010). Similar conceptions have been presented in
celebrity studies, in which authentic celebrity brands are seen to appear genuine in their
consumer relationships when they behave according to their values (Ilicic & Webster, 2016;
identify with the message and the sender (Brown et al., 2003). In strategic corporate
it allows organizations to operate and take a stance in open, undefined areas as opposed to
shaping audience perceptions within narrowly defined contexts (Molleda & Roberts, 2008).
Authenticity is, however, not a straightforward attribute, in particular for a corporate actor,
whose communications always come with a less or more hidden agenda (Gilmore & Pine,
2007). Molleda (2010) reminds us that the context of communication and the technology used
to communicate affect what the recipient perceives to be authentic; various cues can be used
to craft an authentic identity and voice, which are then subjectively judged and negotiated by
and with the audience (Beverland & Luxton, 2005; Grayson & Martinec, 2004; McLeod,
1999). Arguably, as sociotechnical systems, social media have many contextual features that
allow (or deny) authentic communications between an organization and its stakeholders. In
the case of celebrity endorsements, social media provides a way for brands to be presented
authentically through the celebrity’s own words, pictures or videos rather than through
From the perspective of authentic communication and its strategic planning, celebrity
endorsements lead to the existence of two distinct but interdependent relationships: one
between the organization and the celebrity, and another between the celebrity and his or her
audience. Macnamara and Zerfass (2012) argue that there is a lack of general knowledge
regarding the ways in which organizations seek to utilize social media in their strategic
and the online philosophies of openness, participation and democratization. The dilemma
becomes even more evident when a social media celebrity participates in the framing of the
message; the organization should strategically decide with whom it can and should cooperate
with, and in what ways the delivery of the messages can be outsourced and opened to
external communicators. At the same time, social media celebrities themselves are constantly
negotiating between their own credibility and the sponsored content they post (Abidin & Ots,
A CALL FOR AUTHENTICITY 7
2016; Audrezet et al., 2018). Partly because of this problem, celebrities seek authenticity by
“staging” it (Hou, 2018; Thomas, 2014). They purposefully share intimate and private photos,
interact with their followers and strive for a sense of equality and commonality (Abidin,
2015). All these considerations make organizations’ collaboration with social media
Next, we utilize the ideas presented above to make hypotheses regarding the effect of
communication (Hovland, Janis & Kelley, 1953) and of strategic communication (de Vries,
Gensler & Leeflang, 2012; Wang, Qiao & Peng, 2015), while purchase intention and
purchases are currently the most common organizational goals of sponsored social media
content—and therefore reasonably operationalized for the study of actual social media
content. By and large, we expect that the effects on photo attitudes are parallel with purchase
intentions (Spears & Singh, 2004; Wang et al., 2015), despite some inconsistencies in the
that explain the influence of a celebrity endorser. The most well-known and repeatedly-used
model was developed by Roobina Ohanian (1990), and is composed of three main
regarding their relative importance (cf. Deshpandé & Stayman, 1994; Ohanian, 1990; 1991;
Pornpitakpan, 2004). Other studies have emphasized the mediating role of social
social media celebrities (Gong & Li, 2017; Ilicic & Webster, 2016; Jin, 2018; Labrecque,
2014). The main claim put forward by these studies is that celebrity endorsements are
effective because they facilitate a feeling of belonging and social connectedness with their
audiences.
meta-analysis, see Knoll & Matthes, 2016), the exploration of social media celebrities,
particularly in the context of visual platforms such as Instagram, has only recently begun to
interest researchers (e.g., Djafarova & Trofimenko, 2018; Gong & Li, 2017; Phua, Lin &
Lim, 2018). It is reasonable to expect differences in the dynamics in which the effectiveness
of celebrity attributes is played out. For example, instead of being a meritocratic attribute,
expertise in social media is a cumulative construct based on the thematic branding and
communication strategies pursued by the celebrity (Tolson, 2013). Therefore, rather than
factors that are expected to be particularly relevant in social media: perceived authenticity
and attractiveness.
media celebrities, as well as for corporate branding more generally. Studies have explored the
effects and outcomes of brand authenticity, showing, for example, that brand authenticity
positively impacts brand trust (Schallehn, Burmann & Riley, 2014) and consumers’
behavioral intentions (Fritz, Schoenmueller & Bruhn, 2017). Further, Ilicic and Webster
(2016) showed that the perceived authenticity of a celebrity was a more important predictor
of consumers’ purchase intentions than familiarity with the celebrity. It is assumed in the
particularly important attribute as the channel is abundant with sponsored endorsements and
other advertisements (Audrezet et al., 2018). Therefore, we suggest the following hypothesis:
A CALL FOR AUTHENTICITY 9
attractiveness has a positive impact on how influential he or she is (Chaiken, 1979). With
regards to celebrity endorsement, the evidence is inconsistent. Kahle and Homer (1985) and
purchase intentions, while Ohanian (1991) and Silvera and Austad (2004) found that
within certain themes and topics, social media is highly focused on users’ looks and
appearances. Djafarova and Rushworth (2017) found that, for some users, the visual
attractiveness and the quality of photos is one of the main reasons to follow a celebrity on
Instagram. It therefore stands to reason that celebrity attractiveness has a positive effect on
Celebrity type. Compared to the traditional forms of media or the traditional logic of
the entertainment industry, social media has created various opportunities for individual
actors to gain fame (Labrecque, Markos & Milne, 2011; Khamis et al., 2017). Various terms
have been used to describe people who have become famous through social media:
microcelebrity (Senft, 2008; Marwick, 2013), the instafame (Marwick, 2015) and vlogger
celebrity (Hou, 2018), for instance. In this study, we use the term social media influencer
(Abidin, 2015; Khamis et al., 2017) to emphasize the origin of the fame without restricting
the person’s area of influence to any specific channel or taking a stand on the size of the
A CALL FOR AUTHENTICITY 10
influence. Several authors suggest that these persons influence consumer preferences more
authenticity and credibility (de Vries et al., 2012; Giles, 2017; Djafarova & Rushworth,
2017).
general celebrities, such as actors, athletes and models. Social media influencers often
represent different styles and demographics than general celebrities (Marwick, 2015), and
their communication style is usually more personal and direct (Keel & Nataraajan, 2012).
These differences are also reflected in how consumers identify with them and react to their
social media content. Giles (2017) showed that social media influencers achieve an important
role as legitimizing agents within their field through so-called micro-celebrity practices.
Djafarova and Rushworth (2017) suggest that social media influencers have a particularly
strong influence on the purchase behavior of young female users in Instagram. Further,
researchers have identified a mediating effect of social identification with the endorser (Jin &
Phua, 2014; De Veirman, Cauberghe & Hudders, 2017), which is arguably higher in the case
of social media influencers due to the close relationship they often have with their followers
(see also Gong & Li, 2017; Verhellen, Dens & De Pelsmacker, 2013). Therefore, we
hypothesize:
H3: Compared to general celebrities’ photos, social media influencers’ photos are
more positively related to followers’ (a) photo attitudes and (b) purchase intentions.
consistent in one’s behavior and true to one’s character (e.g., Vannini & Franzese, 2008;
Marwick, 2013). In the context of celebrity endorsement, the assessment of such consistency
A CALL FOR AUTHENTICITY 11
is expanded to the relationship between the celebrity and the product being endorsed.
Advertising research has repeatedly found that a celebrity endorsement is effective when
there is a good fit, or congruence, between the endorser and the product (Kahle & Homer,
1985; Kim & Na, 2007; Kamins, 1990; Till & Busler, 1998). For example, a tennis player is a
good endorser of tennis rackets. Research also shows that lack of congruence is not only a
missed opportunity but can also have negative effects on brand recall (Evans, 1988).
Moreover, a lack of a relationship between the endorser and the product leads consumers
more likely to believe that the endorsement was bought than in the case of appropriate
product-celebrity pairs (Silvera & Austad, 2004). Perceived congruence has been found to be
in effect even when a consumer is unfamiliar with the endorser or the product (McCormick,
2016). Further, a recent study of celebrity endorsements on Instagram showed that mere
image congruence between a celebrity and a product improved consumer responses (Phua et
al., 2018). In other words, the product does not need to be directly associated with the
celebrity’s profession, for example, but it needs to fit an abstract idea of what the celebrity is
like and what his or her style is. Based on these considerations, we hypothesize:
H4: Perceived congruence between a social media celebrity’s style and her photo is
positively related to followers’ (a) photo attitudes and (b) purchase intentions.
endorsements in social media and product placements in entertainment media among both
consumers and policy makers (Cain, 2011; Dekker & Reijmersdal, 2013). There is some
research evidence that doing so would also be preferable from the organization’s perspective.
Nebenzahl and Secunda (1993) found that consumers have positive attitudes towards product
placements when they are transparent. Consumers may even prefer sponsored online content
over alternative forms of on-screen advertisements due to their unobtrusive nature (Gupta &
A CALL FOR AUTHENTICITY 12
Gould, 1997; Nelson, Keum & Yaros, 2004; Tutaj & van Reijmersdal, 2012). In the case of
sponsored product endorsements in blogs, research shows that consumers are indifferent as to
whether the blogger has received cash or a product discount for an endorsement (Lu, Chang
& Chang, 2014). Some studies, on the other hand, show that disclosing a sponsorship causes
consumers to develop distrustful beliefs about the content (Boerman, Willemsen & Van Der
Aa, 2017).
In many cases, however, the consumer is unable to know if a celebrity has received a
compensation for endorsing a product, and it is suggested that there are numerous
undisclosed sponsored endorsements (cf. Roettgers, 2016). Thus, if a person is unaware that a
product endorsement in social media has been sponsored, he or she may easily perceive it as
basis on which consumers make purchase decisions (e.g., Cheung & Thadani, 2012; Erkan &
Evans, 2016) so it can be argued that consumers are more strongly influenced by such
endorsements if sponsors are left undisclosed. We therefore posit the following hypothesis:
H6: Compared to sponsored photos that reveal the sponsor, a sponsored photo that
does not reveal the sponsor is more positively related to followers’ purchase
intentions.
A CALL FOR AUTHENTICITY 13
Method
Celebrity survey
The social media celebrities for the study were chosen from a Finnish Instagram
marketing agency’s influencer network. The agency is focused on hosting influencers who
post mainly fashion, beauty, home décor and family-themed photos. The agency collaborates
with celebrities who have at least 2000 followers. For the purpose of this study, two persons
who work at the agency and know the celebrities well categorized them into general
celebrities (actors, models, etc.) and social media influencers (bloggers, Instagram stars, etc.).
In many cases, the social media influencers have gained fame beyond social media but the
The survey was sent to 75 social media celebrities by email in November 2016. The
celebrities were asked to answer questions regarding each of their nine latest photos,
excluding the very newest photo (for which the number of likes was still likely to change).
Among other things, we asked if the photos had been a part of a commercial collaboration
(i.e., if they had been sponsored) either through the agency or otherwise. The researchers
collected what information they could via Instagram (number of followers, number of likes
per photo).
45 social media celebrities in all responded to the survey (response rate 60%). 93% of
the respondents were women, and 7% were men. The distribution largely aligns with the
gender distribution of all the celebrities the agency cooperates with. The average age of the
respondents was 30.1 years (SD = 6.8). Their average number of followers was 23,997 (SD =
35,432). On average, the photos within the sample had attracted 1841 likes each (SD = 3203).
Of the celebrities, 24% were categorized as general celebrities and 76% social media
influencers. The average number of followers was 43,122 (SD = 38,854) for general
A CALL FOR AUTHENTICITY 14
celebrities and 17,810 (SD = 31,950) for social media influencers, and the difference is
Of the photos, 75.8% had not been sponsored, 16.3% had been sponsored but with the
sponsor not explicitly disclosed, and 7.9% had been sponsored with the sponsor disclosed. As
signs of disclosure, we accounted for texts and hashtags such as “ad”, “collaboration”,
“sponsored”, either in Finnish or in English (at the time of the study, Instagram did not allow
the marking of a sponsored photo as an application feature on top of the photo). There was no
significant difference in photo sponsorship (X2(1) = .305, p = .581, N = 405) and, among
sponsored photos, sponsor disclosure between general celebrities and social media
influencers (X2(1) = 1.476, p = .224, N = 98). Figure 1 presents three example photos that
were part of the sample. Most of the photos in the sample followed similar aesthetics.
FIGURE 1 HERE
Follower survey
A follower survey was conducted between November 2016 and February 2017. Using
the Instagram inbox, a survey was sent to 90 randomly selected followers of each social
media celebrity who had responded. The same nine photos were presented to the respondents
who were asked to answer questions about the celebrity and each of the nine photos. For each
photo, the survey showed the name of the Instagram account, the photo and its caption, and
the number of likes (as shown in example photos in Figure 1). With the survey, we measured
on a 1–7 Likert-type scale the followers’ perceptions about celebrity authenticity and
attractiveness (Vannini & Franzese, 2008; Ilicic & Webster, 2016; Ohanian, 1990), photo-
celebrity congruence (Kim & Na, 2007), photo attitudes (Silvera & Austad, 2004) and
purchase intentions invoked by the photos (Erkan & Evans, 2016). The item wordings were
made to fit the specific research context. Cronbach’s alphas were computed to evaluate the
construct reliability of the scales. All varied between .850 and .942, indicating good
A CALL FOR AUTHENTICITY 15
reliability (Malhotra, Birks & Wills, 2012). See Table 1 for a summary of the survey
The 4050 follower surveys attracted responses from 592 respondents (response rate
13.4%). The average age of the respondents was 25.6 years (SD = 8.8), 88% were women,
and 12% men. According to Omnicore (2017), approximately 68% of Instagram users are
women. Nevertheless, as most of the celebrities in the sample were women, the
TABLE 1 HERE
explaining followers’ photo attitudes using all the photo evaluations (N = 5119) and a model
explaining followers’ purchase intentions using only those evaluations that concerned
sponsored photos (N = 1312). We used hierarchical multiple regression to model the data,
which met the assumptions for the analysis. Neither non-linearity nor heteroscedasticity was
identified in the residuals of the research variables (Hair, Anderson, Tatham & Black, 1998).
When testing for multicollinearity, no problems arose based on Tolerance and VIF statistics
(thresholds > .10 and < 10, respectively) (Hair et al., 1998).
Results
Followers’ photo attitudes were modelled first (see Table 2). In the first step of the
model, the number of likes a photo had received (logarithmic scale) was included in the
model as a control variable. The results indicate that the number of likes had a small positive
relationship with followers’ photo attitudes (β = .068, p < .01). Then, in the second step,
Authenticity and attractiveness both had a positive effect (β = .227, p < .01, β = .240, p < .01,
respectively). Hence, H1a and H2a are supported. Celebrity type also had a positive, yet
small, relationship with photo attitudes, which was higher for social media influencers’
A CALL FOR AUTHENTICITY 16
photos (β = .029, p < .05). This finding provides support for H3a. In the third step, photo-
related variables were included in the model. Photo-celebrity congruence related strongly
with photo attitudes (β = .529, p < .01), which indicates support for H4a. Finally, the results
suggest that photo sponsorship had a negative relationship with photo attitudes (β = -.056, p <
.01). This supports H5; non-sponsored photos are perceived more favorably than sponsored
photos. Including the photo-related variables in the model diluted the effect of the celebrity
type, which suggests that, finally, there was no direct relationship between celebrity type and
photo attitudes. Thus, H3a is not supported. The final model explained 41.5% of the
TABLE 2 HERE
With regards to purchase intention, we included only sponsored photos in the model
(see Table 3). First, the results suggest that there was no significant relationship between the
number of likes and followers’ purchase intentions. In the second step, photo-related
variables were included in the model, which improved R2 by 11.6%. Both perceived
authenticity and attractiveness had a positive relationship with purchase intentions (β = .254,
p < .01 and β = .118, p < .01, respectively), thus supporting H1b and H2b. Social media
influencers’ sponsored photos were more positively related with purchase intentions than
general celebrities’ sponsored photos (β = .136, p < .01), which supports H3b. In the third
step, photo-celebrity congruence had the strongest relationship with purchase intentions (β =
.458, p < .01), meaning that H4b is supported. Sponsor disclosure did not have a significant
relationship with purchase intentions. H6 is thus not supported. Finally, including the photo-
related variables in the model diluted the direct effect of perceived attractiveness, and H2b is
therefore not supported. The final model accounted for 29.4% of the variability of purchase
intention.
TABLE 3 HERE
A CALL FOR AUTHENTICITY 17
photo attitude and purchase intention, Figures 2 and 3 illustrate their mean differences. As
shown, photo attitude remained the same for social media influencers regardless of whether a
photo was sponsored or not, while general celebrities’ non-sponsored photos were more
favorably perceived than their sponsored photos. Purchase intention was consistently higher
in case of social media influencers’ sponsored photos than general celebrities’ sponsored
photos, regardless of whether the sponsor was disclosed. The followers still considered
sponsor-disclosing photos more commercial than those that did not disclose the sponsor,
suggesting that they paid attention to the disclosure. Measured with a statement “I think
commercial content is central in this photo”, the sponsor-disclosing photos were more highly
rated (M = 5.67, SD = 1.68) than photos that did not disclose the sponsor (M = 4.82, SD =
1.79) (F(1,1332) = 64,832, p < .01, N = 1334). It is also worth noting that perceived
authenticity was not dependent on celebrity type. Both general celebrities (M = 5.61, SD =
1.05) and social media celebrities (M = 5.36, SD = 1.15) were perceived as authentic
Discussion
As with many previous studies, we found that certain celebrity characteristics have an
effect on how favorably the audience perceives them and the endorsements they make.
Perceived authenticity was found to have a positive effect on both photo attitudes (all photos)
and purchase intentions (sponsored photos). Attractiveness, on the other hand, had a positive
effect on photo attitudes, which seems reasonable considering the large role of visual
elements and appeal in Instagram. Attractiveness did not, however, have any effect on
(Pornpitakpan, 2004; Till & Busler, 1998). The accentuated effect of authenticity on purchase
intentions supports the idea that, in order to be effective, the sponsored content that a
celebrity posts should align with his or her ideals and values. Therefore, in accordance with
existing studies on the relevance of authenticity for social media celebrities (e.g., Thomas,
2014; Audrezet et al., 2018), our results highlight the importance of the ‘aura of authenticity’
Second, congruence between the style of the photo and the celebrity had the strongest
positive relationship with photo attitudes and purchase intentions, which is in line with
previous research findings (Keel & Nataraajan 2012; Knoll & Matthes, 2017; McCormick,
2016; Phua et al., 2018). Photo-celebrity congruence may be the dimension that captures the
essence of photo authenticity and makes the photo genuine and credible. The follower knows
what kind of content the celebrity has posted before, so if a new photo is not consistent with
that style, he or she reacts negatively. Further, observations on congruence are in line with the
existing literature that emphasizes authenticity both for celebrity brands (Ilicic & Webster,
2016) and brands in general (Molleda, 2010; Molleda & Roberts, 2008): communication is
more effective when the image of the celebrity also remains consistent with regard to the
products endorsed. The congruence effect is present not only for sponsored content, but also
for the full set of images tested in the attitude model, which, together with the large effect
size, indicates that congruence is an important aspect when followers evaluate visual
messages.
and non-disclosing photos). This is unsurprising given the positive attitudes that consumers
have towards spontaneous and real WOM messages (Cheung & Thadani, 2012; Erkan &
Evans, 2016). However, an interesting finding is that, among the sponsored photos,
indicated by our second model. This finding is supported by Nebenzahl and Secunda (1993),
who argue that there are no strong objections when product placements are disclosed as
advertisements. Lu and colleagues (2014) suggest a similar effect in the case of sponsored
blog posts. According to the present study, the same applies to followers of celebrities on
Instagram—neither the celebrity nor the sponsor benefits from not disclosing the sponsor
Traditionally, celebrity endorsement literature has studied the effect of well-known big stars
(Erdogan, 1999), but, in the context of social media, regular-seeming people can also hold
celebrity status. Thus, the opportunities for influence have become more fragmented. In the
current study, sponsored photos of social media influencers elicited higher purchase
intentions than sponsored photos of general celebrities. With regards to photo attitudes, the
celebrity type did not have an effect, indicating that general celebrities’ photos also contain
interesting content. The findings suggest, thus, that when it comes to behavioral intentions,
social media influencers’ endorsements are more relevant than general celebrities’
endorsements, as also suggested by Djafarova and Rushworth’s (2017) interview study. The
results also lead us to believe that the influence of social media influencers is based on other
factors than what existing research has determined in the context of traditional celebrities and
Moreover, as social media influencers arguably have a closer relationship with their
followers as their fame has arisen from social media, the effectiveness of their endorsements
may also be based on the effectiveness of WOM communications (Cheung & Thadani, 2012;
Erkan & Evans, 2016; Jin, 2018). Followers may regard social media influencers’ sponsored
endorsements as more like peer endorsements, which increases their effectiveness. The fact
that most social media influencers are rarely featured in mass media might accentuate this
A CALL FOR AUTHENTICITY 20
perception (Labrecque, 2014). It thus seems that social media influencers might enable
identification more easily than general celebrities (Gong & Li, 2017; Jin & Phua, 2014) and
that the transfer of desired attributes is more effective for the messages endorsed by social
media influencers rather than general celebrities (McCracken, 1989; Roy, 2016). These
observations make social media influencers attractive for strategic communication, them
being ‘people like me’ readily present in the digital environments in which the stakeholders
While the findings of this study carry several implications, some limitations need to
be addressed. First, there is not yet an established framework to measure the influence of
social media celebrities, nor influence in social media more generally. In this study, we
followed recent scholarship on social media celebrities and focused on the source
(e.g., Ohanian, 1990). Our results show that the characteristics selected are significant
that more research and scale development is needed in order to define the characteristics that
influence followers of social media celebrities, and followers of social media influencers in
particular. It also remains unclear how the perceptions of social media influencers develop in
the future if they are featured in mass media more often, reach even more followers and start
Further, one dependent variable in this study was purchase intention, and it is clear
that this variable does not equal realized purchase decisions, even if the former often has
predictive power over the latter (De Cannière, De Pelsmacker & Geuens, 2009). So, the
findings regarding purchase intentions need to be interpreted with care. Future research
should study real purchase decisions, for example by utilizing the specific discount codes that
A CALL FOR AUTHENTICITY 21
social media celebrities often use to track consumers’ purchases. Moreover, strategic
communication research could benefit from other dependent variables as well, such as more
would also account for other than commercial contexts, for example political or health
expect that social media celebrities will be used more frequently in non-commercial areas of
communication in the future. However, for this study, we chose to focus on purchase
intention because it fits to the current mainstream practices of celebrity endorsement in social
media.
As the present study did not account for the visual content of the photos studied,
future research could focus on the content of the visual messages in greater detail.
Furthermore, after the completion of the survey, a feature called “Stories” (short videos
available only for a limited time) became popular on Instagram, and future research should
consider the effect of such content as well. Further, as product endorsements have become
more commonplace in social media, research should explore the accepted rate of sponsored
content in one celebrity’s social media feed. Does the effectiveness of an endorsement
decline as more sponsored endorsements appear and does photo-celebrity congruence, for
endorsement and influencer collaboration. The findings illustrate the change in how influence
ramifies in our mediated society; social media celebrity endorsements are a prime example of
a novel mechanism to promote products, worldviews and ideologies in contexts that were
originally considered and promoted as spaces for genuine peer communication and
empowerment (e.g., Borchers, 2016; Curran, 2012). Several contributions to the area of
A CALL FOR AUTHENTICITY 22
strategic communication can be identified. First, we found that social media influencers’
photos are more influential than general celebrities’ photos when it comes to purchase
intentions evoked. It is assumed that the social media influencers generate a greater sense of
social identification (Djafarova & Rushworth, 2017; Gong & Li, 2017), and thus they gain
more defined and relevant followers than general celebrities, which improves the impact of
their endorsements. This means organizations will most likely benefit from strategic
On the other hand, social media influencers typically have fewer followers than
number of social media influencers in order to reach as large of an audience. Moreover, while
social media influencers have relatively higher influence among their own social media
followers, it is likely that their influence is significantly lower among the general population,
who might not even recognize them. Therefore, if the company intends to use a celebrity
endorser in mass media advertisements in addition to sponsored social media photos, general
celebrities (with a good message-celebrity match) can be better choices as they are known to
a wider population. Our results indicate, however, that social media influencer endorsements
work effectively if aimed to reach specific stakeholders for a given strategic communication
goal.
both photo attitudes and purchase intentions. The importance of a match between the
celebrity and the product has been shown in previous studies, and our findings show that the
importance of congruence also applies in the context of Instagram and to the style of the
photo, rather than just the product. Hence, in order to achieve the communication goals and
also to address the right stakeholders, organizations need to select the celebrities they work
with carefully and ensure that the sponsored photos and messages fit the celebrity’s usual
A CALL FOR AUTHENTICITY 23
style. This is also an important implication for the celebrities, who engage in strategic
All in all, our findings highlight that the strategic planning of influencer relationships
should account for the type of the celebrity and the match between the celebrity profile and
the content he or she produces. Authenticity in the celebrity profile should be highlighted
more than the person’s attractiveness. Further, our findings show that sponsored photos were
less favorably perceived than non-sponsored photos. Therefore, celebrity endorsement and
influencer collaboration are not shortcuts to effective communication, and strategic planning
References
Abidin, C., & Ots, M. (2016). Influencers tell all? Unravelling authenticity and credibility in
a brand scandal. In M. Edström, A. T. Kenyon & E.-M. Svensson (Eds.), Blurring the
Gothenburg: Nordicom.
Audrezet, A., De Kerviler, G., & Moulard, J. G. (2018). Authenticity under threat: When
Aula, P., & Heinonen, J. (2016). The Reputable Firm: How Digitalization of Communication
Beverland, M., & Luxton, S. (2005). Managing integrated marketing communication (IMC)
through strategic decoupling: How luxury wine firms retain brand leadership while
Boerman, S. C., Willemsen, L. M., & Van Der Aa, E. P. (2017). “This post is sponsored”:
Brown, S., Kozinets, R. V., & Sherry, J. F. Jr. (2003). Teaching old brands new tricks: retro
branding and the revival of brand meaning. Journal of Marketing, 67(3), 19–33.
De Cannière, M. H., De Pelsmacker, P., & Geuens, M. (2009). Relationship quality and the
Chaffey, D. (2017, November 23). Global social media statistics summary 2017. Smart
marketing/social-media-strategy/new-global-social-media-research/
Colliander, J., & Erlandsson, S. (2015). The blog and the bountiful: Exploring the effects of
disguised product placement on blogs that are revealed by a third party. Journal of
Curran, J. (2012). Reinterpreting the Internet. In Curran, J., Fenton, N., & Freedman, D.
Deshpandé, R., & Stayman, D. M. (1994). A tale of two cities: Distinctiveness theory and
Djafarova, E., & Rushworth, C. (2017). Exploring the credibility of online celebrities’
doi:10.1080/1369118X.2018.1438491
Elberse, A., & Verleun, J. (2012). The economic value of celebrity endorsements. Journal of
Erkan, I., & Evans, C. (2016). The influence of eWOM in social media on consumers’
Fitch, K. (2017). Seeing ‘the unseen hand’: Celebrity, promotion and public relations. Public
Fritz, K., Schoenmueller, V., & Bruhn, M. (2017). Authenticity in branding – exploring
51(2), 324–348.
Giles, D. C. (2017). How do fan and celebrity identities become established on Twitter? A
study of ‘social media natives’ and their followers. Celebrity Studies, 1–16.
Gilmore, J. H., & Pine, B. J. (2007). Authenticity: What Consumers Really Want. Boston:
Gong, W., & Li, X. (2017). Engaging fans on microblog: the synthetic influence of parasocial
Grayson, K., & Martinec, R. (2004). Consumer perceptions of iconicity and indexicality and
Gupta, P. B., & Gould, S. J. (1997). Consumers’ perceptions of the ethics and acceptability of
Hair, J. F., Anderson, R. E., Tatham, R. L. & Black, W. C. (1998). Multivariate Data
Hallahan, K., Holtzhausen, D., Van Ruler, B., Verčič, D., & Sriramesh, K. (2007). Defining
35.
A CALL FOR AUTHENTICITY 27
Hartmans, A. (2018). The 50 most followed Instagram accounts in 2018. Business Insider,
1–20.
Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and Persuasion. New
Ilicic, J., & Webster, C. M. (2016). Being true to oneself: Investigating celebrity brand
Jin, S. V. (2018). “Celebrity 2.0 and beyond!” Effects of Facebook profile sources on social
Jin, S. A. A., & Phua, J. (2014). Following celebrities’ tweets about brands: The impact of
43(2), 181–195.
Kahle, L. R., & Homer, P. M. (1985). Physical attractiveness of the celebrity endorser: A
Kamen, J. M., Azhari, A. C., & Kragh, J. R. (1975). What a spokesman does for a sponsor.
advertising: When beauty may be only skin deep. Journal of Advertising, 19(1), 4–13.
A CALL FOR AUTHENTICITY 28
Kapitan, S., & Silvera, D. H. (2016). From digital media influencers to celebrity endorsers:
Keel, A., & Nataraajan, R. (2012). Celebrity endorsements and beyond: New avenues for
Khamis, S., Ang, L., & Welling, R. (2017). Self-branding,‘micro-celebrity’ and the rise of
Kim, Y. J., & Na, J. H. (2007). Effects of celebrity athlete endorsement on attitude towards
the product: The role of credibility, attractiveness and the concept of congruence.
Knoll, J., & Matthes, J. (2017). The effectiveness of celebrity endorsements: a meta-analysis.
28(2), 134–148.
Labrecque, L. I., Markos, E., & Milne, G. R. (2011). Online personal branding: Processes,
Lu, L.-C., Chang, W.-P., & Chang, H.-H. (2014). Consumer attitudes toward blogger’s
product type, and brand awareness. Computers in Human Behavior, 34, 258–266.
Macnamara, J., & Zerfass, A. (2012). Social media communication in organizations: The
Malhotra, N., Birks, D., & Wills, P. (2012). Marketing Research: An Applied Approach (4th
Marwick, A. E. (2015). Instafame: Luxury selfies in the attention economy. Public Culture,
27(1), 137–160.
Marwick, A. E. (2013). Status Update: Celebrity, Publicity, and Branding in the Social
Mathur, L. K., Mathur, I., & Rangan, N. (1997). The wealth effects associated with a
McLeod, K. (1999). Authenticity within hip-hop and other cultures threatened with
Molleda, J. C., & Roberts, M. (2008). The value of “authenticity” in “glocal” strategic
Molleda, J. C. (2010). Authenticity and the construct's dimensions in public relations and
Moulard, J. G., Garrity, C. P., & Rice, D. H. (2015). What makes a human brand authentic?
32(2),
Nebenzahl, I. D., & Secunda, E. (1993). Consumers’ attitudes toward product placement in
Nelson, M. R., Keum, H. & Yaros, R. A. (2004). Advertainment or adcreep game players’
39–52.
Omnicore (2017). Instagram by the numbers: Stats, demographics & fun facts. Retrieved
from https://www.omnicoreagency.com/instagram-statistics/
Park, S., & Cho, M. (2015). Celebrity endorsement for nonprofit organizations: The role of
Phua, J., Lin, J. S., & Lim, D. J. (2018). Understanding consumer engagement with celebrity-
93–102.
Roettgers, J. (2016, August 22). Kardashians in trouble over paid product endorsements on
instagram-paid-ads-product-placements-1201842072/
381.
Schallehn, M., Burmann, C., & Riley, N. (2014). Brand authenticity: model development and
Senft, T.M. (2008). Camgirls: celebrity and community in the age of social networks. New
Silvera, D. H., & Austad, B. (2004). Factors predicting the effectiveness of celebrity
Spears, N., & Singh, S. N. (2004). Measuring attitude toward the brand and purchase
Sudha, M., & Sheena, K. (2017). Impact of influencers in consumer decision process: the
Thomas, S. (2014). Celebrity in the “Twitterverse”: history, authenticity and the multiplicity
Till, B. D., & M. Busler (1998). Matching products with endorsers: Attractiveness versus
Chouliaraki (Ed.), Self-Mediation – New Media, Citizenship and Civil Selves (pp. 93–
Tutaj, K., & van Reijmersdal, E. A. (2012). Effects of online advertising format and
18(1), 5–18.
Vannini, P., & Franzese, A. (2008). The authenticity of self: Conceptualization, personal
De Veirman, M., Cauberghe, V., & Hudders, L. (2017). Marketing through Instagram
Verhellen, Y., Dens, N., & De Pelsmacker, P. (2013). Consumer responses to brands placed
in Youtube movies: the effect of prominence and celebrity endorser expertise. Journal
De Vries, L., Gensler, S. & Leeflang, P. S. H. (2012). Popularity of brand posts on brand fan
Wang, Y., Qiao, F., & Peng, W. (2015). Is the size or the valence of proactive engagement
Zerfass, A., Verčič, D., Nothhaft, H., & Werder, K. P. (2018). Strategic communication:
Defining the field and its contribution to research and practice. International Journal
Table 1
Item Scale
Cronbach’s
Construct Item M SD M SD Source
α
Questions about the celebrity in general. "I think this person is…"
Authenticity Genuine. 5.12 1.39 5.18 1.18 .926 Vannini & Franzese,
Photo-celebrity This photo fits well 5.76 1.38 5.64 1.31 .850 Kim & Na, 2007
account’s overall
style.
photo.
Photo attitude I like this photo. 5.53 1.52 5.31 1.49 .884 Silvera & Austad,
interesting.
Purchase I get ideas for new 3.92 1.95 4.29 1.70 .853 Erkan & Evans,
photo.
purchases.
buying something I
Table 2
β t ∆F ∆R2
Overall R2 = .415
Adjusted R2 = .415
Table 3
evaluations, N = 1312)
β t ∆F ∆R2
Overall R2 = .297
Adjusted R2 = .294