SciELO - Scientific Electronic Library Online

 
vol.53 número3Editorial: the revise & resubmit (R&R) processTaxation, corporate governance and dividend policy in Brazil índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados

Journal

Artigo

Indicadores

Links relacionados

Compartilhar


RAUSP Management Journal

versão On-line ISSN 2531-0488

RAUSP Manag. J. vol.53 no.3 São Paulo jul./set. 2018

http://dx.doi.org/10.1108/rausp-04-2018-011 

Research Paper

The effectiveness of celebrity endorsement in aspiring new celebrities: Examining the effects of brand, congruence, charisma and overexposure

Otávio Freirea  * 

Filipe Quevedo-Silvab 

Diego Senise and Pedro Scrivanoc 

aUniversidade de São Paulo, São Paulo, Brazil and Universidade Nove de Julho, São Paulo, Brazil

bSchool of Business and Administration, Universidade Federal de Mato Grosso do Sul, Campo Grande, Brazil

cUniversidade de São Paulo, São Paulo, Brazil

Abstract

Purpose

Celebrity endorsement is a phenomenon widely used by companies and studied by researchers. Despite the generally positive aspects of endorsement on the evaluation of products, in some cases, celebrities cannot substantially help promote products. The purpose of this paper is to analyze the effectiveness of celebrity endorsement effect in an aspiring new celebrity.

Design/methodology/approach

Four studies involving 664 respondents were conducted to analyze the effectiveness of the endorsement. In addition to verifying the existence of the effect (Study 1), studies were conducted focusing on contextual variables with the potential to modify effect. Study 2 analyzes the brand and charisma effects, Study 3 analyzes the congruence between celebrities and Study 4 analyzes the exposure level of the endorser celebrity.

Findings

The results demonstrate the effectiveness of this celebrity endorsement, as well as the effect of different contexts on endorsement, demonstrating that although new celebrities are less affected by acclaimed celebrities, as they become better known, to use the endorsement of celebrities with charisma and who relating in some way to the aspiring celebrity, can be an effective strategy, especially for the beginners in the career.

Originality/value

This research contributes to the knowledge of celebrity endorsement to fill the lack pointed out in previous studies in the field over the effectiveness of this effect and, above all, the moderator variables that can influence or even annul this effect. In addition, by developing its own image and reputation, the aspiring new celebrity receives less influence from the endorser.

Keywords Celebrity endorsement

Introduction

Traditionally, the concept of celebrity can be understood as a well-known person recognized by the public and considered a role model for society (McCracken, 1989), due to professional competences or physical appearance (Kahle and Homer, 1985). Very famous actors, models, athletes and singers are all considered celebrities (Friedman and Friedman, 1979). Knowing the influence that celebrities exert over consumers, companies are increasingly investing in their use in advertising, to persuade consumers (McNamara, 2009).

Studies in marketing have been conducted to investigate how celebrity endorsement impacts consumer purchase intentions (Wei and Lu, 2013). Also, several authors have investigated a series of related factors, such as the effect of celebrity endorsement on product recall, the effect of the credibility and expertise of the celebrity endorsement and the effect of celebrity image (Amos et al., 2008).

Despite the generally positive aspects of endorsement on the evaluation of products, which has been verified in other studies (Silva et al., 2015; Ambroise et al., 2014), in some cases, celebrities cannot substantially help promote products (Sliburyte, 2009). They may even damage a brand, if consumers learn that an existing endorser has committed a transgression (Um, 2013). These contrasting results demonstrate how celebrity endorsement is complex and can be influenced by several variables.

A meta-analysis performed by Knoll and Matthes (2017) show that studies about the subject have found positive, negative and neutral effects, leaving in doubt the actual effectiveness of celebrity endorsement. The authors demonstrated that the differences found in the studies may have occurred due to moderating variables, such as the endorser type (Wei and Lu, 2013), congruence (Silva et al., 2015) or endorser sex (Bergkvist and Zhou, 2016). Knoll and Matthes (2017) suggest that future studies should focus precisely on the conditions under which the endorsement effect may change: for example, endorsement of non-conventional products (Myrick and Evans, 2014), endorsement of persons (Van Steenburg, 2015), non-profit endorsement (Wheeler, 2009) and endorsement in emerging countries (Chou, 2014) such as China, India or Brazil, all factors which will be addressed in this study.

Along with the lack of studies noted by Knoll and Matthes (2017), there has been another global phenomenon related to endorsement and celebrity theory as a whole. With the broad popularization of the internet, social networks, social media and reality television, it has been proposed that there has been an extension of the concept of celebrity, so that through these new media, ordinary people can also achieve such status. As a result, the number of aspiring celebrities has grown (Keel and Nataraajan, 2012). Therefore, knowing that the effectiveness of celebrity endorsement effect may vary completely depending on the context of the endorsement, and given the phenomenon of new celebrities emerging from different media, the purpose of this paper is to analyze the effectiveness of an “acclaimed” celebrity endorsement effect in a context where an aspiring new celebrity is being endorsed.

In line with recent research propositions for this area (Knoll and Matthes, 2017), to analyze the effectiveness of the endorsement and verify the existence of the effect (Study 1), studies were conducted focusing on contextual variables with the potential to modify effect. Study 2 analyzes the brand and charisma effects (Kowalczyk and Royne, 2013; Louie et al., 2001); Study 3 analyzes the congruence between celebrities (Fleck et al., 2012); and Study 4 measures the exposure level of the endorsing celebrity (Carrillat et al., 2013).

The results demonstrate the effectiveness of this celebrity endorsement, as well as the effect of different contexts on endorsement. They show that although new celebrities are less affected by endorsements from established celebrities as they become better known, employing the endorsement of celebrities with charisma and who are related in some way to the aspiring celebrity can be an effective strategy, especially for newcomers on the scene.

Celebrity endorsement

Traditionally, the concept of celebrity can be understood as a person who is recognized by the public and considered to be a role model for society (McCracken, 1989), due to his or her professional competence or beauty (Kahle and Homer, 1985). Celebrities can inspire consumers’ desires, hopes and dreams (Rockwell and Giles, 2009). Furthermore, celebrities can work as the personification of a brand, creating bonds and facilitating connections with consumers (Thomson, 2006).

Despite the positive aspects of endorsements, the use of celebrities in campaigns may or may not be effective (Misra and Beatty, 1990). Celebrities have been used to positively influence consumer behavior toward a product (Tripp et al., 1994), and advertising campaigns create a link between the product and the celebrity, causing a transfer of meaning that can either be positive or negative to the product (Till and Shimp, 1998).

To select an appropriate celebrity, the advertiser or marketer should take into account, among other factors, the longevity of the campaign, the acceptance of the celebrity and their relevance, the target consumer’s opinion on the use of celebrities for communication and the receptivity of the target consumer when associating the celebrity with the product or brand. Several factors are considered when choosing a celebrity endorser or brand sponsor, when attempting to establish a campaign for the brand:

  • fame – wide recognition by consumers or by specific groups;

  • adjustment – combination or match between brand attributes and celebrity attributes regarding consumer perceptions;

  • financial features – costs and returns from the use of celebrities as endorsers; and

  • roles – different ways to use celebrities in marketing communications (Pringle and Binet, 2005).

Celebrity endorsement efficacy is sustained by credibility, expertise and attractiveness (Ohanian, 1991). Credibility refers to the confidence that the celebrity conveys to the public; expertise is linked to the knowledge and experience that the endorser has on a certain subject; and attractiveness is associated with physical appearance, beauty and sympathetic nature.

When consumers believe that the endorser reflects their idealized self-concept and self-image, the evaluation of the advertisement is positive and increases product purchase intentions (Choi and Rifon, 2012). In addition to this positive assessment, such an endorsement makes these consumers more likely to show loyalty to the brand or product.

Studies also show that the use of celebrities in campaigns is linked to various strategies. Rumschisky (2009) found that people are willing to pay up to 20 per cent more for a product, depending on who endorses it, generating greater revenues for the company. Advertisements featuring celebrities tend to increase the value of the company on the stock exchange, as such advertisements also influence investors’ perceptions of the endorsed company (Agrawal and Kamakura, 1995).

The effect of celebrity endorsements on the evaluation of products, demonstrating that celebrities influence their fans, has also been studied (Sliburyte, 2009; McNamara, 2009). Researchers found that much of this effect results from consumers’ associations between the celebrity and the endorsed object (Choi and Rifon, 2012; Till and Shimp, 1998).

New social media platforms such as Facebook, Instagram and Twitter, as well as reality television programming, are reaching a worldwide audience. These media have provided a new method for creating aspiring celebrities (Keel and Nataraajan, 2012). In reality, music shows, for example, celebrities from the world of music both judge and coach the contestants. It can be conceptualized that established celebrities eventually “lend” their images to candidates who choose them as mentors. The dynamics of the shows reinforce this concept by presenting the candidates as part of the celebrity teams. At all times, the association between participant and celebrity is presented to the public. Because the contestants are aspiring celebrities and unknown to the public up to that point, being associated with a celebrity with an established image can improve their assessment. Similarly, a new product launched on the market with the endorsement of a celebrity with credibility in the area can also improve its reception. Thus, it is expected that:

H1.

The evaluation of aspiring new celebrities will be more positive when a well-known celebrity endorses them.

Study 1

The purpose of Study 1 is to analyze whether the effect of celebrity endorsement also occurs in the evaluation of an aspiring new celebrity, as it does with tangible products. The stimuli were defined by the researchers. To give greater validity to the study, all studies were conducted during season two of the program The Voice Brazil, in which candidates compete for the preference of the public while seeking successful careers in the music industry. They are divided into teams led by established singers who act as coaches. Thus, in this program, candidates seeking to become new celebrities are valued by the public (potential consumers of their music) and are endorsed by the celebrity that chooses them to compete for their teams.

The authors used one of the program’s candidates and the celebrity that chose to coach the candidate. An online questionnaire was used, and the link was distributed on an online panel provided and maintained by a national research institute. In all studies of this research project, the first section of the instrument presented stimuli for manipulation. Respondents evaluated the candidate using a Likert scale ranging from 1 (Completely Disagree) to 7 (Strongly Agree) points. Demographic data from the respondents were collected. The final sample of this first study was 100 viewers, who were assigned to a 2 (celebrity endorsement: With endorsement vs No endorsement) × 1 (Participant of the program The Voice Brazil) condition. First, participants had to answer questions about their involvement with the program, and people who did not watch, or did not know the participant, were excluded from the study. All spectators assessed the participant through a series of questions. As all candidates on The Voice usually have a coach, to select participants for the condition “no endorsement”, after evaluating the candidate they had to select who the candidate’s coach was from a list of celebrities. The participants who failed to choose the correct coach were assigned to the condition “no endorsement”. Participants who chose correctly associated the candidate with his or her celebrity coach (“endorsement”). To assess the candidates, we used scales to measure an intention to consume music (α = 0.908) and performance in the program (α = 0.925).

Study 1 – results

The study was conducted with 193 viewers of the show, with an average age of 28.3 (19-62). In accordance with the hypothesis of Study 1, the aspiring candidates for becoming a new celebrity were evaluated more favorably when they were endorsed by an established celebrity (their coach or technical advisor during the program), both for the intention to consume their music (Mwith endorsement = 5.8 (1.4) and = Mwithout endorsement 5.2 (1.3); p < 0.05), and performance in the program (Mwith endorsement = 5.3 (1.6) and Mwithout endorsement = 4.4 (1.7); p < 0.01) (Figure 1).

Figure 1. Evaluation of candidates with and without endorsement 

Study 1 – discussion

The results show that the effect of celebrity endorsement can also occur in a context where the evaluated object is a person aspiring to be a future celebrity. Despite the generally positive aspects of endorsement on the evaluation of general products, as has been verified in other studies (Choi and Rifon, 2012; Silva et al., 2015), in some cases celebrities cannot substantially help promote products (Knoll and Matthes, 2017; Sliburyte, 2009). These contrasting results demonstrate the complexity of celebrity endorsement, and reinforce the relevance of verifying this effect in a context of an acclaimed celebrity and an aspiring new celebrity.

According to the results, the celebrity endorsement influences not only the assessment of the candidate in the program but also the intention to consume future work (music and television shows) by this candidate. This is another indication of the effectiveness of this type of endorsement. According to Knoll and Matthes (2017), the effectiveness of celebrity endorsement has been discussed in recent years in terms of positive and negative effects, and the influence of this effect is subject to several moderating variables, such as congruence, charisma (Silva et al., 2015) and endorser sex (Bergkvist and Zhou, 2016), which demonstrates the importance of studies that address different aspects and scenarios of this phenomenon.

Charisma and brand effects on celebrity endorsement

Another dimension of the endorsement effect is the influence that the brand has on this phenomenon. A brand aims, among other factors, to differentiate the product from other offerings and to add value, contributing to achieving competitive advantage (Keller and Lehmann, 2006; Aaker, 1991). Strong brands grant greater awareness, consciousness and consumer loyalty (Rust et al., 2004). In addition, for consumers, product quality is not necessarily related to the technical characteristics of the product, but to an intangible assessment of the brand in question, thereby turning the brand into a key element in the process of purchase decisions (Hoeffler and Keller, 2003).

People can also be considered brands (Aaker, 1991). Their assets, as in the case of product brands, can be built through recognition and remembrance, positive associations, perceived quality in the relationship and loyalty level generated by the relationship. Brands can also reflect human dimensions of personality, based on the evaluation of consumers (Aaker, 1997). If brands can have human characteristics, it can be inferred that people can be brands and that these dimensions would be prominent.

Consumers generally perceive and react to stimuli generated by people just as they do in relation to brands. This phenomenon is triggered by automaticity and by the level of evaluation for that particular object – in this case, people and brands (Alba and Hutchinson, 1987). If this occurs, the object captures consumers’ attention, assisting in the processing and selection of information to be considered in the perception of the object. For celebrities, this phenomenon is further evidenced by the fact that they are role models for society in general (McCracken, 1989). Prominent celebrities in their professional areas do not only lend their names to endorse products and services; their own production can be understood as endorsed by their names. Also, they may create new products and extend their product lines (Kowalczyk and Royne, 2013; Luo et al., 2010).

The development and exponential growth of new media, combined with the diversification and growth of new media programming formats, has allowed the emergence of new celebrities (Keel and Nataraajan, 2012). Series on free and subscription TV channels; comedy and music programs on YouTube; and reality shows based on simulations of real-life events such as musical and sporting competitions, have captivated audiences, making room for actors, musicians, athletes and participants in these reality shows to be considered aspiring celebrities, thus achieving certain social recognition. In line with Alba and Hutchinson (1987); Aaker (1991, 1997); Kowalczyk and Royne (2013) and Luo et al. (2010), these aspiring celebrities may also be considered brands, with greater or lesser degrees of recognition.

Recognized brands are considered superior and better evaluated than unrecognized brands or more poorly evaluated brands, with or without the use of celebrity endorsement (Silva et al., 2015). According to the authors, a well-evaluated, recognizable brand does not need celebrity endorsement because the effect of such endorsement would not add significant value to brands perceived as strong by consumers.

In a reality show, candidates are aspiring celebrities and are also building their images and their brands (Keel and Nataraajan, 2012), some with more ownership and competence than others. Given that the image of the celebrity endorser influences the product – and a negative image can negatively influence the evaluation of the product (Louie et al., 2001), but a recognized brand can change the endorsement effect (Silva et al., 2015) – it is expected that:

H2.

A weak candidate (worst evaluated brand) endorsed by a celebrity with a positive image will be better evaluated than a weak candidate endorsed by celebrity with a negative image.

H3.

The evaluation of a strong candidate (best evaluated brand) does not depend on the image of the celebrity endorser.

Study 2

The purpose of Study 2 is to analyze whether the effect of the brand on celebrity endorsement also occurs in the evaluation of an aspiring new celebrity, and whether the influence of the celebrity image has an effect. For this study, we used a 2 × 2 full factorial design between subjects (Candidates: Strong vs weak) × (Celebrity Endorser: well evaluated celebrity vs poorly evaluated celebrities). For data analysis, we used ANOVA.

The stimuli were defined by the researchers, and Sam Alves was chosen as the contender (Strong Brand), for being one of the most popular candidates and most downloaded on the official website of the program. Gabby Moura was chosen as the poor candidate (Weak Brand) because she is less popular and has a significantly lower number of downloads on the official website of the program. Claudia Leitte, who was the technical advisor of both candidates, was used as the celebrity endorser. The instrument was similar to that used in Study 1.

The final sample of Study 2 was 152 individuals. As in Study 1, spectators first answered questions about their involvement with the program, and people who did not watch or did not know the participants were excluded from the study. Participants then assessed the candidate using a series of questions. To check celebrity influence on candidates, the participants of this study also evaluated the celebrity. We used the same scales as Study 1 for candidates’ assessment, and an adapted attitude scale (α = 0.986) to evaluate the celebrity.

Study 2 – results

Study 2 was conducted with 152 viewers with an average age of 27.7 (18-60). The manipulation check confirmed that Sam Alves is perceived as a stronger candidate than Gabby Moura (MSam Alves = 5.5 (1.4) and MGabby Moura = 4.4 (1.6); p < 0.01). Regarding the assessment of the celebrity, respondents were divided by the median to form the two groups (well evaluated celebrity vs poorly evaluated celebrity). The celebrity was assessed with an adapted attitude scale (α = 0.986). We found 80 respondents with a positive assessment (well evaluated celebrity) and 72 with negative assessment (poorly evaluated celebrity). Confirming H2 of this study, the weak candidate (weak brand) was better assessed on intention to consume her music when spectators positively evaluated the celebrity endorser (positive attitude toward celebrity) (Mwith Well Evaluated Celebrity = 5.6 (1.0) and Mwith Poorly Evaluated Celebrity = 4.3 (1.6); p < 0.01), as compared to her performance in the program (Mwith Well Evaluated Celebrity = 5.1 (1.5) Mwith Poorly Evaluated Celebrity = 3.4 (2.0); p < 0.01).

Concerning the strong candidate (strong brand), there was no difference whether spectators perceived his advisor as a good or bad endorser. This effect was observed both on intention to consume his music (positive Mwith Well Evaluated Celebrity = 5.8 (1.3) and Mwith Poorly Evaluated Celebrity = 5.4 (1.4); p = NS) and performance in the program (Mwith Well Evaluated Celebrity = 5.1 (1.7) and Mwith Poorly Evaluated Celebrity = 4.6 (1.7); p = NS), regardless of the attitude toward the celebrity endorser, confirming H3 (Figure 2).

Figure 2. Candidates’ evaluations with celebrity endorsement with negative and positive image 

Discussion of Study 2 – results

The results of Study 2 show that the strongest candidate’s evaluation was the same regardless of the image rating of the celebrity. The findings of this study are similar to those found by Silva et al. (2015), where the brand factor – in this case, the candidate’s own image – moderates the celebrity endorsement effect. On the other hand, the weakest candidate (worst evaluated brand) was better evaluated when the endorsing celebrity coach was rated as having a positive image by the participant, rather than a negative image, confirming H2 of this study.

In addition, when considering the scenario where the endorsing celebrity has a positive image, the evaluation of the weak endorsed candidate rises to the same level as the strong participant.

This happens for both evaluation in the program and intention to consume the candidate’s products, demonstrating the effectiveness of the endorsement and its greater importance in cases in which the new celebrity has not yet established a strong image.

Based on this result, professionals interested in the new celebrity should propose strategies of dissemination associated with the celebrity endorser, and preferably with the current fans of this celebrity, which would enhance the effect of the endorsement.

The effect of different levels of congruence on celebrity endorsement

Celebrity choice should be guided by the congruence that he or she has with the product endorsed. The concept of congruence is known as finding the best match between the product and any associated variable. The level of congruence refers to how consistent are the endorser’s most relevant features with the product’s most important attributes (Fleck et al., 2012; Huston et al., 2003; Misra and Beatty, 1990).

Congruence depends on two dimensions: relevance and expectation. Relevance reflects the extent to which the stimulus information contributes to a clear identification of the theme or main message being communicated (Heckler and Childers, 1992). Expectation refers to the expected degree to which an item or piece of information fits in a predetermined pattern or structure evoked by the subject. Several authors have demonstrated the relevance of the positive relationship between the celebrity endorser and the advertised product (Batra and Homer, 2004; Misra and Beatty, 1990).

An aspiring celebrity candidate in a musical reality show presents a set of characteristics that must somehow fit in with a certain musical style, which indicates their level of professionalism (Kahle and Homer, 1985). In addition, there are also aspects related to attractiveness and credibility (Ohanian, 1991), which may or may not be considered congruent with their technical advisor in the program, i.e. the celebrity endorser (Fleck et al., 2012). Thus, it is expected that:

H4.

A weak candidate (worst rated) endorsed by a congruent celebrity will be better assessed than a weak candidate endorsed by an incongruent celebrity.

H5.

The evaluation of a strong candidate (best rated) does not depend on congruence with the celebrity endorser.

Study 3

The purpose of Study 3 was to analyze whether different celebrity congruence levels also affect the evaluation of an aspiring new celebrity. For this study, we used a full factorial design between subjects, 2 (candidates: Strong vs Weak) × 2 (Celebrity endorser: Congruent vs Incongruent). For data analysis, we used ANOVA and ANCOVA.

As in Study 2, Sam Alves was chosen as the strong contender and Gabby Moura as the weak candidate. The scenarios of congruence level were manipulated by having spectators read a text showing the congruence (or incongruence) between the candidate and the celebrity that was coaching him or her during the show. For example, in one of the scenarios, the text emphasized how candidate and coach shared the same musical style. The instrument was similar to that used in previous studies.

The final sample of this study was 176 spectators. As in previous studies, participants had to answer questions about their involvement with the program, and people who did not watch or did not know the participants were excluded from the study. The stimulus was presented, and participants assessed the candidates on the same scales as previous studies.

Study 3 – results

Study 3 was conducted with 176 viewers of the show, with an average age of 29.1 (18-62). Manipulation check confirmed that Sam Alves is perceived as stronger than Gabby Moura [MSam Alves = 5.5 (1.6) and = MGabby Morua 4.5 (1.6); p < 0.01], and that participants in the congruent scenario evaluated the celebrity as more congruent with the candidate than participants in the incongruent setting [MCongruent Endorser = 4.9 (2.1) and MIncongruent Endorser = 3.9 (2.0); p < 0.01].

There was no difference in the evaluation of the weak participant (worst evaluated) concerning the congruence level with the endorser celebrity. However, to isolate the effect of attitude toward the celebrity, we conducted an ANCOVA. This test showed that the candidate Gabby Moura (weak candidate) was better evaluated when spectators perceived a more positive congruence level with the endorser, either for intention to consume her music [MCongruent Endorser = 5.0 (1.4) and MIncongruent Endorser = 4.7 (1.5); p < 0.01], or performance in the program [MCongruent = 4.0 (1.8) and MIncongruent Endorser = 3.5 (1.7); p < 0.01], confirming H4 of this study.

However, and as expected, the congruence level with the celebrity endorser had no effect on the strongest participant (best rated). Neither intention of consuming his music [MCongruent Endorser = 5.6 (1.5) and MIncongruent Endorser = 5.4 (1.4); p = NS] nor performance in the program [MCongruent Endorser = 4.7 (1.9) and MIncongruent Endorser = 4.8 (1.6); p = NS] showed significant differences, regardless of the celebrity endorser congruence level, thus confirming H5 of this study (Figure 3).

Figure 3. Candidates’ evaluation with different congruence levels of endorsement 

Discussion of Study 3 – results

The results of the study show that the weaker candidate was better evaluated when there was congruence between her and the celebrity endorser. However, to understand this phenomenon, it was necessary to isolate the effect of attitude toward celebrity, indicating that – despite the fact that congruence affects the assessment in the case of an aspiring new celebrity – celebrity endorser image plays a key role in the celebrity endorsement effect. Regarding the strongest candidate, the evaluation was the same regardless of congruence with celebrity endorser.

This result demonstrates how the endorsement effect can be complex. When choosing a celebrity endorser, it is advisable for consumers to perceive a fit between new celebrity and celebrity. This can be achieved through musical style, regionality, physical appearance, etc. However, the current image of the celebrity endorser should be taken into account when choosing the endorser, as a bad reputation (Louie et al., 2001) may decrease the effectiveness of the endorsement or even reverse this effect.

The effect of overexposed celebrity endorsement

Congruence and charisma effects on celebrity endorsement are essential in terms of advertising effectiveness (Fleck et al., 2012). However, a celebrity involved in negative events may affect the image of the endorsed product (Louie et al., 2001). The impact of these events is directly reflected in the company’s stock returns. Also, the greater the celebrity’s perceived guilt in the episode, the greater will be the influence on their endorsement effectiveness (Carrillat et al., 2013; Louie et al., 2001). Another common fact within the historical context in risk strategy is the loss of relevance that celebrities can suffer with the media after the endorsement contract has been signed (Luo et al., 2010). The image of celebrities can change abruptly, diluting the relationship that existed when the brand association began (Erdogan et al., 2001).

When the celebrity endorses only one product, the endorsement is evaluated more positively and respondents indicate a greater interest in buying the product (Mowen and Brown, 1981). Consumers also show a more favorable attitude toward the advertisement, as compared to celebrities who endorse many products. “Multiple endorsements” refer to two relatively common occurrences in advertising: a celebrity who lends his or her endorsement to many products, and a product that relies on the endorsement of many celebrities in its advertising campaign (Rice et al., 2012). Studies on celebrity overexposure in the media also show that, even if there is congruence between brand and celebrity, the more products that are endorsed by a single celebrity, the less effective the endorsement is to the brand (Kaikati, 1987). Other studies (Subhadip, 2012; Ilicic and Webster, 2011; Hsu and Mcdonald, 2002; Tripp et al., 1994) show that there is a negative influence on the perception of credibility of the endorser, attitude toward the ad and intention to purchase products endorsed by overexposed celebrities, once the credibility of the celebrity and the validity of his or her appeal becomes questionable (Carrillat et al., 2013).

In the context of a reality show, if the coach – an established celebrity endorser – is perceived as someone who considers all candidates as favorites, indiscriminately, this endorsement to the aspiring celebrity may lose its strength. This phenomenon was already observed in the context of products and brands (Rice et al., 2012; Carrillat et al., 2013; Ilicic and Webster, 2011; Subhadip, 2012; Tripp et al., 1994; Kaikati, 1987; Mowen and Brown, 1981). If the celebrity endorser is careful in his or her statements of favoritism, the power of endorsement remains. On the other hand, previous studies already showed that stronger candidates may have already established their image, thereby moderating the overexposure effect of his or her coach.

In line with studies on overexposure of celebrity endorsers, we developed H6 and H7 for this study:

H6.

A weak candidate (worst rated) endorsed by an overexposed celebrity will be evaluated worse than a weak candidate endorsed by celebrity without overexposure.

H7.

The evaluation of a strong candidate (best rated) does not depend on the level of overexposure of the celebrity endorser.

Study 4

The purpose of Study 4 is to analyze whether celebrity endorser overexposure affects the evaluation of an aspiring new celebrity. For this study, we used a full factorial design between subjects, 2 (candidates: Strong vs Weak) × 2 (Celebrity Endorser: Overexposed vs Not Overexposed). For data analysis, we used ANOVA.

As in Studies 2 and 3, Sam Alves was used as the strong contender and Gabby Moura as the weak candidate. Overexposure was manipulated by designing a scenario in which the celebrity chose many candidates for her team (endorsing a large number of candidates) for an overexposed condition, and a scenario where the celebrity chose a few candidates for her team (endorsing few candidates) for a not overexposed condition. The instrument was similar to that used in previous studies.

The final sample of this study was 236 people. As in previous studies, first, the participants answered questions about their involvement with the program, and people who did not watch or did not know the participants were excluded from the study. The stimulus was presented and spectators assessed the candidate through a series of questions, using the same scales as previous studies.

Study 4 – results

Study 4 was conducted with 236 viewers of the show, with an average age of 30.6 (18-58). Manipulation check confirmed that Sam Alves is perceived as a stronger candidate than Gabby Moura [MSam Alves = 5.6 (1.5) and MGabby Moura = 4.7 (1.5); p < 0.01]. Also, participants in the overexposed celebrity condition perceived more over-exposure when compared to the participants in the not overexposed condition [MCelebrity Overexposed = 4.6 (1.8) and MCelebrity Not Overexposed = 3.4 (1.7); p < 0.01].

Confirming H6 of this study, the weak candidate (worst rated) received lower ratings in both intention to consume her music when endorsed by an overexposed celebrity [MGabby Moura+Celebrity Overexposed = 4.9 (1.4) and M Gabby Moura+Celebrity Not Overexposed = 5.4 (1.3); p < 0.10], and performance in the program [M Gabby Moura Celebrity Overexposed = 3.7 (1.8) and M Gabby Moura+Celebrity Not Overexposed = 4.6 (1.7); p < 0.05]. Also, and as expected, the strong participant (best rated) was not affected by the overexposure of his coach in the program. No difference can be observed, either on intention to consume his music [MSam Alves+Celebrity Overexposed = 5.2 (1.5) and MSam Alves+Celebrity Not Overexposed = 5.6 (1.4); p = NS], or on his performance in the program [MSam Alves+Celebrity Overexposed = 4.6 (1.7) and MSam Alves+Celebrity Not Overexposed = 4.8 (1.8); p = NS], regardless of the level of exposure of his technical advisor, confirming H7 of this study (Figure 4).

Figure 4. Candidates’ evaluation with different levels of celebrity endorser overexposure 

Discussion of Study 4 – results

The results of the study show that the weakest candidate was better evaluated when the celebrity endorser was not perceived as overexposed. In a context of overexposure, the endorsement of an established celebrity loses power, showing a significant difference from the lowest exposure context. As for the strongest candidate, the assessment was the same regardless of the level of exposure of the endorser celebrity used. The results support the hypotheses and attest that level of overexposure has an effect on the endorsement, as well as on the context of products and brands, as confirmed in previous studies. However, it is important to notice that when the aspiring new celebrity is successful in rapidly developing a strong image for the audience, this has the same effect as a strong brand, thereby moderating the negative effect of an endorsement by an overexposed celebrity.

General discussion

The purpose of this study was to analyze the effectiveness of an “acclaimed” celebrity endorsement effect in a context where the endorsed object is another person – more specifically, an aspiring new celebrity. In addition, aligned with a proposal for future studies from Knoll and Matthes (2017), we sought to understand the effects of brand, charisma, congruence and celebrity exposure on the effectiveness of the endorsement, and therefore on the spectators’ attitudes and behavioral intentions in this context. To operationalize the study, the authors chose the phenomenon of reality shows which, along with other new media, such as Facebook, Instagram and Twitter, have attracted huge audiences, allowing and contributing to the emergence of new celebrities (Keel and Nataraajan, 2012).

Study 1 showed that when endorsed by an established celebrity, an aspiring new celebrity was better rated than when he or she lacked endorsement. This was revealed both in the assessment of the aspiring celebrity’s performance on the reality show The Voice Brazil, and in the reported intention to purchase the work of this future celebrity (music recordings and shows), confirming the effect of the endorsement. Previous studies had already investigated the endorsement effect on products (Kim and Na, 2007; Silva et al., 2015), but this is the first study to show that celebrity endorsement effect can also be effective for an aspiring new celebrity.

Study 2 demonstrated two effects. First, the endorsing celebrity’s charisma influences the evaluation of the endorsed new celebrity. Another result of this study relates to brand influence on the evaluation, based on the understanding of the aspiring celebrity as brand (Kowalczyk and Royne, 2013; Luo et al., 2010). For a strong candidate (strong brand), celebrity charisma did not change the overall assessment. However, in examining the effect of the endorsement on the weak candidate, it is perceived that the evaluation rises to the same level as the other candidate when the endorsing celebrity is evaluated positively. This demonstrates the effectiveness of the endorsement and its effect enhanced by a contextual variable, in this case, the charisma of the endorser.

In Studies 3 and 4, it was also possible to identify the effects of congruence (Fleck et al., 2012; Batra and Homer, 2004) and celebrity overexposure (Ilicic and Webster, 2011; Subhadip, 2012). In both studies, it was once again possible to identify the influence of contextual factors on the endorsement, in a context of established celebrities endorsing aspiring celebrities.

Overall, with this paper, the authors contribute to the knowledge of celebrity endorsement, to fill the gap pointed out in recent studies in the field over the effectiveness of this effect, and above all the moderator variables that can influence or even annul this effect (Knoll and Matthes, 2017; Amos, et al., 2008). A celebrity can endorse an aspiring new celebrity, and this is enhanced by variables related to the endorser, such as charisma and level of exposure. However, the choice of the endorser cannot be made without taking into account the characteristics of the endorsed. There must be some degree of association between them for the endorsement to be effective. In addition, by developing their own image and reputation, aspiring new celebrities are less subject to influence from the endorser. The effect of the endorsement still exists, but to a lesser extent and with less influence from the endorsing figure. For example, in Study 2, the endorser’s charisma (celebrity character) did not change the endorsee’s evaluation. This result is interesting and contributes to studies about brand (Silva et al., 2015), and in particular those focused on the possible effects of the brand on endorsers (Thomas and Fowler, 2015).

From a practitioner standpoint, reality shows such as The Voice and other similar formats can be seen as a source of potential future celebrities. In recent years, it has been normal in the USA for program participants to top iTunes download charts, demonstrating this potential. Thus, knowing how to manage their career from the very beginning can be decisive to guarantee both longer and profitable careers. Aspiring celebrities that are not very well known by spectators can benefit from celebrity endorsement. Music labels and other companies of this type can use this information to develop strategies to leverage new artists in partnerships with their established artists. However, it is important to verify whether both congruence and charisma are present, as well as no overexposure. When choosing a celebrity to endorse another, it is necessary to emphasize the possible associations between them for the public, and constantly monitor the reputation of the celebrity vis-à-vis consumers and ensure that there is no overexposure of the celebrity, as the effect of the endorsement will be less effective if the same celebrity endorses several aspirants to a new celebrity. On the other hand, if an aspiring celebrity manages to build a strong image with spectators, he or she can be seen as an established brand, and thus will be less and less affected by the endorsement, and in time will have the potential to be an endorser.

Future research must vary the profile of the celebrity endorser and aspiring new celebrities. The profile should be tested in other substantive areas, such as new media and social media (Facebook, Instagram, Twitter, YouTube, etc.), reality shows and new entertainment programming formats. Moreover, it would be important for other researchers to work with different kinds of brands and categories of products and services.

References

Aaker, D. (1991), Managing Brand Equity, The Free Press, New York, NY. [ Links ]

Aaker, J.L. (1997), “Dimensions of brand personality”, Journal of Marketing Research, Vol. 34 No. 3, pp. 347-356. [ Links ]

Agrawal, J. and Kamakura, W.A. (1995), “The economic worth of celebrity endorsers: an event study analysis”, Academic Research Library, Vol. 59 No. 3, pp. 56-62. [ Links ]

Alba, J.W. and Hutchinson, J.W. (1987), “Dimensions of consumer expertise”, Journal of Consumer Research, Vol. 13 No. 4, pp. 411-454. [ Links ]

Ambroise, L., Pantin-Sohier, G., Valette-Florence, P. and Albert, N. (2014), “From endorsement to celebrity co-branding: personality transfer”, Journal of Brand Management, Vol. 21 No. 4, pp. 273-285. [ Links ]

Amos, C., Holmes, G. and Strutton, D. (2008), “Exploring the relationship between celebrity endorser effects and advertising effectiveness. A quantitative synthesis of effect size”, International Journal of Advertising, Vol. 27 No. 2, pp. 209-234. [ Links ]

Batra, R. and Homer, P.M. (2004), “The situational impact of brand image beliefs”, Journal of Consumer Psychology, Vol. 14 No. 3, pp. 318-330. [ Links ]

Bergkvist, L. and Zhou, K.Q. (2016), “Celebrity endorsements: a literature review and research agenda”, International Journal of Advertising, Vol. 35 No. 4, pp. 642-663. [ Links ]

Carrillat, F.A., D’astous, A. and Lazure, J. (2013), “For better, for worse? What to do when celebrity endorsements go bad”, Journal of Advertising Research, Vol. 53 No. 1, pp. 15-30. [ Links ]

Choi, S.M. and Rifon, N.J. (2012), “It is a match: the impact of congruence between celebrity image and consumer ideal self on endorsement effectiveness”, Psychology & Marketing, Vol. 29 No. 9, pp. 639-650. [ Links ]

Chou, H.Y. (2014), “Effects of endorser types in political endorsement advertising”, International Journal of Advertising, Vol. 33 No. 2, pp. 391-414. [ Links ]

Erdogan, B., Baker, Z. and Tagg, M.J. (2001), “Selecting celebrity endorsers: the practitioner’s perspective”, Journal of Advertising Research, Vol. 41 No. 3, pp. 39-48. [ Links ]

Fleck, N., Korchia, M. and Le R, I. (2012), “Celebrities in advertising: looking for congruence or likability?”, Psychology & Marketing, Vol. 29 No. 9, pp. 651-662. [ Links ]

Friedman, H. and Friedman, L. (1979), “Endorser effectiveness by product type”, Journal of Advertising Research, Vol. 19 No. 5, pp. 63-71. [ Links ]

Heckler, S.E. and Childers, T.L. (1992), “The role of expectancy and relevancy in memory for verbal and visual information: what is incongruency?”, Journal of Consumer Research, Vol. 18 No. 4, pp. 475-492. [ Links ]

Hoeffler, S. and Keller, K.L. (2003), “The marketing advantages of strong brands”, Journal of Brand Management, Vol. 10 No. 6, pp. 421-445. [ Links ]

Hsu, C. and Mcdonald, D. (2002), “An examination on multiple celebrity endorsers in advertising”, Journal of Product & Brand Management, Vol. 11 No. 1, pp. 19-29. [ Links ]

Huston, C.R., D’ouville, E. and Willis, T.H. (2003), “Race and gender effects on consumer identification of products with celebrities endorsers”, The Marketing Management Journal, Vol. 13 No. 1, pp. 85-99. [ Links ]

Ilicic, J. and Webster, C.M. (2011), “Effects of multiple endorsements and consumer-celebrity attachment on attitude and purchase intention”, Australasian Marketing Journal, Vol. 19 No. 4, pp. 230-237. [ Links ]

Kahle, L. and Homer, P.M. (1985), “Physical attractiveness of the celebrity endorser: a social adaptation perspective”, Journal of Consumer Research, Vol. 11 No. 4, pp. 954-961. [ Links ]

Kaikati, J.G. (1987), “Celebrity advertising a review and synthesis”, International Journal of Advertising, Vol. 6 No. 2, pp. 93-105. [ Links ]

Keel, A. and Nataraajan, R. (2012), “Celebrity endorsements and beyond: new avenues for celebrity branding”, Psychology & Marketing, Vol. 29 No. 9, pp. 690-703. [ Links ]

Keller, K.L. and Lehmann, D.R. (2006), “Brands and branding: research findings and future priorities”, Marketing Science, Vol. 25 No. 6, pp. 740-759. [ Links ]

Kim, Y. and Na, J. (2007), “Effects of celebrity athlete endorsement on attitude towards the product: the role of credibility, attractiveness and the concept of congruence”, International Journal of Sports Marketing & Sponsorship, Vol. 8 No. 4, pp. 310-320. [ Links ]

Knoll, J. and Matthes, J. (2017), “The effectiveness of celebrity endorsements: a meta-analysis”, Journal of the Academy of Marketing Science, Vol. 45 No. 1, pp. 55-75. [ Links ]

Kowalczyk, C.M. and Royne, M.B. (2013), “The moderating role of celebrity worship on attitudes toward celebrity brand extensions”, Journal of Marketing Theory and Practice, Vol. 21 No. 2, pp. 211-220. [ Links ]

Louie, T., Kulik, R. and Jacobson, R. (2001), “When bad things happen to the endorsers of good products”, Marketing Letters, Vol. 12 No. 1, pp. 13-23. [ Links ]

Luo, L., Chen, X., Han, J. and Park, C.W. (2010), “Dilution and enhancement of celebrity brands through sequential movie releases”, Journal of Marketing Research, Vol. 47 No. 6, pp. 1114-1128. [ Links ]

McCracken, G. (1989), “Who is the celebrity endorser? Cultural foundations of the endorsement process”, Journal of Consumer Research, Vol. 16 No. 3, pp. 310-321. [ Links ]

McNamara, K. (2009), “Publicizing private lives: celebrities, image control and the reconfiguration of public space”, Social and Cultural Geography, Vol. 10 No. 1, pp. 1-23. [ Links ]

Misra, S. and Beatty, S.E. (1990), “Celebrity spokesperson and brand congruence, an assessment of recall and affect”, Journal of Business Research, Vol. 21 No. 2, pp. 159-173. [ Links ]

Mowen, J.C. and Brown, S.W. (1981), “On explaining and predicting the effectiveness of celebrity endorsers”, Advances in Consumer Research, Vol. 8 No. 1, pp. 437-441. [ Links ]

Myrick, J.G. and Evans, S.D. (2014), “Do PSAs take a bite out of shark week? The effects of juxtaposing environmental messages with violent images of shark attacks”, Science Communication, Vol. 36 No. 5, pp. 544-569. [ Links ]

Ohanian, R. (1991), “The impact of celebrity spokespersons’ perceived image on consumers’ intention to purchase”, Journal of Advertising Research, Vol. 31 No. 1, pp. 46-54. [ Links ]

Pringle, H. and Binet, L. (2005), “How marketers can use celebrities to sell more effectively”, Journal of Consumer Behavior, Vol. 4 No. 3, pp. 201-214. [ Links ]

Subhadip, R. (2012), “To use the obvious choice: investigating the relative effectiveness of an overexposed celebrity”, Journal of Research for Consumers, Vol. 22, pp. 41-69. [ Links ]

Rice, D.H., Kelting, K. and Lutz, R.J. (2012), “Multiple endorsers and multiple endorsements: the influence of message repetition, source congruence and involvement on brand attitudes”, Journal of Consumer Psychology, Vol. 22 No. 2, pp. 249-259. [ Links ]

Rockwell, D. and Giles, D.C. (2009), “Being a celebrity: a phenomenology of fame”, Journal of Phenomenological Psychology, Vol. 40 No. 2, pp. 178-210. [ Links ]

Rumschisky, A. (2009), “The value of using famous personalities in advertising communications: a quantitative analysis of prices for a fashionable product”, available at: http://knowledge.wharton.upenn.edu/article/celebrity-advertising-what-is-the-roi/ (accessed 10 July 2014). [ Links ]

Rust, R., Lemon, K. and Zeithalml, V. (2004), “Return on marketing: using customer equity to focus marketing strategy”, Journal of Marketing, Vol. 68 No. 1, pp. 109-127. [ Links ]

Silva, L.A., Lopes, E.L., Freire, O. and Silva, D. (2015), “The brand’s effect on the evaluation of advertising endorsed by celebrities: an experimental study”, Brazilian Business Review, Vol. 12 No. 4, pp. 57-78. [ Links ]

Sliburyte, L. (2009), “How can celebrities be used in advertising to the best advantage?”, World Academy of Science, Engineering and Technology, Vol. 58 No. 2, pp. 934-939. [ Links ]

Thomas, V.L. and Fowler, K. (2015), “Examining the impact of brand transgressions on consumers’ perceptions of celebrity endorsers”, Journal of Advertising, Vol. 45 No. 4, pp. 377-390. [ Links ]

Thomson, M. (2006), “Human brands: investigating antecedents to consumer’s strong attachments to celebrities”, Journal of Marketing, Vol. 70 No. 3, pp. 104-119. [ Links ]

Till, B. and Shimp, T. (1998), “Endorsers in advertising: the case of negative celebrity information”, Journal of Advertising, Vol. 27 No. 1, pp. 67-82. [ Links ]

Tripp, C., Jensen, T. and Carlson, L. (1994), “The effects of multiple product endorsements by celebrities on consumer attitudes and intentions”, Journal of Consumer Research, Vol. 20 No. 4, pp. 533-547. [ Links ]

Um, N.-H. (2013), “Celebrity scandal fallout: how attribution style can protect the sponsor”, Psychology & Marketing, Vol. 30 No. 6, pp. 529-541. [ Links ]

Van Steenburg, E. (2015), “Areas of research in political advertising: a review and research agenda”, International Journal of Advertising, Vol. 34 No. 2, pp. 195-231. [ Links ]

Wei, P.S. and Lu, H.P. (2013), “An examination of the celebrity endorsements and online customer reviews influence female consumers’ shopping behavior”, Computers in Human Behavior, Vol. 29 No. 1, pp. 193-201. [ Links ]

Wheeler, R.T. (2009), “Nonprofit advertising: impact of celebrity connection, involvement and gender on source credibility and intention to volunteer time or donate money”, Journal of Nonprofit & Public Sector Marketing, Vol. 21 No. 1, pp. 80-107. [ Links ]

Further reading

Ghodeswar, B.M. (2008), “Building brand identity in competitive markets”, Journal of Product & Brand Management, Vol. 17 No. 1, pp. 4-12. [ Links ]

Spry, A., Pappu, R. and Cornwell, T.B. (2011), “Celebrity endorsement, brand credibility and brand equity”, European Journal of Marketing, Vol. 45 No. 6, pp. 882-909. [ Links ]

Received: September 22, 2016; Accepted: July 18, 2017

*Corresponding author Otávio Freire can be contacted at: otfreire@usp.br

Creative Commons License Published in RAUSP Management Journal. Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at https://creativecommons.org/licences/by/4.0/legalcode