Examining the Effectiveness of Athlete Celebrity Endorser Characteristics and Product Brand Type: The Endorser Sexpertise Continuum

The Athlete as Celebrity Endorser Defined

Certain athletes are utilized by brands as endorsers of their products because of the celebrity status gained by these athletes as a result of their success in their chosen sports. According to Boorstin (1961), “[t]he celebrity is a person who is known for his well-knownness (p. 57, italics added).” In fact, Boorstin (1961) refers to celebrities as human pseudo-events, claiming that the omnipresent print and broadcast media have provided a “...means of fabricating well-knownness (p. 47).” As such, celebrities rise and fall through the processes and whims of publicity.

McCracken (1989) defines the celebrity endorser “as any individual who enjoys public recognition and who uses this recognition on behalf of a consumer good by appearing with it in an advertisement” (p. 310). According to McCracken, celebrities are believed to increase sales by lending their favorable image to a product through endorsement. Both marketers and celebrities have much at stake in these endeavors and therefore should be keenly interested in factors that contribute to successful endorsements. According to the literature, endorser effectiveness can be determined based on source credibility, source attractiveness, and the attribution of endorser characteristics to the product.

Models of Celebrity Endorser Effectiveness

Most models of celebrity endorser effectiveness proposed by previous researchers have been based upon two basic models: the Source Credibility Model (Hovland, Janis, & Kelley, 1953; Hovland & Weiss, 1951) and the Source Attractiveness Model (McGuire, 1968). The Source Credibility Model purports that the effectiveness of the message is largely determined by the expertise and trustworthiness of the celebrity endorser, which offers explanations for why certain athlete endorsers such as highly successful men’s tennis professionals Andre Agassi, Rafael Nadal, and Roger Federer could easily represent any brand of tennis equipment. In a different vein, the Source Attractiveness Model purports that similarity, familiarity, and liking of celebrity endorsers primarily determine their effectiveness, which attempts to explain why some athletes such as Manning, Woods, and racecar driver Dale Earnhardt Jr. can endorse brands in many different product categories.

A general measure of physical attractiveness has often been used to assess celebrity endorser effectiveness, assuming what is beautiful is good (Dion, Berscheid, & Walster, 1972). However, mixed findings suggest that caution should be exercised in basing endorsement decisions on this model alone, since consumers perceive physical attractiveness differently.

The Product Matchup Hypothesis (Forkan, 1980; Kahle & Homer, 1985; Kamins, 1989; Kamins, 1990) emphasizes the need for a matchup, also referred to as congruency or fit, between the celebrity and the product. For example, the fit between Woods and Nike Golf is undeniable. However, according to sport marketing consultant Marc Cagins, Woods has failed as an endorser for Buick because of a lack of fit: “People always shook their head as to why he did the deal anyway with Buick. If he’s going to do a deal with GM, you would have thought Cadillac more than Buick” (Thomaselli, 2008, p. 28).

Research based on attractiveness matchup and expertise matchup has supported the hypothesis that celebrity/product fit remains critical to endorsement success. Kamins (1990) tested the matchup hypothesis based on attractiveness and found that the physical attractiveness of a source only mattered when an attractive source was paired with an attractiveness-related product, resulting in increased perceived source credibility and attitude toward the advertisement. Current examples would include the female tennis professionals and sisters Serena and Venus Williams’ Avon cosmetics campaign and the European men’s’ soccer star Thierry Henry/Woods/Federer Gillette campaign. In a study examining the impact of gender and physical attractiveness on responses to motorsports sponsorships (Roy, Stewart, & Goss, 2003), attractiveness played a greater role for female drivers than for male drivers, with the more attractive female drivers perceived as possessing greater expertise with products such as shampoo, tanning products, and dietary products. However, Bower and Landreth (2001) found that highly attractive models were not more effective than normally attractive models for problem-solving, attractiveness-related products and attributed these results to the premise that consumers perceive highly attractive models to have never faced an attractiveness-related problem and therefore have little expertise using these products. To illustrate, this premise would suggest that the current campaign utilizing female tennis star Serena Williams for ProActiv acne medication would be unsuccessful. However, a moderately attractive model may be more likely to be perceived to have faced an attractiveness-related problem and overcome it using the advertised product, fully illustrating the premise that the expertise of the source is more important than the attractiveness of the source.

In Till and Busler’s (2000) examination of the attractiveness and expertise matchup, while an attractiveness/matchup effect was not found, an attractiveness effect and an expertise/matchup effect were both discovered, suggesting the possibility of expertise trumping physical attractiveness in matching celebrities and products for endorsement effectiveness. These findings also suggest the possibility that physical attractiveness is itself considered a type of expertise. To illustrate, a beautiful female model may be considered an expert on makeup, which is a product she could use to enhance her beauty. However, she may not be perceived as an expert on cars, because while she looks beautiful in the car, the car is not connected to her physical beauty.

McCracken (1989) suggests that a matchup of attractiveness or expertise alone is insufficient for achieving successful endorsements, positing a Model of Meaning Transfer. In this model, he proposes that celebrity endorsers are special cases of meaning transfer: “The effectiveness of the endorser depends, in part, upon the meanings he or she brings to the endorsement process” (p. 312). To elaborate, he explains that numerous cultural meanings exist in the world, and that celebrities draw meanings from the culturally constituted world when developing their images, which are accumulations of meanings from the roles they assume in various aspects of their careers, both in media and in the public eye. McCracken (1989) suggests that, similar to typecasting of actors in Hollywood, the consistency of the images that celebrities develop is the notion that makes them useful to marketers. These meanings are then transferred to the products that celebrities endorse. Consumers then recognize that the celebrities possess compelling or useful cultural meanings, causing them to buy the endorsed products and acquire these meanings upon consumption of the products. Therefore, for an endorsement to become successful, products must match well with the bundle of meanings or overall image of celebrity endorsers rather than with the element of attractiveness or expertise alone.

The Multidimensional Endorser

Since the overall image of the celebrity must be taken into consideration in endorsement decisions, McCracken’s (1989) theory suggests that celebrity endorsers possess a multidimensional nature. Erdogan, Baker, and Tagg (2001) confirm “managers do not see celebrities as unidimensional individuals (e.g., attractive, credible)” (p. 45). Rather, their findings indicate that advertising agency managers should consider a range of criteria when choosing celebrity endorsers, including trustworthiness, expertise, physical attractiveness, familiarity, and likeability of the source. Further, the importance of the criteria depends on the product brand to be endorsed.

The Roles of Expertise and Likeability

Celebrities “...succeed by skillfully distinguishing themselves from others essentially like them (Boorstin,1961, p. 65).” However, based on the theory that celebrity endorsers possess a multidimensional nature, the Endorser Sexpertise Continuum proposes a distinct difference in the products with which a celebrity endorser can be appropriately matched, a conclusion based upon whether the celebrity endorser’s fame stems from his/her area of expertise, or whether the source of the celebrity endorser’s fame stems from his/her physical attractiveness or personality. For example, women’s tennis star Anna Kournikova’s popularity stems almost entirely from her sex appeal, whereas Woods is known for his athletic prowess on the golf course. Each of these athletes anchors a different end of the continuum. Some athletes such as female swimmers Amanda Beard and Dara Torres fall in the middle, possessing both sex appeal and athletic success.

Celebrity endorsers provide the tools that consumers seek to affect meaning transfer by availing the cultural meanings encapsulated in the material forms of the celebrities’ images through product endorsement. According to McCracken (1989), “The celebrity is supplying not just an example of self-creation, but the very stuff with which this difficult act is undertaken” (p. 317).

Some celebrities possess an acquirable expertise, i.e., other people can learn the same expertise possessed by certain celebrities. As previously discussed concerning cultural meanings, celebrity endorsers provide consumers with the tools or skills needed to acquire that same expertise. For example, Denise Austin and fitness guru/martial artist Billy Blanks can provide other people with the tools needed for meaning transfer, so their celebrity images focuses more on the characteristic of acquirable expertise through learning their fitness regimens. These celebrities tend to gravitate toward the left side of the continuum, but some may move to the right end of the continuum, depending on their degree of likeability. For example, in addition to wide-ranging fitness options and abilities, Austin is also generally known for her outgoing personality, beauty, and calm, encouraging approach to fitness, so she would fall further to the right side of the continuum than Blanks, who is primarily known for his intense TaeBo fitness regime and is presented as a less dynamic, less personable figure.

In contrast, some celebrities possess an inherent expertise unique to that person alone and therefore cannot be achieved by another person. For example, Manning’s football abilities cannot be imitated, even if someone were to take lessons directly from him. People are attracted to such celebrities. In essence, these celebrities cannot provide others with specific, tangible tools they need for meaning transfer. Accordingly, their celebrity images focus more on the characteristic of likeability, causing these celebrities to occupy the right side of the proposed continuum. Furthermore qualifying celebrities for the right side of the continuum would be the popularity and endurance of the celebrity image. Celebrities may also move to the right as their images become more solidified. For example, while Jordan was a highly successful basketball player, he transcended that label to be a likeable celebrity far beyond the basketball court.

Logically, if the image of a celebrity is focused on an acquirable expertise, the range of products that he/she can endorse is limited to that area of expertise. However, if the image of a celebrity is focused on likeability, then he/she can endorse a much broader range of products, since the image transfer is based on personality feature rather than a particular area of expertise.

Consumer Needs

In developing their Brand Concept Management (BCM) model, Park, Jaworski and MacInnis (1986) identified three types of brand concepts (or images created in a brand) based on consumer needs: functional, symbolic, and experiential. Functional needs are defined as “those that motivate the search for products that solve consumption-related problems” (Park et al., 1986, p. 136). Symbolic needs are defined as “desires for products that fulfill internally generated needs for self-enhancement, role position, group membership, or ego-identification” (Park et al., 1986, p. 136). Experiential needs are defined as “desires for products that provide sensory pleasure, variety and/or cognitive stimulation” (Park et al., 1986, p. 136). As with BCM, the characteristics of each of these brand concepts must be taken into consideration in developing an effective endorsement. Till and Busler (2000) suggested that involvement could be a factor to consider in matchup. Since these brand concepts are based on consumer needs, a closer look at consumer involvement and ways they process information may lead to a better understanding of this phenomenon.

Consumer Involvement and Information Processing

Petty, Cacioppo, and Schumann (1983) investigated the moderating role of consumer involvement in advertising effectiveness. Based on the Elaboration Likelihood Model (ELM), they identified two distinct routes to attitude change based on likelihood of the consumer elaborating the message: the central route and peripheral route to persuasion. When defining attitude change as a representation of whether a consumer has favorably changed his/her attitude toward the brand after encountering some of its communication (e.g., an advertisement), higher involvement (the central route) results in more scrutiny of the message, whereas lower involvement (the peripheral route) results in consumers giving little or no thought about the product qualities presented in the message. In the central route, attitude changes result from consumers carefully analyzing information for a particular situation. In contrast, attitude changes in the peripheral route result from inferences made from available cues. The findings of Petty et al. (1983) indicate that the persuasive message itself has a greater impact under high involvement conditions, whereas the endorser has a greater impact under low involvement conditions. Therefore, the central route is a more cognitive dimension, while the peripheral route is a more affective dimension. This premise most likely explains why Woods failed as an endorser for Buick. Since cars are high involvement products that lead consumers to carefully analyze information, the likelihood that a celebrity endorser will work in this situation declines greatly. Woods’ failed Buick endorsement can also be coupled to a mismatch between the celebrity and brand (Thomaselli, 2008). A hypothetical example of the central versus peripheral route could be found the use of racecar driver Dale Earnhardt Jr. as an endorser. Earnhardt works well for Mountain Dew’s Amp energy drink (a low involvement product) that serves as a co-primary sponsor of his racecar, since he races with a youthful, energetic style. Theoretically, however, in the same role for the National Guard (a high involvement product), his sponsorships may prove less effective, since no apparent linkage exists between Earnhardt’s racing career and military service.

While Petty et al. (1983) indicated that the expertise or attractiveness of a message source is a peripheral cue, their research purports that, when product involvement is high, an expert endorser will serve as an argument for consideration. However, when product involvement is low, the endorser then serves a more affective, holistic function and is indeed a peripheral cue. In their hypothesis test of attractiveness as a peripheral cue for shampoo, Petty and Cacioppo (1980) explained that they found no attractiveness and involvement interaction and commented that some subjects’ lack of interaction might have been attributed to model’s appearance as a relevant, persuasive argument.

Sujan (1985) identified two types of consumer processing strategies that complement the central and peripheral routes. In the piecemeal approach, consumers analyze each attribute of a product to determine its overall value. Alternatively, in the categorical approach, Sujan (1985) notes that consumers perform more efficient processing by utilizing previously defined categories: “The belief is that affect is cued by the categorization of stimuli rather than through a constructive attribute review process” (p. 31).

Based on this discussion, since functional brand concepts are related to problem solving, they would require more piecemeal processing and therefore would belong with the cognitive route of the ELM. On the other hand, symbolic brand concepts are related to reference groups and ego enhancement. They require more categorical processing and therefore would belong with the peripheral route of the ELM. Experiential brand concepts are more internal driven by consumers’ need for cognitive or sensory stimulation and therefore are more context-driven. As such, they may belong in either route, depending upon their context.

Endorser as a Brand

Kotler (1991) defines a brand as “a name, term, sign, symbol, or design, or combination of them which is intended to identify the goods and services of one seller or group of sellers and to differentiate them from those of competitors” (p. 442). In essence, a brand is represented as an image that has been developed and that exists in the minds of consumers. Therefore, since a celebrity endorser is essentially a bundle of meanings that combine to form an overall image, a celebrity endorser can also be considered a brand.

To further discuss celebrities as their own brands, marketers often attempt to personalize their brands by endowing them with certain human characteristics or brand personalities. Aaker (1997) defines brand personality as “the set of human characteristics associated with a brand” (p. 347). For example, the Marlboro cigarette brand exhibits a high level of brand personality. According to Lohof (1969), “The Marlboro image is a cultural symbol which speaks to the collective imagination of the American people. It speaks of virgin frontier and of brutal efficacy and constant vigilance which the frontier exacts from its residents” (p. 447; cf. Reddy, Holak, & Bhat, 1994). However, in the case of a celebrity endorser, the brand is an actual person. Rather than developing a brand personality for a product, a celebrity endows his/her clearly defined, existing personality to the product through endorsement. Therefore, a celebrity endorser can be considered the epitome of a brand personality. O’Donnell (2008) claims that the pairing of Michael Jordan and Nike set the standard for athlete/product endorsement matchup, and one can argue that Nike’s success is rooted in that partnership decision. Nevertheless, all brands have lifecycles (Ries & Ries, 1998). Nike has experiences problems selling the Jordan brand since Jordan’s retirement in 2003 because many consumers in younger demographics have only seen Jordan play on highlight reels and do not perceive him as relevant (Intini, 2008). Thus, the brand’s new focus will be on the myth of Michael with “Become Legendary” as the new tagline (Intini, 2008). Intini (2008) believes that if Nike handles this new strategy well, Jordan could become sport’s first immortal brand.

Endorsement as a Brand Alliance

If an endorser can be considered a brand, then the pairing of a celebrity endorser and a product brand is analogous to the concept of a brand alliance. According to Rao & Ruekert (1994), because brand names are valuable assets, they can combine with other brand names in a synergistic alliance in which the sum is greater than the parts. In the case of a celebrity endorsement, the product brand associates with the endorser in an effort to develop a stronger positive consumer attitude toward the product brand, essentially hoping to capitalize on positive spillover effects. The work of Simonin and Ruth (1998) supports this strategy. Based on familiarity, Simonin and Ruth (1998) found that spillover effects might not always affect the partners equally. If two highly familiar brands ally, they experience equal spillover effects. However, in an asymmetrical relationship the less familiar brand experiences greater spillover effects. In essence, celebrities should be compensated based on their contribution towards such a spillover effect.

However, the threat of negative spillover effects necessitates consideration by both endorsers and product manufacturers, when engaging in partner selections. An action by an athlete is considered a scandal if it is illegal or unethical, involves multiple parties over a sustained period of time, and/or whose impact affects the integrity of the sport with which he or she is associated (Hughes & Shank, 2005). Hughes and Shank (2005) reason that a one-time violation by an admired sports celebrity will carry less impact with the consumer than repeated illegal or unethical events. In contrast, Behr and Beeler-Norrholm (2006) posit that the notion of celebrities-gone-bad is part of the fun for consumers and question whether it truly does serious damage to the consumers’ views of the endorsed brand. This may be true for the celebrities who are liked specifically for their rebellious personalities but is not generally applicable for all endorsers. When measuring consumer skepticism, Bailey (2007) found the strength of the association between the endorser and the brand affects whether the negative information will affect consumer perceptions and attitudes.

Positive and negative spillover effects impact brand alliances. Since the image of the brand is composed of an accumulation of meanings, each time a brand associates with another brand, the experience impacts and contributes to its overall image (Rodrigue & Biswas 2004). To maintain its marketing value, the product brand and the endorser brand must each strive to maintain a consistent individual image in the public eye.

Fit

The concern of fit in a brand alliance can also be applied in the case of an athlete celebrity endorsement. Simonin and Ruth (1998) found that product fit and brand fit are related positively to attitudes toward the brand alliance. In the endorsement literature, expertise had a stronger effect on attitude toward the brand and purchase intentions than physical attractiveness (Ohanian 1991; O’Mahoney & Meenaghan 1998; Till & Busler 1998). When the fit between the endorser and the endorsed product is incongruent, only attractiveness remains important (Kim & Na 2007). Further, the work of Broniarczyk and Alba (1994) supports the idea that a specific brand attribute, such as the expertise of the celebrity, will have a greater influence on the effectiveness of the endorsement than the overall image of the brand. Experience with a product does not always equate to expertise with the product (Siemens, Smith, Fisher, & Jensen 2008). Sport celebrity endorsers especially must carefully choose their endorsements, since they could be held liable if they do not use the product they endorse (Moorman 2006). If consumers are more engrossed in a product, then they will be paying closer attention to the significant attributes of the brand rather than the holistic affect related to the brand. However, less involved consumers will not be as interested or knowledgeable about relevant product issues and may look more for a holistic affect.

As shown in Figure 1, the functional brand concept has been paired with the celebrity endorser known for an acquirable expertise. If the consumer is in the central route and is processing information in a piecemeal fashion, then the expertise of the endorser may be an important consideration. The symbolic brand concept has been paired with the celebrity endorser known for likeability. While these celebrities could endorse a product related to their original area of expertise (e.g., Woods and Nike Golf), they have transcended this level and are also able to endorse unrelated products based on affect (e.g., Woods and Gillette). If the consumer is in the peripheral route and is processing information categorically, then the overall image of the celebrity will play an important role in the purchase decision. However, these celebrities would not be successful endorsers for functional products in which they have no expertise (e.g., Woods and Buick automobiles).