Customers' need for uniqueness theory versus brand congruence theory: The impact on satisfaction with social network sites

Social media brands may experience a lack of competitiveness and attraction due to the silent negative increase of customers' need for uniqueness (NFU). This is the result of a tension between the theory of brand congruence, that most brands endeavor to establish with their customers, and the theory of consumers' NFU, that many consumers aim to fulfill, that has not been previously examined. Whilst the theory of brand congruence states that consumers have a favorable attitude towards brands that greatly match their self-concept (brand similarity), the theory of uniqueness, when brand similarity is shared with others, generates avoidance of similarities with others, leading to a reduction in engagement and satisfaction. Social media platforms including Facebook are at risk from customers' NFU which significantly reduces users' satisfaction, hence reducing the attractiveness and usage of Facebook. To examine this, the paper identifies the paradox of similarity in relation to both theories and examines the relationship between consumers' socialization within Social Network Sites (SNSs), brand similarity, consumers' need for uniqueness, and consumers' satisfaction with the SNS. The conceptual model was tested using data from 341 consumers in the US. The findings show that whilst satisfaction with the SNS is enhanced by brand similarity and friend liking, it is undermined by customers' need for uniqueness. Implications for online brand communities and SNSs (Facebook) are discussed and guidance for future research is provided.


Introduction
Brands' continuous efforts to engage with their consumers through Social Network Sites (SNSs) have proven to have a significant and positive effect on the success of brand engagement (e.g. Hanna, Rohm, & Crittenden, 2011;Morgan-Thomas & Veloutsou, 2013). Most studies on consumer engagement within SNSs have focused primarily on online communities within an SNS such as Facebook, Instagram, and Twitter, and rarely have these studies looked at the brands of SNSs themselves. Few studies have examined the role of the SNS on consumer engagement with brands (e.g. Algesheimer, Dholakia, & Hermann, 2005;Shih, 2009;Thorbjørnsen, Supphellen, Nysveen, & Pedersen, 2002;Wetsch, 2012). Much of the literature on online consumer psychology has focused on understanding consumer online brand relationships (e.g. Hayes, Luoma, Bond, Masuda, & Lillis, 2006;Morgan-Thomas & Veloutsou, 2013), online shopping attitudes (e.g. Hausman & Siekpe, 2009;Overby & Lee, 2006), and e-word of mouth (e.g. Chen, Tang, Wu, & Jheng, 2014;Okazaki & Taylor, 2013). However, the impact of consumer psychology and SNSs' brand similarity on customers' satisfaction with SNSs (e.g. Facebook) lacks examination. Studies that have discussed customers' satisfaction with SNSs (e.g. McAlexander, Kim, & Roberts, 2003;Rose, Clack, Samouel, & Hair, 2012) have not considered the impact of consumers' social experiences on SNSs as well as brand on consumer self-concept.
Most users of social media platforms seek to fulfill their need for uniqueness (NFU) through membership with recognized and leading social media brands including Facebook and Instagram. Users' social experiences on SNSs (e.g. Kim and Ko, 2010;Wetsch, 2012) and their similarity with brands (e.g. Kabadayi & Price, 2014;Rowley, 2004) are key determinants of satisfaction with those SNSs. Hence, a consumer's self-concept within an SNS and how an individual customer sees him/ herself in comparison to others within the online brand community can impact their satisfaction with the SNS. According to Sirgy (1982) there are four elements of self-concept that explain consumer behavior. These include the ideal self-image, the actual self-image, the social self-image, and the ideal social self-image. In this study, we focus on the ideal selfhttps://doi.org/10.1016/j.jbusres.2019.03.016 Received 30 December 2017; Received in revised form 6 March 2019; Accepted 10 March 2019 image and the ideal social self-image. The ideal self-image was consistently found to explain consumers' attitudes and behaviors towards brands (Aaker, 1999;Aaker & Schmitt, 2001), whilst the ideal social self-image influences consumers' socialization and friend liking within SNSs (Muniz & O'Guinn, 2001;White & Dahl, 2007). Ideal self-image is related to the "fit between how consumers would like to see themselves in relation to brand personality" (Kuenzel & Halliday, 2010, p. 170). Ideal social self-image is related to comparing one's own-self with referent others, with whom one shares a social identity (Festinger, 1954).
The theory of brand congruence and the theory of uniqueness are employed in this study. Each theory provides good insights into the competing aspects that enhance or damage consumers' satisfaction with SNSs. Such conflicting effects that consumers are exposed to when on SNSs (online brand community on Facebook) need to be better understood. Whilst the theory of brand congruence argues that consumers who are congruent with a brand are likely to feel better about themselves (better ideal self-image) (see e.g. Alpert & Kamins, 1995;Dolich, 1969;Grubb & Hupp, 1968;Sirgy, 1982), the theory of the customer's NFU argues that once a good deal of similarity with others is apparent and feelings of indifference are aroused (the self-perception of uniqueness is undermined), satisfaction reduces due to the absence of any effect on both ideal-self and social images (Snyder & Fromkin, 1977a). Therefore, customers' NFU can have a significant negative impact on consumer satisfaction with the SNS -Facebook. Customers' NFU is defined as "a positive striving for differentness relative to other people" (Snyder & Fromkin, 1977a, p. 519). Customers' NFU has the potential to affect individuals' identification within online brand community.
To demonstrate this, this study develops a conceptual model that integrates the effect of both theories and tests their influence on satisfaction with SNSs (Facebook). The conceptual model includes constructs that reflect consumers' socialization (friend liking) within Facebook, brand similarity, customers' need for uniqueness, and satisfaction with Facebook. The findings provide good support to the tension that the theory of brand congruence and the customers' need for uniqueness put on consumer behavior within Facebook. Importantly, the findings show that whilst brand similarity increases customers' perceptions of uniqueness (as predicted by the theory of brand congruence), consumers' need for uniqueness reduces satisfaction with Facebook. The paper starts by discussing the theoretical foundation of the study. It then discusses and justifies the hypotheses within the conceptual model, followed by detailed discussion on the methods used and analysis performed. Discussion, implications, and future research are then presented.

Theoretical background
In developing the conceptual model, the study draws on recent developments in literature, including studies on online brand relationships (e.g. Morgan-Thomas & Veloutsou, 2013), online brand communities (e.g. Algesheimer et al., 2005;Chan & Li, 2010), customers' experiences in online communities (e.g. Novak, Hoffman, & Yung, 2000;Rose et al., 2012), and social identification (e.g. Bhattacharya & Sen, 2003). However, the theoretical foundation of the conceptual model was based on two key overarching theories, namely the theory of uniqueness (e.g. Snyder & Fromkin, 1977a) and the theory of brand congruency (e.g. Sirgy, 1982). Alongside these two theories, the theory of social comparison (Festinger, 1954) also contributes well to the identification and understanding that the paradox of similarity can have significant implications on consumers' satisfaction with SNSs (Facebook).

The theory of uniqueness
Individuals are motivated to see one own self as different and distinctive from others for a meaningful self-identification (Vignoles et al., 2000). The work by Snyder and Fromkin (1977b) empirically establish why people need to see themselves as different and unique compares to their social group. People with high NFU tend to feel unpleasant about perception of similarity with other and seek to deviates from the group aiming to differentiate themselves from others (Snyder & Fromkin, 1977a). Hence, people naturally are motivated to maintain a level of uniqueness in order to enjoy improving their self-image that they want others to see (Tianetal., 2001).
The theory of uniqueness (Snyder, 1992;Snyder & Fromkin, 1977a) argues that when the need to feel different from other people is aroused and, in events where the self-perception of uniqueness is undermined, the need to feel different competes with other motives to protect and enhance such uniqueness. According to Tian, Bearden, and Hunter (2001, p. 50), Need for Uniqueness (NFU) is defined as "an individual's pursuit of differentness relative to others that is achieved through the acquisition, utilization, and disposition of consumer goods for the purpose of developing and enhancing one's personal and social identity". The very essence of 'the need for uniqueness' is fundamentally based on counterconformity, as consumers with a high level of NFU tend to deviate from others in their group. Feeling differentiated from other people has long been recognized to have a significant impact on consumer behavior (e.g. Chan, Berger, & Boven, 2012;Simonson & Nowlis, 2000).
Such behavior generates avoidance of similarities with others, leading to a loss of interest in possessing products or brands that are commonly used by others. Such behavior results in moving away from the shared norms with others, attempting to re-establish self-uniqueness/differentness (Snyder, 1992). Avoiding similarity refers to "devaluing and avoiding the purchase of products or brands that are perceived to be commonplace" (Tian et al., 2001, p. 52). The enhancement of selfconcept and uniqueness occurs through acquiring symbolic meanings from purchased products or engaging with brands through internal and personal processes. The search for self-enhancement is intensive during low self-esteem or during high perception of similarity with others (Lynn & Snyder, 2002), and can lead to a negative emotion (Snyder & Fromkin, 1977b, Snyder & Fromkin, 1977a. Social comparison theory (Festinger, 1954) argues that an individual compares him/herself with referent others, with whom they share a social identity. Such comparison has a significant impact on customers' NFU as well as their consequent behavior towards brands.
However, the feeling of being left out from social groups may trigger a need for belonging, but still a certain desire for uniqueness may lead consumers to engage with brands that may not necessarily be anthropomorphized. People aim at achieving a certain level of uniqueness from and similarity to others (Lynn & Harris, 1997). People are naturally motivated to build and maintain relationships with others (Baumeister, 2012). Thus, such need for uniqueness may be constrained by the need for social assimilation (Brewer, 2012). People who feel very unique and different from others within their social group may increase their interaction within the group to maintain relevance and selfidentification with the group (Brewer and Pickett, 1999). However, people who perceive that they have highly similar to others tend to behave in ways that enable them to feel different and unique (Tian et al., 2001).

Brand congruence theory and the paradox of similarity
The theory of brand congruence argues that individuals are more comfortable and satisfied with products/brands that are congruent with their actual or desired self-concept (Grubb & Grathwohl, 1967;Sirgy & Su, 2000). While the theory of brand congruence recognizes the actualself and the ideal/desired-self, there are two relevant motives that impact brand congruence: self-consistency and self-esteem (Alpert & Kamins, 1995;Sirgy, 1982). Self-consistency motivates individuals to avoid behaviors that are incompatible with the self-concept, whilst selfesteem motivates individuals to achieve greater uniqueness of selfimage. There is clearly competition or tension between the two aspects of the self (self-consistency versus self-esteem) that influences individuals' buying decisions (e.g. Sirgy, 1982;Snyder & Gangestad, 1986). However, individuals experiencing NFU are more likely to focus on achieving the ideal image and uniqueness by avoiding products/ brands that, while congruent with their self-concept, are widely shared by many other individuals.
Hence, brand congruence/similarity is, on the one hand, key in providing the desired feeling of uniqueness searched for by a customer, as engaging with and possessing the brand reflects the 'unique self'. But, on the other hand, brand similarity when shared with others will not fulfill customers' NFU as no differentiation or specialness is achieved. Instead it can lead customers to avoid engaging with the brand as it will not contribute to the feeling of uniqueness. Therefore, whilst brand similarity can contribute positively to customers' NFU, when it is shared with others from within the same social group, customers' NFU has negative consequences.
Unless a brand acts as a supporter to consumers' need to be different and unique, it may end up being a product like many others (Puzakova & Aggarwal, 2018). There is a difference to how different social groups perceive things; such as consumers who purchase for hedonic or utilitarian reasons. Consumers feel that hedonic purchases are done with unique preferences compared to utilitarian, simply because consumers would each like a product to suit their own pleasures compared to products that serve a general purpose (Whitley, Trudel, & Kurt, 2018). Consumers with incidental pride are more likely to seek uniqueness in options that would show off the pride in their achievements or personal traits (Huang, Dong, & Mukhopadhyay, 2014). Moreover, the need for uniqueness is directly related to the level of knowledge and involvement that consumers put into a product. The more they relate and feel positive towards a certain product, the more they are driven to purchase it to help further their uniqueness (Bhaduri & Stanforth, 2016).

The context of social network sites
Different SNSs target different people; some cater for a diverse audience while others attract people based on commonalties (e.g. language, gender, nationalities, etc.) amongst users (Ellison, 2007). In addition, the degree to which an SNS incorporates new information and communication tools (e.g. mobile connectivity, photo/video sharing) can vary from one site to another (Ellison, 2007). The persistence of an SNS heavily depends on the continued use of its members (Chang & Zhu, 2012).
While SNSs have attracted users from around the globe, retaining these members is not easy and requires continuous innovation (Chiang, 2013). Indeed, for an SNS to flourish it needs to maintain a competitive advantage by satisfying the 'gratification' purpose(s) of its users that motivate continuous use (Chang & Zhu, 2012;Wu, Wang, & Tsai, 2010). Once one or more of these purposes are fulfilled, the satisfaction of members within the SNS can then be ensured.
For any user to become a member of an SNS, a profile needs to be created. This profile acts as a self-representation of the personal information that members share with friends or other SNS members (Cheung, Chiu, & Lee, 2011;Lin & Lu, 2011;Tapscott, 2008). The profile of each member is unique and is considered as an expression of one's identity, in which an individual can "type oneself into being" (Sundén, 2003, p. 3). Establishing a profile helps members identify the level of like-mindedness between individuals, which determines the degree of likeability between friends (Matzler, Pichler, Füller, & Mooradian, 2011;McAlexander, Schouten, & Koenig, 2002). These relationships can be formed based on interactions between individuals and pre-existing friends or other SNS members (Choi, Kim, Sung, & Sohn, 2011;Thelwall, 2008) that share a state of "consciousness of kind" (Giddings, 1896). Attitude similarity is a key element in forming impressions (Byrne, 1971) and creating friendships (Kandel, 1978) on an SNS. In fact, it is the foundation for a lasting friendship (Werner & Parmelee, 1979). Once members become familiar with one another, a sense of belonging is formed, where individuals strive to be part of a community that represents their interests and builds upon their social capital (Lin, Cook, & Burt, 2001;McKenna, Green, & Gleason, 2002;Peter, Valkenburg, & Schouten, 2005). This means that becoming a member of an SNS offers individuals an opportunity to capitalize on their social networks with friends they like (Boyd, 2004). In fact, the maintenance of social capital is one of the main reasons behind why people continue using SNSs (Hu & Kettinger, 2008). However, to maximize this social capital, individuals should strive to become members of an SNS that includes not only their social circle but also individuals they like (Lin et al., 2001). This demonstrates that there would be an attitude formation, not only between the members of the SNS, but also between these members and the SNS itself. Nonetheless, both relationships are interrelated, as interpersonal relationships play a sizeable role in forming such attitudes (Eaton, Majka, & Visser, 2008). These attitudes can be either positive or negative, depending on users' experiences online (Eaton et al., 2008), as well as the degree of interpersonal attraction between the individuals of the SNS (Byrne, 1969(Byrne, , 1971. Facebook has been deemed as the leading SNS since 2016 to date with a consistency of usage throughout the different age groups (Murnane, 2018). Facebook is being used as a social network that allows people to communicate and stay in touch with connections from all over the world. More specifically, Facebook has proven to have the closest connections -whether with people that they know from high school and daily interactions, or people they have never met in person (Hampton, Goulet, & Purcell, 2014). With almost 2 billion monthly active users, Facebook has become the most important platform for marketers, using photographs and videos in order to garner the interest of the various age groups, genders, and social groups available on the website; the only challenge here is to find the perfect strategy to reach all intended audiences (Standberry, 2018). With the growth of social advertising featuring endorsed brands throughout the years, people are more willing to interact with brands more positively once they find their friends taking action as well. This shows that the importance of Facebook to users no longer stops at interacting and socializing, but rather scoping out the value of the brands that can be found on the site (Ramadan, Abosag, & Zabkar, 2018).

Friend liking and customers' NFU
It has long been known that consumers engage in activities on SNSs to enhance their self-presence, self-promotion, and uniqueness (e.g. Bibby, 2008;Kramer & Winter, 2008;Underwood, Kerlin, & Farrington-Flint, 2011), as well as their social identity expressiveness (Pagani, Goldsmith, & Hofacker, 2013). Consumers on SNSs regularly choose to interact to maintain close-knit social friendships (Skinstad, 2008). Consumers often share information, opinions, and feelings with others if it contributes to and enhances self-presentation and uniqueness (Kolek & Saunders, 2008;Stutzman, 2006). However, online social identification and friend liking is defined as a high degree of similarity amongst individuals of the social group within the SNSs (Kabadayi & Price, 2014;Rowley, 2004). Perceived similarity amongst members of an online community brand promotes positive emotions and feelings (Biel & Bridgwater, 1990), and results in a higher level of liking within the SNSs (Beukeboom, Kerkhof, & de Vries, 2015;Kim, Lee, & Hiemstra, 2004). It is evident from the extant literature that similarity between individuals is a crucial determinant of interpersonal attraction and liking (Berscheid & Walster, 1969;Byrne, 1971;Huston, 1974;Lott & Lott, 1965). In fact, for a liking relationship to form between individuals of an SNS, interpersonal similarity and familiarity between the users need to exist (Kaptein, Nass, Parvinen, & Markopoulos, 2013). This similarity is defined by other members' personality traits, attitudes, values, behaviors, or physical appearances (Berscheid & Walster, 1969;Byrne, 1971;Huston, 1974;Lott & Lott, 1965). However, to be able to identify these attributes it is vital that one first joins an SNS that can represent the members' interests to the public.
Social interaction on SNSs that promote similarity and liking amongst their members is no different to social interaction offline (Fournier, 1998;Tufekci, 2008), in that it can trigger individuals' NFU (Chan et al., 2012;Miremadi, Fotoohi, Sadeh, Tabrizi, & Javidigholipourmashhad, 2011). Hence, a high perception of liking and similarities amongst friends on SNSs will inevitably lead to an increase of customers' NFU, as suggested by the theory of uniqueness (Tian et al., 2001). In addition, as predicted by the theory of social comparison, individuals who socialize and identify themselves with a social group tend to consciously and unconsciously compare themselves to their peers within the social group with whom they interact (Festinger, 1954). This is true on both SNSs and in offline interaction (Tufekci, 2008). Hence, a comparison of the self-concept within online brand communities and the high perception of friend likeability and similarities increases consumers' NFU which can have negative implications on SNSs. Recently, López, Sicilia and Moyeda-Carabaza (2017, p. 29) argued that "the extent that brand communities satisfy the need for affiliation (similarity), consumers may seek to enhance their distinctiveness from others". Therefore, we hypothesize the following: H1. : High friends' likeability negatively increases customers' need for uniqueness.

Friend liking and brand similarity
Relationships are mainly based on liking and similarity between users that share the same interests within a certain community (Matzler et al., 2011;McAlexander et al., 2002). In fact, for any relationship to thrive, a sense of liking between any two entities needs to exist (Hendrick & Hendrick, 1992). This sense of liking is a mixture of interpersonal similarity and familiarity between any two individuals (Kaptein et al., 2013). Indeed, we tend to like people that we frequently interact with and who are similar to us (Byrne, 1971;Strauss, Barrick, & Connerley, 2001), leading to closeness (Laurenceau, Barrett, & Pietromonaco, 1998;Ledbetter et al., 2011;Park, Jin, & Jin, 2011;Valkenburg & Peter, 2007) and trust (Sheldon, 2009).
The sharing of information on digital platforms develops an online impression (Good, 2013;Liu, 2007) that other users might feel similar with each other, that gives rise to a sense of liking (Cialdini, 2009), which in turn leads to the formation of relationships between SNS members (McKenna et al., 2002;Peter et al., 2005). Strong relationships and social ties with individuals on an SNS are established when a feeling of high similarity between the members of the SNS is found (McPherson & Smith-Lovin, 1987). Forming such a relationship will lead to more frequent interaction grounded by the similar interests, which are based on previous brand ownership experiences (Brown & Reingen, 1987;Matzler et al., 2011;McAlexander et al., 2002).
A brand can establish a relationship with its consumers if the consumers identify with the brand and develop a sense of brand similarity (Torres, Augusto, & Godinho, 2017). This similarity is a representation of both the brand and the consumer, which is mainly driven by likeable friends over social media platforms (Rowley, 2004). Friend likeability is a key motivator for brand similarity, given the sheer volume of brand endorsements and eWOM on social media platforms (Moran & Muzellec, 2017). On that basis, we hypothesize the following: H2. : The higher the SNS's friend likeability, the stronger the feeling of similarity with brands.

Friend liking and satisfaction with SNSs
There are four main attributes that define likeability between individuals: physical attractiveness, similarity, compliments, and association (Cialdini, 1993). Not all attitudes are equal; some attitudes are stronger and more durable than others (Petty & Krosnick, 1995;Levitan & Visser, 2009). To determine the level of interpersonal attraction, one should first assess the interpersonal reward gained from this attraction (Newcomb, 1956). In the SNS context, similar attitudes act as a reward for members because they satisfy the effectance motivation of members (Byrne & Clore Jr, 1967;Byrne, Clore, & Worchel, 1966). If members achieve this interpersonal similarity, an interpersonal attraction of liking will form between SNS friends, which satisfies their effectance motivation (Byrne et al., 1966;Byrne & Clore Jr, 1967). This means that the stronger the interpersonal attraction between individuals, the stronger the attitude towards the SNS. Only when rewarding members by satisfying their effectance motivation (Byrne et al., 1966;Byrne & Clore Jr, 1967), will satisfaction with the SNS be achieved.
There is no agreement on how customer satisfaction should be defined (Rogers et al., 1992). To give examples of this, He, Li, and Harris (2012) argued that customer satisfaction is the fit between the performance of a brand and the expectations of customers, whereas Delgado-Ballester and Munuera-Alemán (2001) and Schau, Muñiz, & Arnould, 2009 argued that customer satisfaction is related to brand trust. Further to this, customer satisfaction is linked to quality as Liljander and Strandvik (1992), p. 113) argued that these "have evolved along parallel tracks". Olsen and Johnson (2003) argued that customer satisfaction is a good predictor of customers' intentions and behaviour, whereas Kotler (1994, p. 20) argued that customer satisfaction is "the key to customer retention". In this study, we define satisfaction within SNSs as the contentment of users and their fulfilment response based on their cumulative experience within the SNS. The more consumers are satisfied with an SNS the higher their intentions will be to continue using it (Shi, Lee, Cheung, & Chen, 2010). This shows us that the satisfaction of members on an SNS is crucial for the survival of the social platform.
Users perceive enjoyment from belonging, and social capital on the SNS plays a big role in the continuance of the SNS (e.g. Hu & Kettinger, 2008;Kwon & Wen, 2010). If members perceive an SNS as being useful then they are more likely to be satisfied with the social network and be more engaged (Kang & Lee, 2010;Kwon & Wen, 2010;Lin & Lu, 2011;Sledgianowski & Kulviwat, 2009).
Studies have shown that there is a positive psychological outcome gained from engagement on an SNS (e.g. Burke, Marlow, & Lento, 2010;Ellison, 2007;Kim & Lee, 2011;Steinfield, Ellison, & Lampe, 2008;Valenzuela, Park, & Kee, 2009;Valkenburg, Peter, & Schouten, 2006). This positive psychological outcome is a mix of social support and sense of community, which determines the psychological well-being of members within a social network (Vieno, Santinello, Pastore, & Perkins, 2007). In fact, achieving this sense of community and social support can lead to satisfaction with SNSs (Kutek, Turnbull, & Fairweather-Schmidt, 2011;Manago, Taylor, & Greenfield, 2012;Oh, Ozkaya, & LaRose, 2014). Social support is defined by Cohen and Hoberman (1983, p. 100) as the "resources provided by one's interpersonal ties". Joining brand communities on an SNS can allow members to share their interests and form interpersonal similarity with other members of their community (Matzler et al., 2011;McAlexander et al., 2002). Interpersonal similarity between members gives rise to a form of interpersonal liking between members (Nicholson, Compeau, & Sethi, 2001). Friends' liking on SNSs provides a sense of social support, leading to greater satisfaction with SNSs (Vieno et al., 2007;Young, 2006). On that basis, we hypothesize the following: H3. : The higher the SNS's friend likeability, the stronger the users' satisfaction with the SNSs (Facebook).

Similarity with brand and satisfaction with SNSs
There are two main factors that lead to member satisfaction with regard to engaging on SNSs: social support and sense of community (Kutek et al., 2011;Manago et al., 2012). Establishing a sense of community is considered a crucial factor that determines the social well-being of members within a social network (Vieno et al., 2007). Brand similarity that gives a sense of belonging to a community is defined by the similarity between the characteristics of its members, their dependence on each other (Sarason, 1974), and their similarity with the brand of the online community they joined (Lam, Ahearne, Hu, & Schillewaert, 2010). This means that the more similar members feel to the online brand community they are part of, the more likely they are to be satisfied (Kutek et al., 2011;Manago et al., 2012). According to the brand congruence theory (e.g. Aaker, 1999;Alpert & Kamins, 1995;Sirgy, 1982), the greater the brand similarity, the more likely such similarity reflects the self-concept of individuals, leading to greater satisfaction. This is true not only off-line but also online on SNSs, as consumer engage more with brands on online communities in order to achieve such self-concept (Moran & Muzellec, 2017;Oh et al., 2014;Zhang, Hu, Guo, & Liu, 2017).
Consumers formulate a sense of similarity with the brand that they associate themselves with (Anselmsson, Johansson, Maranon, & Persson, 2008;Kuksov, Shachar, & Wang, 2013;Langner, Bruns, Fischer, & Rossiter, 2014). Once consumers realize the degree of similarity they have developed with the brand, they are more likely to develop a greater bond and experience greater satisfaction, not only with the brand itself but also with the platform on which its consumers and the brand engage. Since consumers join the brand online community in the first place because of their perceived brand similarity and interests in the brand related activities, consumers are likely to develop greater satisfaction with the SNS within which the consumers and the brand exist. If brands are able to maintain and enhance such similarity between the brand and the consumers, heightened satisfaction and engagement not only with the brand but also with the SNS will be experienced by those consumers. Hence, we hypothesize the following: H4. : The stronger the feeling of similarity with brands, the stronger the satisfaction with the SNS (Facebook).

Similarity with brand and customers' NFU
Brand similarity can motivate consumers who lack a feeling of uniqueness to look for a brand by which they can better reflect themselves and become more closely associated with. Thus, brand similarity can contribute to a customer's NFU only if its customers do not perceive that the similarity with the brand is shared with others within the social group. In such a situation, social comparison theory (Festinger, 1954) is in action, as the NFU drives the customer to avoid similarity with others (Snyder & Fromkin, 1977a;Tian et al., 2001), and to seek distinctive product designs (Bloch, 1995). However, in the case when brand similarity is not shared with others within the community then it can positively fulfill customer NFU. This is because customers' NFU is typically fulfilled when consumers associate or possess a brand that is perceived to be similar to their selves (e.g. Tian et al., 2001).
Within brand communities, it has long been argued that an implicit or explicit feeling of similarity with brands tends to develop amongst like-minded individuals who share similar interests (e.g. Chan & Li, 2010;Mathwick, Wiertz, & de Ruyter, 2008;McAlexander et al., 2003). As discussed under the brand congruence theory, the fit between the customer's own-self and the brand's personality positively impacts the customer's engagement and response to brands (Kim et al., 2004;Malär, Krohmer, Hoyer, & Nyffenegger, 2011). Successful brands are able to mirror customers' ideal or actual selves and are congruent with consumers, motivating such consumers to continue to verify and validate their self-image with the brand (Swann, 1983). Such similarity between the brand and the customer will have a significant positive impact on the way members of brand communities feel about themselves. Hence, a greater perception of brand similarity contributes well to achieving the uniqueness sought by consumers who are experiencing a NFU. Thus, we hypothesize the following: H5. : The stronger the feeling of similarity with brands, the better the consumers' NFU can be enhanced (enhances feeling of uniqueness).

Customers' NFU and satisfaction with SNSs
SNSs have improved the communication not only between brands and their consumers, but also between consumers and the brands they associate themselves with. Consumers who identify with certain brands join brand communities on SNSs to build upon their social capital (Lin et al., 2001;McKenna et al., 2002;Peter et al., 2005), and capitalize their social network (Boyd, 2004). Achieving the desired social capital ensures satisfaction with the SNS and its continued use (Hu & Kettinger, 2008). However, satisfaction with the SNS can be undermined when consumers do not achieve the social capital they aim for (Anselmsson et al., 2008;Kuksov et al., 2013;Langner et al., 2014), and when consumers' NFU is not fulfilled/enhanced within the online brand community. This can lead to reduced consumer engagement and satisfaction with the SNS itself.
The theory of the need for uniqueness states that whilst consumers avoid similarity with others in their social group (Snyder, 1992;Snyder & Fromkin, 1977a;Tian et al., 2001), failing to enhance their selfconcept and achieve the needed uniqueness can increase consumers' negative feelings and dissatisfaction. While this raises a question regarding the effectiveness of online brand communities in fulfilling consumers' NFU, a high level of consumer NFU reduces satisfaction with the SNS within which the brand community exists. Therefore, we hypothesize the following: H6. : The higher the customers' NFU, the lower the satisfaction with the SNS.

Methodology
The study focused on members of online brand communities that exist on Facebook. In order to demonstrate how friend liking and brand similarities impact customers' NFU and satisfaction with the SNS (Facebook), the study asked the target respondents to relate their answers to the brand community that they most prefer. This was important in ensuring that respondents had the psychological experience that reflected the constructs within the conceptual model. The type of brands the respondents related to is not regarded as important to this study. Because the conceptual model argues that consumers' NFU on SNSs are likely to be similar across different brand communities. In addition, the conceptual model was not developed to specifically reflect a particular brand.

Data collection
Prior to collecting the data a small pilot study was conducted in order to assess face validity. Seven respondents were asked to comment on the length of the questionnaire, clarity of the questions, and overall structure. Participants found the questionnaire to be adequate but suggested small modifications to the wording of three items. Thus, slight changes were made while ensuring that the meaning of these items was not undermined by the modification.
Data used to test the hypotheses was collected from a sample of Facebook users in the United States using Amazon's Mechanical Turk (MTurk) crowd sourcing marketplace. The questionnaire was posted in September 2016 and remained open for one month. The questionnaire contained three main parts. The first part contained general questions regarding the respondents' use of Facebook. Part 2 contained all item scales for the constructs in the conceptual model and the final part contained general questions asking for demographic information (e.g. as age, gender, and occupation). A total of 363 responses were received. A visual inspection of the data was made and it was determined that 22 surveys did not contain full information so they were dropped, thus resulting in a usable sample of 341 survey.

Sample profiling
The average reported Facebook usage of the respondents was less than 1 h per day (53%), followed by 1-5 h (39%), and 5-10 h (7%). The majority of respondents had used Facebook for over 5 years (72%), followed by 3-5 years' use (19%), and just 6% of respondents selecting 1-3 years' use and 3% less than 1 year's use. The respondents reported that they mainly used Facebook to stay in touch with friends (45%) and interacting with new friends (35%), followed by the need to stay up to date with information (13%), and other reasons (7%). The gender split was 51% female, 49% male. The majority of respondents were under 30 years of age (68%). The age group split resulted as follows: age 18-20 years (26%), 21-29 years (42%), 30-39 years (17%), 40-49 years (11%), 50-59 years (3%), and over 60 years (1%). This sample description demonstrates a good match with the actual population of users of Facebook in the US, as young adults, including students, aged 18 to 29 years constitute 88% of users of all social media platforms. Having 68% of our sample in the youngest age bracket is deemed to be representative as it takes into consideration the split per usage in social media platforms by age bracket (18-29 being the biggest, then falling systematically per older age brackets).

Measures
All constructs were measured using seven-point Likert scales and all were adopted from previous research. All items used in the study can be found in Table 1. The scale used to measure friend liking was adopted from Reysen (2005) and consists of five items. Brand similarity was measured using a 4-item scale reported in Thorbjørnsen et al. (2002). As for the need for uniqueness construct, the scale was adopted from Tian et al. (2001). The original scale consisted of twelve items measuring three dimensions. One of their dimensions, 'avoidance of similarity' best reflects the conceptualization of NFU utilized in this study. Accordingly, we utilized the five items developed by Tian et al. (2001) to measure NFU. Finally, user satisfaction was measured using a three item scale developed by defined by Lin (2008).

Analyses
All analyses were conducted using SPSS 20 and SEM using Mplus Version 7.4. Prior to testing the hypotheses, we assessed the validity and reliability of the items used to measure the constructs. To determine how well our measurement and structural models provided a fit to the data, we followed suggestions by Hu and Bentler (1999) to use CFI and IFI as incremental fit measures and SRMR as a measure of absolute fit, in addition to the χ 2 statistic. Following convention, we also reported RMSEA. We submitted the 17 items used to measure the constructs in our model to a confirmatory factor analysis using Mplus Version 7.4 (Muthén & Muthén, 2015). The χ 2 was significant (χ 2 = 262.51 (113) , p < .001) but the CFI (0.98) and IFI (0.97) values were indicative of very good fit, as was the SRMR (0.04). The RMSEA value was 0.06. In addition, all completely standardized parameter estimates were above 0.7 and all t-values for the item loadings were much greater than 2.0, both of which provide some evidence of convergent validity (Segars, 1997). Furthermore, the composite reliability results for all constructs were quite high and the average variance extracted values were all far above the acceptable cut-off value of 0.5. Table 1 presents the completely standardized factor loadings along with the t-values, composite reliability, and average variance extracted values. Discriminant validity was assessed using the AVE-SV method suggested by Fornell and Larcker (1981). This approach calls for the shared variance between two constructs to be compared with the average variance extracted for each construct. Discriminant validity is evident if the average variance extracted is greater than the shared variance (i.e. correlation squared) between two constructs, or alternatively if the square root of the AVE is greater than the correlation between two constructs. Table 2 provides the square root of the AVE values that can be compared to the correlations. As can be seen, for each pair of constructs the square root of the AVE is much greater than the correlations, thus providing evidence of discriminant validity. Finally, we assessed the reliability of the constructs using construct reliability. As can be seen in Table 2 these values are quite high, indicating some degree of confidence in the reliability of the scales. Based on the foregoing, we believe our measures are very adequate to be utilized for the hypotheses tests.
The test of the structural model represented in Fig. 1 provided an adequate fit to the data (χ 2 = 262.51 (113) , p < .001; CFI = 0.98; IFI = 0.97; SRMR = 0.04; RMSEA = 0.06). The first hypothesis proposes that friend liking would be negatively related to the need for uniqueness (NFU), a contention that was supported by our data (γ = −0.2, p < .001). This result is significant as it proves for the first time that similarities within online brand communities can negatively increase customers' NFU. The theory of uniqueness has long proved that this is the case with offline samples. Hypothesis Two suggests that friend liking would be positively associated with brand similarity perceptions and receives support (γ = 0.301, p < .001). This result is not surprising given that respondents are members of the community of their preferred brand. Hence, this result reflects what the brand congruency theory predicts. We also find support for H3, which proposes that friend liking will be positively related to the SNS (γ = 0.43, p < .001). In other words, friend liking and similarities within Facebook positively increase members' satisfaction with Facebook. This particular result can partly explain why Facebook is the leading and most successful social network site in the world. We also suggest that perceptions of brand similarity will be positively related to SNA (H4) and NFU (H5). Both hypotheses received support (H4: β = 0.42, p < .001; H5: β = 0.31, p < .001). The conceptual prediction for both hypotheses is confirmed by these results which reflect the importance of brand similarity in positively reducing the impact of customers' need for uniqueness and positively contribute to and enhance satisfaction with the social network site (Facebook). These results further confirm the crucial role that brands with communities on Facebook play in members' experience within social network sites. Finally, in H6 we suggest that there is a negative relationship between NFU and the SNS and find support for this (β = −0.09, p = .044). This result provides an interesting contribution to the literature, not because it is the first time such a hypothesis has been tested, but because it highlights the risk that customers' need for uniqueness can have on social network sites (Facebook). Such a result is important as most social network sites are not aware of this risk. (See Fig. 2.)

Discussion
The study employed the theory of brand congruence and the theory of uniqueness in order to generate insights on the impact of customers' need for uniqueness (NFU) and brand similarity on consumers' satisfaction with SNSs. By focusing on online brand communities within the SNS (Facebook), the conceptual model focuses on examining whether the paradox of similarity amongst liked friends within the online community exists in that it provokes an increase in customers' NFU, and the impact of this on consumer satisfaction with the SNS -Facebook. The findings provided full support for the conceptual model, proving that a high level of customers' NFU within online brand communities can be increased by perceived similarity amongst liked friends and can have significant negative impact on consumers' satisfaction with SNSs (Facebook).
Consumers engage in an online community to enhance self-presence and their own uniqueness (Bibby, 2008;Underwood et al., 2011), and to develop social identification (Pagani et al., 2013). However, in this study, it was argued that online social identification and friend liking reflect a high degree of similarities amongst members of the community within the SNSs (Kabadayi & Price, 2014;Rowley, 2004). The findings confirm that friend liking negatively increases consumers' NFU. The theory of uniqueness argues that perceived similarity can activate the NFU. Hence, it is not surprising that the results confirm that friend liking, which developed because of a high perception of similarity (Kaptein et al., 2013), indeed increases the NFU. While this is an interesting new understanding of customers' NFU on the online brand community, this finding was predicted by the combined power of the theory of uniqueness and the theory of social comparison in other contexts.
It is not surprising that liking between friends within an online brand community identifies and develops similarity with the brand. Previous studies have already shown that members of online communities formulate a sense of similarity for a brand which they identify themselves with (e.g. Kuksov et al., 2013;Langner et al., 2014). The sense of community and liking that develops between members of an online brand community facilitates identification with the brand leading to greater brand similarity (Torres et al., 2017). This finding fits well with the prediction of the theory of brand congruence in that consumers are likely to be congruent with brands that reflect either the ideal or actual self. Having said this, it is apparent that the feeling of liking within the community encourages consumers to be congruent with the brand. The theory of brand congruence has largely focused at the individual level and has not directly considered the impact of group/friends liking on the level of congruence with the brand. The work by Keller (1993) has already shed some light on the associations/ similarities between the social group identity and brands, which "can .97 a a Values on diagonals are the square root of average variance extracted for the construct. b p-Values in Parentheses. I. Abosag, et al. Journal of Business Research 117 (2020) 862-872 vary according to their favorability, strength, and uniqueness" (Keller, 1993, p. 5). Hence, brand similarity is contingent on the level of liking within the online community. Friends liking within online brand communities is found to influence consumers' satisfaction with SNSs (Facebook). Achieving liking and a sense of community has been predicted to increase satisfaction with SNSs (Manago et al., 2012;Oh et al., 2014). Since many consumers join online brand communities, to not only keep with their favorite brand but also socialize with other members of the community, friends liking within these brand communities is important to their satisfaction. Friend liking is an important factor in contributing to the overall psychological state within SNSs. Hence, this finding shows clearly that friends experience within online brand communities is not only important to the brand itself but also SNSs (Facebook). As rivalry amongst different SNSs is on the increase to engage more users, the ability to enable consumers to develop better engagement and interaction has important implications on consumers' satisfaction with the SNS (Facebook). Previous studies have largely focused on the psychological gains by consumers from engaging within SNSs (e.g. Burke et al., 2010;Ellison, 2007;Kim & Lee, 2011;Valenzuela et al., 2009;Valkenburg et al., 2006). However, our finding shows that this positive psychological gain is also important for SNSs (Facebook).
Brands that are able to successfully build online communities reflect the success by the SNSs to engage consumers and brands. Hence, the relationship between similarity with the brand and consumers' satisfaction with the SNSs (Facebook) is critical in showing whether the SNS (Facebook) is satisfying members of online brand communities. The finding shows that consumers who develop a sense of similarity with the brand are also able to develop satisfaction with the SNS (Facebook). This finding is interesting in that consumers who are congruent with brands and achieve self-concept within online brand communities develop satisfaction for the SNS (Facebook) within which the community exists. Existing literature (e.g. Moran & Muzellec, 2017;Oh et al., 2014;Zhang et al., 2017) has made the connection between consumers' perceived similarity with the brand in online communities and the achievement of self-concept. However, this study went further to make the link with the SNS (Facebook).
As predicted by the theory of brand congruency, brand perceived similarity is found to positively enhance customers' feeling of uniqueness. Such fit between customer and brand personality has long been found to reinforce and strengthen self-concept and self-expressiveness (e.g. Aaker, 1997;Kim et al., 2004;Malär et al., 2011). Online brand engagement that is able to reflect the ideal or actual-self encourages consumers to validate their own images with the brand they are congruent with. This is not different to the role played by brands offline. However, such congruence between the brand and consumers within online communities clearly significantly enhances consumers feeling of uniqueness which has positive implications for the success of online brand community.
Perhaps the most significant finding from this study is related to the negative impact of customers' NFU on customer satisfaction with the SNS (Facebook). Customers with high NFU tend to lower their satisfaction with the SNS (Facebook). Hence, high customers' NFU can undermine the social capital that consumers' aim to generate values through their engagement with online communities and through being interactive with the SNS (Facebook). Existing studies have argued that consumers engage with SNSs to increase their social capital gains (e.g. Langner et al., 2014;Peter et al., 2005), which increases their satisfaction with the SNS (Hu & Kettinger, 2008). However, the finding from this study clearly raises concerns for SNSs as consumers with high NFU do not realize or achieve the social capital they are after. The theory of uniqueness argues that consumers who are less successful achieving uniqueness will tend to avoid similarity with others and may withdraw from direct engagement (Snyder, 1992;Snyder & Fromkin, 1977a;Tian et al., 2001), which is the direct result of the negative feeling and dissatisfaction with the environment within which their engagement takes place. While the brands positively impact consumer as expected, the question is why has the community of that brand led to the negative increase in customers' NFU?

Implications
Consumers' experience on SNSs, namely Facebook, is mainly driven by (1) socialization with friends, (2) the brand experience based on friends' endorsements, and (3) the social experience itself. The base of the social experience is driven predominantly by friends on the SNS. The more similar they feel, the higher the need for uniqueness. Likewise, the less people feel similar to their friends on the SNS, the less they will feel the need to be unique. Our findings show that an increased level of customers' need for uniqueness within SNSs should be a key concern for SNSs and brands alike, as it will significantly affect the monetization model of the site. Although studies on the self-perceptions of uniqueness are limited, almost all studies have focused on pure consumer behavior, and little or no attention has been paid to the role of brand similarity in enhancing/undermining consumer self-perception of uniqueness, especially on SNSs. Consumers' self-perception of uniqueness influences their responses to the brands' engagement efforts as well as their relationships with these brands. The paradox of brand similarity in relation to customers' uniqueness seeking behavior has not been sufficiently examined in previous literature. This study contributes to the existing brand literature by identifying and examining such a paradox, with significant implications for brands and for SNSs, which are becoming the favorite platforms for brand engagement.
As the competition amongst SNSs intensifies, maintaining satisfaction with users who may have a high level of NFU is going to be challenging. This would particularly happen amongst SNSs that tend to attract popular brands that have a high similarity level with users. Whilst Facebook has a clear gap over other competing SNSs, the negative impact of customers' NFU on satisfaction with SNSs can be serious and may lead customers to join other SNSs that they may perceive to enable them to fulfill their NFU better, e.g. leaving Facebook to join Instagram even when both SNSs offer different interaction and services. Hence, the negative impact of customers' NFU on SNSs is worthy of further investigation. The managerial implications for this research are straight forward: first, we suggest that NFU is an important asset which SNSs can make use to their advantage. However, as the competition amongst SNSs intensifies, maintaining satisfaction with users who may have a high level of NFU is going to be challenging. This would particularly happen amongst SNSs that tend to attract popular brands that have a high similarity level with users. As such, brands need to develop strategies to respond to customers' NFU, even for those with high levels of similarity.
Second, our findings imply that the negative impact of customers' NFU on satisfaction with SNSs can be serious and may lead customers to join other SNSs that they may perceive to enable them to fulfill their NFU better, e.g. leaving Facebook to join Instagram even when both SNSs offer different interaction and services. Hence, managers much take measures to mitigate the negative impact of customers' NFU on SNSs.
Furthermore, extant research indicates that managers of highly engaged brands on SNSs have little or no knowledge of how to manage customers with high levels of NFU within their online brand communities. To mitigate the negative impact of customer's NFU on SNSs, we suggest that managers need to develop their awareness about the challenge that customers' NFU creates. Similarly, managers of SNSs including Facebook need to develop their awareness and knowledge about customers' NFU and develop strategies to successfully maintain their relevance, not only of the SNS itself, but also the relevance of brands that have created successful online brand communities.

Limitations and future research
The study focused on Facebook without a specific focus on individual brands. We suggest that focusing on similarity with a few key brands may bring further insights into the paradox of similarity and its impact on brands and satisfaction with SNSs. This study did not examine whether the reduced satisfaction will actually lead those affected customers to switch to other SNSs, with implications not only for Facebook, but also for brands within Facebook; therefore future studies should examine this further, as this study only includes satisfaction with Facebook. Other output constructs should be considered including intention to switch, perceived image, and loyalty.
The findings from this study show that friend liking was the main cause of the negative increase in customers' NFU and not the similarity with brand. Hence, future research should examine further the role of customers' interactivity within online communities to further understand how such a decrease in a feeling of uniqueness happens. Whilst the theory of uniqueness successfully explains customers' psychology on how the gap of uniqueness comes to exist, future research needs to investigate whether or not there are factors/variables that contribute to the increase of NFU within online brand communities, particularly factors/variables that are directly related to the brand of that community.
In addition, future research should retest the conceptual model on other SNSs that have different community dynamics such as Instagram, Twitter, Snapchat, etc., especially with a longitudinal research design. Testing the conceptual model across different SNSs with a longitudinal design will be useful not only to validate the findings but also reveal the mechanism through which similarities and customers' NFU impact of customer satisfaction as SNSs evolves with time. In particular, future research may also need to assess the degree of interactivity and similarities within other online non-Facebook based communities and find whether such similarities increase customers' NFU, as found by this study.
Furthermore, in this current study we did not specify any brand community, hence future research may want to consider examining customer NFU on specific brand(s). As such research will be very helpful to help managers to understand why customer NFU is relevant to a particular brand and to help them to formulate further strategies to use customer NFU as a strategic asset. In addition, whilst this study has focused on the preferred brand community by respondents, future research may examine respondents who are members of more than one brand community.
Finally, the findings from this study clearly show that customer NFU can itself trigger consumers' resistance to the SNS platform itself, which could lead to less engagement. Hence, future research needs to examine the relationship between customers' NFU and consumers' resistance behaviour, which to our best knowledge, has not been examined previously. Additionally, the direction of the relationship between customers' NFU and friend liking and similarities should be examined as in this study we tested, as suggest by the theory of uniqueness, the impact of friend liking on customers' NFU. The opposite should be considered.