Company versus consumer performance: does brand community identification foster brand loyalty and the consumer’s personal brand?

This study investigates the effect of community identification in building brand loyalty (attitudinal and behavioural) and a personal brand via social networks. The proposed model explained the main community identification antecedents and how identification can lead to enhanced company and consumer performance. This study adopts the theoretical lens of uses and gratifications theory and identifies three motivational drivers of community identification: relationship-oriented motives, self-oriented motives and brand content-oriented motives. The data were analysed with a structural equation modelling method based on a convenience sample collected through a survey. This study extends the body of knowledge about the outcomes from an active social media usage, based on a UGT perspective. It relates community identification and its drivers to loyalty. Besides, it links community identification with personal branding, which is considered as a vital outcome expected by social media users. The findings suggested that self-oriented motives represented a key driver for taking part in an online community. In addition, community identification represents to be an important antecedent to build attitudinal loyalty rather than behavioural loyalty. Consequently, community identification was also found to be a significant driver for building a user’s personal brand.


Introduction
In 2019, 72.4% of the worldwide online population had accessed social networks, up from 69.6% in 2016 (Statista 2020). This increase justifies the constantly growing attention given to virtual communities that integrate people. Today, virtual communities, including increasingly popular brand communities, are a special form of consumer communities, defined as a specialized, non-geographically bound link between consumers, which are part of a structured set of social relationships among fans of a brand containing three common markers: consciousness of kind, shared rituals and tradition, and moral responsibility (Muniz and O'Guinn 2001).
The relationships between virtual communities' members and brands, and the consequent community identification, are still confusing to companies and have become a major issue for brand research (e.g. Zaglia 2013;Wolter et al. 2016;Tuškej et al. 2013;Manchanda et al. 2015;Coelho et al. 2018;Pedeliento et al. 2020). One key point is the existence of two relationships: one relationship between consumers' brand identification and brand attachment and one among consumers that join a community and build such relationships with the community members (Millán and Díaz 2014). Thus, it is important for brand studies to explain one of the most important motivations for joining online communities, supporting the thesis that consumers who are affiliated with a group tend to be less radical in rejecting brands than are individuals who are independent from a social group (Escalas and Bettman 2005). In fact, brand enthusiasts simultaneously perceive a social identity with both the brand community and its social network environment (Zaglia 2013). In addition, beyond social ties with community members and the brands, the previous research claims that another major motivation for using social media is self-presentation (Seidman 2013;Teichmann et al. 2015). To support this claim, Orehek and Human's (2017) findings suggested that social media platforms could be an effective outlet for self-expression, which is a form of affirmation of one's self as originally defined by Ashforth and Mael (1989). As Hwang and Kandampully (2012) pointed out, self-concept is particularly fundamental for young customers' relations with brands and other customers.
These dynamics are critical not only for the survival and success of social networks but also to determine the intensity with which consumers identify themselves with a particular brand (Millán and Díaz 2014); such identification is built around a brand, its core asset, but ultimately grows and persists because of the relationships among its members (Jang et al. 2008). Therefore, it is important to consider and to separately evaluate how these two dimensions, namely consumer-brand identification and consumer-other consumers identification, are influenced by different motives that lead consumers to join online communities and subsequently how these two identifications can help to build the overall community identification. Community identification considers the strength of the consumer's relationship with the brand community (Demiray and Burnaz 2019). Referring to Bagozzi and Dholakia (2002), the community identification can be influenced by internalization (i.e. congruence of one's goals with those of group members) and identification (i.e. the conception of one's self in terms of the group's defining features).
Despite the recognized importance of this topic, the composition of community identification, formed partly by consumer-other consumers ties and partly by consumer-brand identification, has not been measured yet. Additionally, there is a lack of understanding concerning what drives and builds community identification and how such drivers influence its two parts. Moreover, further research is needed regarding the outcomes of community identification (Mandl and Hogreve 2020).
Stokburger-Sauer (2010) were among the earliest to find a significant relationship between consumer-brand identification and brand loyalty, but their research was applied to the offline context without considering such identification built and developed through social media brand pages. The same is true for Marzocchi et al. (2013), who explored the impact of identification constructs on loyalty-related outcomes within an offline brand community of motorcyclists during an international "brandfest" in Italy.
Similarly, Wolter et al. (2016) found that consumer-brand identification leads to brand loyalty; also exploring the opposite relationship, they found that consumer-brand disidentification leads to brand opposition. Additionally, Millán and Díaz (2014) discovered that brand identification has an effect on consumer loyalty and word-of-mouth communications. The same was found in a recent study by Elbedweihy et al. (2016). However, none of the cited studies were related to the online context or focussed on the identification via social media brand pages.
A literature review by Alves et al. (2016) showed that this topic is still unexplored. Similarly, Black and Veloutsou's (2017) findings suggested that there was extensive borrowing of identities among the brand, the individual and the brand community offline and online.
Considering the online context, a recent study found that consumer identification with a brand is stronger when consumers perceive it as more prestigious and more humanlike (Tuškej and Podnar 2018). The same study also showed a positive impact of consumer identification with a brand on their engagement with the brand, while a very recent study found that when customers' identification with a brand community becomes salient, they strengthen their emotional attachment to the brand and improve their centrality in the network (Chang et al. 2020). In addition to company/brand performances, community identification can provide benefits for consumers in terms of a stronger personal brand. This less explored phenomenon represents a tailored set of selfmarketing activities (Malone and Fiske 2013). Therefore, social media can enable self-expression and self-presentation and, thanks to the strength of community relationships with other consumers and with the brand, can help to build a strong personal brand for social media users.
Hence, the present study contributes to extant brand management research, examining users' motives that lead them to identify themselves with brands but also to identify with other consumers and to assess the influence of such brand community identification on loyalty generation processes (attitudinal and behavioural) and users' personal brand building.
This investigation sheds more light on what drives social media users to identify themselves with a brand community and with its members and how these ties lead from one-side brands to reach brand loyalty via an effective hosting of the community, from the other side they help users to build their personal brand within this community. This is not only a gap in the literature, but it represents a relevant aspect to explore for social media managers that strive to create a fertile context to gain brand outcomes such as loyalty or attachment.
This study adopted and extended the uses and gratifications theory (UGT) to explore the motives that drive consumers to stay connected with other members and to use brand pages on social media. The uses and gratifications theory grants that when a particular medium fulfils the expected satisfaction, this leads users to constant use of this medium (Katz et al. 1974). Hence, UGT attempts to explain not only why social networking sites usage, ultimately leading to gratification (Bae 2018), but also it is helpful in understanding more in depth the consequences of the active usage of a specific media.
UGT theory tries to explain why individuals have different media usage patterns, holding that people actively use media to satisfy various needs and to achieve their goals or specific gratifications. Katz et al. (1973) introduced five general categories of motives: cognitive (information), emotional (entertainment), social (connection), personal integrative and escape. Socialization and social support (Bae 2018) are noted as key categories that bring users satisfaction from social media. Such gratification motives applied to the research context of social media are related to the so-called next generation of gratification where users seek interactivity which is related to activity, responsiveness, choice, control and flow deriving from social media usage and the interactions with other users (Sundar and Limperos 2013).
To summarize, the main motives for social media usage of socialization and social support identified by Bae (2018) have been allocated into three main sub-categories, namely relationship-oriented motives such as social interaction; selfrelated motives such as self-expression; and brand contentoriented motives such as brand distinctiveness, hedonic and utility functions (Saboo et al. 2016;Stokburger-Sauer et al. 2012).
From a theoretical perspective, this study aspires to expand significantly the current body of knowledge related to UGT measuring how different motives to join online communities would lead consumers to identify with the brand community and its members and how through such identification, they will build their personal brand and at the same time increase their brand loyalty. Such relationships within the context of social media have rarely been investigated. Recently, apart from Bae (2018), also Abid and Harrigan (2020) have explored social media-enabled voter relationships through uses and gratifications theory via qualitative methods. Li et al. (2015) focused on the gratification motives of online gamers, whereas Gan and Li (2018) identified media appeal as a critical motive of WeChat communicator users. Furthermore, Rokito et al. (2019) explored the gap between gratifications sought from Facebook use and gratifications obtained from it as a key mechanism of frequent returns to Facebook. Although all they are relevant as they help to understand the value and the outcomes of the relationships created via online communities from a UGT perspective (Claffey and Brady 2017), there is the need to better investigate the outcomes obtained via the gratifications perceived through users' identification with social media. Hence, with the direct reference to Bae (2018) findings, the research aims are the following: • To understand the influence of (a) relationship-oriented motives such as social interaction, (b) self-related motives such as self-expression and (c) brand contentoriented motives such as brand distinctiveness, hedonic and utility functions of SNS-bc on consumers' identification with the brand and other consumers; • To explore the impact of consumer-other consumers and consumer-brand identification on building community identification; • To investigate the effect of the brand community identification on enhancing brand loyalty (both attitudinal and behavioural) as a company-expected effect of community hosting and on reinforcing a personal brand, which is an effect of community identification awaited by consumers in the social network context.
In doing so, this paper begins with a background section on the literature followed by a presentation of the conceptual model for this research, along with the related hypotheses. Then, the methodology part discusses the appropriateness of the method and sample adopted, and the main findings of the research are presented. The paper concludes with a discussion of these findings, their implications and the future direction of this stream of research.

Uses and gratifications theory
The uses and gratifications theory (UGT) is the theoretical lens through which this study views this deep, complex and broad study on social network communities and related brands. Cantril (1941), who studied how and why individuals use different mass media, developed this theory. Since then, several studies have adopted the UGT. Although differing in their starting points and in their frameworks, they have been based on the assumption that people are active, are aware of their needs and choose different media to satisfy them. Based on this, UGT has become suitable to be extended to other media such as the Internet and social media (Ko et al. 2005;Ruggiero 2000;Whiting and Williams 2013).
As noted above, UGT theory tries to explain that individuals use media to satisfy various needs and to achieve their goals and gratification from this usage. Gratification is typically defined as a characteristic of satisfaction experienced by individuals that is related to the active use of the media (Stafford et al. 2004). In general, this theory provides explanation for users' motives of using media, factors that influence these motives and the outcomes of media usage (Liu et al. 2020).
Among the UGT fundamental assumptions, one is that people are in an active position, while they are choosing the communication media, which satisfy their needs based on a set of gratifications (Eginli and Tas 2018).
The theory focuses more on what individuals do with the media than what the media has on the person (Katz et al. 1974). Based on that, individuals adopt media per specific reasons, and they obtained satisfaction from such choice (Stafford et al. 2004).
According to this, within the context of social media, UGT is helpful in understanding which social media are adopted and the related motivations (Quan-Haase and Young 2014). Not only, UGT also explains how social media are used to satisfy needs and the main motivations behind certain behaviours deriving from this usage (Eginli and Tas 2018). For instance, Iyer et al. (2018) adopted UGT to investigate how the gratification derived from social media explains consumer attitudes. Besides, Schivinski and Dabrowski (2016) adopted UGT to examine users' outcome behaviour from social media usage. In particular, they showed the intention to purchase products and services as a result of the benefits consumers perceived not only via the interaction with the brand in the online community, but also thanks to the positive attitude they have for a particular brand obtained via the social interaction they have with the online brand communities members. To summarize, UGT is useful not only to understand the main motives users have to join and be part of social media but also to explain their consequent active behaviours in this context (Corrada et al. 2020). The previous research explored different categories of motives/gratifications when investigating the context of social media via the adoption of UGT.
For instance, Sicilia and Palazon (2008) explored the main motivations for users to join virtual communities and found that the main reasons are related to social support and entertainment rather than informational value motives. Furthermore, Park et al. (2009) examined the link between users' gratification perception from joining Facebook groups and their participation in political and civil life in the offline context. The main motives that emerged from this research were related to socializing, entertainment, self-status seeking and information.
Jahn and Kunz (2012) provided a clear distinction among three groups of needs: a relationship-oriented set of needs based on social interaction among users; a self-oriented group of needs based on the specific needs of individuals, such as achieving status or diversion; and a third group related to content-oriented needs regarding the information delivered by the brand through the media, in this case, from the brand SNS.
Following Jahn and Kunz (2012), De Vries and Karlson (2014) adopted functional, social, hedonic values and brand relationship characteristics measured by the brand strength. These drivers were linked to community engagement; therefore, co-creation value was measured as an antecedent of engagement.
Drawing from the UGT framework, Claffey and Brady (2017) identified from the literature five motivational drivers of consumer engagement that relate to utilitarian, hedonic, personal self-enhancement, and social and personal integrative needs. This study summarized the previous research supporting the choice of the three motive categories adapted from Jahn and Kunz (2012), which is the study with the highest number of citations. It clearly divides the motives to follow and participate in social media and encompasses the most adopted motive categories from the previous research in the same field. Bae (2018) identified socialization and social support as the main motives for social media usage. Hence, following her and being inspired by Jahn and Kunz (2012) these motives has been expanded into three main sub-categories, namely relationship-oriented motives such as social interaction; self-related motives such as self-expression; and brand content-oriented motives such as brand distinctiveness, hedonic and utility functions (Saboo et al. 2016;Stokburger-Sauer et al. 2012). Hence, the main gratification motives are explained as follows.

Relationship-oriented gratification: social interaction motive
Social connection and interaction are considered one of the primary functions of social media. This category includes connections with friends and family (Heinonen 2011), finding old acquaintances and keeping in touch with people who live far away (Whiting and Williams 2013). In addition, users connect and make friendships with people whom they do not know but with whom they share similar interests (Whiting and Williams 2013). To support this finding, the previous research found users may use social media as a substitute for companionship (e.g. Lee and Cho 2011;Zolkepli and Kamarulzaman 2015).
Social media allows people to create and manage a network in which members can collaborate and interact with everyone at the same time (Pedeliento et al. 2020). Users can benefit from connecting with other consumers (Hajli 2014;Stephen and Toubia 2010), celebrities and businesses (Hajli 2014;Ilicic and Webster 2016;Saboo et al. 2016), which also provide online consumer services (Chua and Banerjee 2013;Karakaya and Barnes 2010). Social connection fosters community building (Ngai et al. 2015) and a sense of belonging and bonding (Heinonen 2011;Zolkepli and Kamarulzaman 2015), not only with community members who can become friends but also with a brand itself via its community (Saboo et al. 2016;Kaur et al. 2020).
From a UGT perspective, the use of media can be explained in two ways: one derives from a content-related gratification and one from an experience-related gratification (McGuire 1974). When applying this statement to social media use compared with traditional media, it can be expressed that content-related gratification is obtained not only by brand content but also by creating more chances for communicating and interacting with the people (Eginli and Tas 2018). This is in line with the previous research where the work by Ellison et al. (2007) on Facebook's effects on social capital, found that social gratifications are the motivating factor for people using Facebook, providing social ties to people and especially creating connections with members of the community.
Hence, social interaction with other community members can be perceived as a gratification motive that leads consumers to feel a part of an online community and consequently strengthens the ties among members of this community and with the brand that hosts such a social network.
Based on the above, the following hypotheses were proposed:

H1
The social interaction motive positively affects the relationships with other members of the same SNS-bc (consumer-other consumers).

H2
The social interaction motive positively affects the relationships with the brand on a particular SNS-bc (consumer social network brand identification-CsnBI).

Self-oriented gratification: self-expression motive
In addition to social interaction, consumers can decide to participate in a particular community because they expect it to have an influence on their image or status. Based on UGT, this form of gratification, known as "self-expression", can also be gained from connecting with, liking and sharing a particular brand (Saboo et al. 2016). Social media can be used by users to promote themselves, share information and build a status that makes them feel important (Lee and Ma 2012). This point is consistent with Leung et al. (2013), who emphasized how social media helps narcissists in their self-presentation, such as through the sharing of selfies. Therefore, the display of the self on social media can be helpful for enhancing self-confidence and self-esteem (Phua et al. 2017), as can helping others with advice and suggestions. Such self-orientation gratification is not only a driver to understand why users join and use social media but, based on UGT assumption, can explain one potential outcome deriving from this gratifying motive, namely the identification with a specific online brand community where the user actively belongs to and the relationship she/he has with its members. This is consistent with a recent work by Simon et al. (2016), who pointed out that self-image expression positively affects consumer-brand relations and identification. In fact, these authors posited that identification with a brand is derived from its functionality, which means that brands are identified with the extent to which they fulfil important personal needs/desires of the consumer, such as self-representation (e.g. Belk 2013;Wolter et al. 2016). In addition, as individuals strive for positive self-esteem, they are motivated to maintain a positive self-defining view of themselves and of their relationship to others (Abrams and Hogg 1988;Wolter et al. 2016;Kucharska 2019). Therefore, coherently with UGT, the highest horizon of gaining the expected self-expression gratification is, the more intensive is the identification with the brand and simultaneously, the more intensive is the interaction with other fan page members who follow the same brand. Based on these premises, the following hypotheses were proposed: H3 The self-expression motive positively affects the relationships between other members of the same SNS-bc that they follow (consumer-other consumers).

H4
The self-expression motive positively affects the stronger identification between the SNS-bc particular member and the brand (CsnBI).

Brand content-oriented motives
Brand content-oriented gratification motives concern the functional and hedonic values that are delivered from the brand (Hirschman and Holbrook 1982). The previous social media research has confirmed that both entertainment and information are important issues when users decide to actively participate in an online community (Khan 2017;Fernandes and Castro 2020).
The brand needs to provide the best possible content and to engage the consumers to foster the consumers' sense of identification with the brand. This goal can be achieved if a brand is perceived as being different from other brands and is easily identifiable. That is, brand distinctiveness can be seen as the perceived uniqueness of a brand's identity in relation to its competitors; consequently, it can become a driver of consumer-brand identification (Stokburger-Sauer et al. 2012). Consistent with this statement and based on the UGT perspective, one of the possible justifications of social media participation and usage by users is the perceived distinctiveness and uniqueness of a brand, which could represent a pivotal antecedent to explain a further outcome deriving from an active social media usage (Wang et al. 2012). That is the consumer's desire to identify herself/himself with that brand and to take part in its community and related members. Hence, we proposed the following hypothesis: H5 The more a consumer perceives a particular brand to be distinctive, the more the person will identify with that brand through social media (CsnBI).
In addition, the other two benefits categories can justify not only the reason why users join online communities but also they are useful to explain why users identify themselves as part of a particular brand community and its members. These two benefits are based on the perception of hedonic and utilitarian gratifications, where the hedonic benefit comes from the whole impression derived from the experience with a particular brand page reflecting the level of affective attachment. The utilitarian benefit is derived from the page performance indicating the perceived functionality (Voss et al. 2003) and its convenience (i.e. easy to use this page to find information) (Shao 2009). These benefits are consistent with UGT perspective, through which the previous research has identified that among the prominent needs of users' media choice and usage there are: a relationshiporiented area based on social interaction with others (which is reflected on the experience and engagement users have with the brand page and with other members) and a content-oriented area based on the information delivered by the media (Jahn and Kunz 2012).
This has been supported by the previous research on the important role played by entertainment and information for social media users (e.g. Sánchez-Casado et al. 2018;Raacke and Bonds-Raacke 2008;Sheldon 2008).
Based on this definition, it can be assumed that hedonic benefits can be associated with entertainment, whereas practical benefits are more connected to the user's pragmatism.
Considering these benefits, the identification with a specific group/community may derive from the firm/brand and its initiatives via the online brand communities (Demiray and Burnaz 2019). For instance, elements that are managed by firms which are critical in the assessment of membership of a brand community are related to the richness and reliability of the information provided (Hausman and Siekpe 2009), a user-friendly navigation (Casaló et al. 2010) and attractive content and visuals (Kim et al. 2003). These features require minimum efforts to users who have limited time, and so, they can maximize their utility and gratification from the media usage.
According to previous studies, which adopted similar perspectives to the online branding context (Jahn and Kunz 2012;Pongpaew et al. 2017;Carlson et al. 2018), it was hypothesized that both of these benefits can become vital users' motives that strengthen their identification with a brand. Hence, these benefits are not only seen as drivers that justify social media adoption but also they can lead to users' willingness to feel part of a particular brand community. Thus, we hypothesize that: H6 Higher hedonic functions of the brand page content lead to a higher degree of consumer-brand identification through SNS-bc.
H7 Higher utilitarian functions of the brand page content lead to a higher degree of consumer-brand identification through SNS-bc.

Community relationships in social media: brand versus consumer-other consumers identification
If it can be argued that there is a sense of identification towards a social media brand page, this sense can be derived from the relationship that consumers have with other consumers and with the tie they can have with the brand that hosts the community. Muniz and O'Guinn were the first to introduce the concept of brand community, defining it as "a specialized, non-geographically bound community, based on a structured set of social relationships among admirers of a brand", which emphasized the brand-mediated relationships among consumers (1995,2001). Based on this definition, the concept model of the brand community triangle was introduced, replacing the traditional single-dimensional "consumer-brand" relationship model and providing more emphasis on the relationship among consumers.
Social media represents a fertile context that allows community members to interact with each other at their convenience (Habibi et al. 2014). The essence of social media is the relationship networks that are built by consumers (Pedeliento et al. 2020). For this reason, maintaining harmonious relationships among consumers is a crucial driver of success and the survival of a social media brand page ). In addition, consumer-brand relationships that integrate consumer-other consumers relationships contribute to enhance such harmony , and the engagement consumers have with the brand depends on the benefits they perceived from brand actions on social media (see, for instance, Schivinski et al. 2019;Khan et al. 2019). The former focus on interactions with consumers and the brand or the company behind the brand; a point that has been confirmed in the literature concerning the brand relationship that has shown that consumers tend to invest in a relationship to a brand (Algesheimer et al. 2005;Fournier 1998). These consumers are willing to identify themselves with that brand in a type of consumer-brand identification (Stokburger-Sauer et al. 2012). There are several definitions of consumer-brand identification, ranging from "a consumer's psychological state of perceiving, feeling and value in his or her belongingness with a brand" (Lam et al. 2010, pp. 130) to including not only the dimensions of positive emotional connection and self-brand integration in their construct of brand love but also positive brand evaluations (Batra et al. 2012).
Social network users can identify with brands (e.g. on a Facebook brand page) through many approaches to virtual consumption contribution or the creation of brand-related content (Shao 2009). Hence, understanding people's motivations to engage in brand-related social media use is imperative (Muntinga et al. 2011).
With respect to the CBI construct, which regards consumer-brand identification in the real world (Bhattacharya and Sen 2003;Stokburger-Sauer et al. 2012;Wolter et al. 2016), CsnBI differs from the former, as the identification in the virtual world is reached via acclamation, while identification in the real world is gained by acclamation and physical possession.
The previous research has found that the individual's social identification with a brand community depends on a relationship with the other community members and the relationship with the brand (c.f., Bagozzi and Dholakia 2002;Carlson et al. 2018;Zhou et al. 2012).
However, most of the brand community research predominantly focuses on customer-brand relationships; customer-customer relationships in the community have received less attention despite its contribution to building community identification .
What is different in the current study is that the consumer-brand identification in a social network (CsnBI) has been separated from the necessity of possession of the branded product to make this identification visible for others. Namely, the identification with the brand in the online environment, contrary to offline (CBI), does not require spending money to use the brand image for the users' purpose. Therefore, the stronger ties between users focused on the same brand, the most substantial influence of these ties on community identification. This is in line with the previous research by users and gratification researchers investigating social interaction motivation such as Ko et al. (2005) or other similar categories, such as social motivation explored by Korgaonkar and Wolin (1999). What the present study aims at exploring is not only to investigate whether the gratification deriving from social interaction and the relationship with other members differs from the one obtained via the relationship users have with a brand, but also to understand how the two relationships contribute in building the identification users have with the brand community. Hypotheses 8 and 9 aimed to verify this issue: H8 Stronger consumer-other consumers relationships positively influence community identification driven by SNS-bc.
H9 Stronger consumer-brand identification in a social network (CsnBI) positively influences community identification.

Attitudinal and behavioural loyalty in social media
As discussed earlier, online communities that confer a strong sense of belonging and identification on their members can provide several benefits and gratifications. Consequently, it is assumed that identification with an online community has a positive effect on both the user and the company (Rosenbaum 2008; Hartmann et al. 2015). For instance, Thompson and Sinha (2008) found that higher levels of participation and longer-term membership in a brand community not only increase the likelihood of adopting a new product from the preferred brand but also decrease the likelihood of adopting new products from opposing brands. This result can be interpreted as a result of the perceived gratification users have from joining a specific online brand community, and it can be explained from a UGT perspective. Thus, it highlights that not only users adopt a medium as they perceive gratification from its usage, but it further explains the potential outcome deriving from the active usage of such media based on its contents and attributes (Katz et al. 1973). In the social media context, except few exceptions, there are no studies, which have explored how the gratification generated by the adoption of social media promotes the intention to purchase and repurchase products and services (Corrada et al. 2020). To the best of our knowledge, only Corrada et al. (2020) very recently found a positive relationship between the purchases and repurchases of products and services as the result of the satisfaction of using social media and in accordance with the gratification that consumer experiences when using it.
Furthermore, not only were purchase intention and new product adoption investigated from the previous research but also one of the central dimensions for understanding the effect on a company of good relationship marketing derived from social media is provided by brand loyalty (Kim and Lee 2019).
When considering the definition of loyalty, Oliver (1999) stated that this factor consists of "a deeply held commitment to rebuy or repatronize a preferred product/service consistently in the future, thereby causing repetitive same brand or same brand set purchase, despite situational influences and marketing efforts having the potential to cause switching behaviour" (pp. 34). Specifically, this definition highlights the importance of two important components of loyalty: an attitudinal component (i.e. commitment) and a behavioural component (i.e. purchase, patronage). Bowen and Shoemaker (1998) described attitudinal loyalty as a customer's feasibility to return to a particular provider recommended to him by, e.g. word of mouth, references or publicity about a product or service to others. These alternatives present attitudinal loyalty as the first step before behavioural loyalty. The attitudinal component of loyalty reflects customers' psychological attachments and advocacy towards a particular product or service (Chaudhuri and Holbrook 2001).
The previous marketing research has discussed several issues with regard to the antecedents of brand loyalty and social media, from brand trust (Laroche et al. 2013) to consumer engagement (Dimitriu and Guesalaga 2017) and long-term membership in the community (Thompson and Sinha 2008), and from community commitment to consumer identification with the brand . The latter factor is a strong brand loyalty antecedent, contributing to enhancing a company's performance and its longterm success (Stokburger-Sauer et al. 2012;Kim and Lee 2019). Consumer-brand identification inhibits consumers from switching brands, and a recent study demonstrated that consumers who highly identify with a brand evaluate the price of a product/service more favourably than do other consumers (Popp and Woratschek 2017). Tuškej et al. (2013) found that consumer-brand identification leads to a higher commitment to a brand and to higher willingness to generate positive WOM. Similarly, Fatma et al. (2016) explored the role of consumer-brand identification in enhancing satisfaction, affective commitment and, therefore, brand loyalty. Additionally, a recent study found that consumer-brand identification is "able to immunize brands from negatively charged emotions elicited by unfavourable brand comparisons and protect the consumer's self from discomforting psychological stages induced by suboptimal purchase decisions" (Davvetas and Diamantopoulos 2017, p. 225).
Despite its importance and power, there has been little previous research on the topic of identification with a community and its effect on brand loyalty in the online context. For instance, Kuikka and Laukkanen (2012) examined attitudinal and behavioural loyalty in social media, but they correlated them and skipped the community relationship context.
Two recent studies proposed that consumers' identification with the brand community is a mediator for brand loyalty without the distinction between attitudinal and behavioural loyalty (López et al. 2017;Popp et al. 2016). Hence, the present study aims at investigating the community identification effect on attitudinal and on behavioural loyalties separately. This separation will shed more light on the different effects these dimensions might bring. Namely, exploring two various forms of loyalty to a brand can help understand better the essential purpose of the online brand community management, distanced from purchase transactions. In doing so, this study is among the first to investigate under the UGT perspective the result of the gratification derived from the identifications consumers have with the brand community they join in the online context. In particular, it extends very recent research on purchase and repurchase intention as a result of the gratification perceived from an active social media use (Corrada et al. 2020) and does so considering two outcomes, namely attitudinal and behavioural loyalty.
Therefore, the following hypotheses were developed for this research: H10 The more consumers identify themselves with the particular brand community, the more loyalty (attitudinal) they have to that brand.

H11
The more consumers identify themselves with the particular brand community, the more loyalty (behavioural) they have to that brand.

Building personal brand in the social media context
To enhance the understanding of the impact community identification and the derived perceived gratification have on other behavioural intentions, a very recent study suggested the need for continued research, which could investigate customer outcomes other than repurchases (Mandl and Hogreve 2020).
In fact, in addition to brand loyalty, community identification can also provide benefits for consumers in terms of a stronger personal brand. This less-explored phenomenon represents a tailored set of self-marketing activities (Malone and Fiske 2013). The crux of personal branding is that it is a planned process (Karaduman 2013;Khedher 2014). Personal branding perfectly reflects the theory of planned behaviour (Ajzen 1991(Ajzen , 2002. Kucharska (2017, p. 5) describes a personal brand as "a personal intangible asset defined as a name of a real person combined with all the notions intended to identify this person and to differentiate him/her from others".
With regard to the idea of "prosumerism" by Toffler (1981), it can be hypothesized that there is a certain degree of probability that brands present in social media are predominantly a tool for the self-presentation of the network users.
Recent research found personal identity to be developed in a social environment where individuals have the opportunity to be recognized in a context of reciprocal relationships with other users (e.g. Brooks and Anumudu 2016). Social media represent a fruitful environment for individuals to develop their own personal brand (e.g. Labrecque et al. 2011;Karaduman 2013 andOrehek andHuman 2017).
This can be analysed under a UGT perspective, which helps in the understanding of audiences seen as active media users who choose media, in this case, social networks based on different needs (Wang et al. 2012). This theoretical perspective is useful when-as for social media context-a user-oriented approach is required to understand the production and consumption of social media which are most of the time user-driven (Shao 2009). Finally, gratification also refers to how social media can satisfy personal needs (Quan-Haase and Young 2014).
Therefore, social media enable users' self-expression and self-presentation and, thanks to the strength of community relationships and identification with other consumers and with brands, can help them to build strong personal brands. It is assumed that personal brand is a crucial consumer performance, which derives as an outcome and a benefit of social network participation. If so, it is also the critical company's driver for cultivating consumers' engagement in online brand communities. All of the above led to the following hypothesis.
H12 Stronger community identification positively influences users' personal brands in reference to a particular social group.

Control variables
In addition to the main variables of the model, this research takes into account important control variables (CVs).
CVs allow including extraneous variables in the model. They are not the focal point of the study, but remain theoretically important (Kish 1959;Becker et al. 2016;Nielsen and Raswant 2018).
A global web index survey noticed that online social media usage tends to be driven by national and cultural factors (Alalwan et al. 2017;Nitish et al. 2012;Smith 2010). Additionally, industry insights have an influence on communication in social media (Nitish et al. 2012;Alalwan et al. 2017;Gutiérrez-Cillán et al. 2017). Thus, in reference to the phenomena of interest, and in light of the literature, this research project incorporated controls such as "nationality" and "industry". On the one hand, national cultures may determine how "personal brand" as a phenomenon can be perceived and shaped by social network users. On the other hand, following Kozinets (2020, p. 281), "electrified, digitized, desire-magnifying media unleash global and local imaginariums, liquefying the localities of culture and turning them into intermingling streams". Therefore, intentionally "nationality" has been imputed only as CVs instead of gathering separated national samples to compare them. Moreover, each particular industry has a set of characteristics that create its image, which can also have an influence on brands' image and loyalty. Including them in the model will allow us to gain relevant knowledge-mainly, to control how "nationality" affects "personal brand" and how "industry" affects "behavioural loyalty". Becker et al. (2016) suggest including control variables in hypotheses. Thus, we propose the following: HCV 1 Nationality significantly influences "personal brand".
HCV 2 Industry significantly influences behavioural loyalty.
The model presented in Fig. 1 was developed for the empirical study and was based on the above theoretical assumptions including constructs such as personal branding, self-expression, brand distinctiveness, hedonic and utility functions of a particular brand SNS, social interaction, CsnBI, consumer-other consumers and community identification, behavioural and attitudinal loyalty and control variables.

Method
To select the criteria to collect the data, this study took into account previous studies that pointed out that on the one hand, social media is perceived as a personalized medium of communication but is at the same time as international as any other previous medium (Okazaki and Taylor 2013;Zhang and Vos 2014). They stressed that for developing a social media body of knowledge, effective usage and crosscultural studies are very welcome. Hudson et al. (2016) proved that cultural differences influence social media usage. Due to this inspiration, the Italian and Polish populations of young social media users have been selected by convenience and interest. In addition, European culture is very coherent in many aspects, but it also varies. It was fascinating to check how personal brand and loyalty are perceived by Italian and Polish society just due to scientific curiosity, as many researchers do (e.g. Heikamp et al. 2014or Laconi et al. 2018). This sampling method has been chosen to avoid too small samples or too low a level of user expertise of the study subject.
Thus, bearing in mind all of the above, the study was based on the data originally collected from mainly young users of the social networking service Facebook through a questionnaire. Young users (aged 18-34) is a group more likely to use Facebook (Statista 2020), which is why the sampling process focussed on people in this age group. The Facebook platform was chosen because of its widespread use.
Among social media, Facebook is by far the most popular, with 2.05 billion monthly active users. In August 2017, the other most used social network platforms worldwide were YouTube (1.5 billion) and instant messaging application WhatsApp (1.2 billion). Boasso and Saracino (2016) stated that Facebook is the most powerful marketing platform that has ever existed because users (and therefore also using Facebook consumer brands) have the power to participate, contribute and share their opinions and their own content online.
The detailed sources of the measurement scales used in the questionnaire are presented in "Appendix 1". The respondents reacted to statements based on a 7-point Likert scale, from 1 (definitely not), through 4 (neither yes nor not) to 7 (definitely yes). In particular, respondents were requested to answer concerning eleven constructs with related items regarding three main topics: (1) relationship-oriented, self-oriented and brand contentdriven motives to follow a brand in social media; these motives were measured with five constructs related to social interaction (adapted from Jahn and Kunz 2012), Brand performance/ outcomes Self-oriented motives The statements and related sources are presented in "Appendix 1". The questionnaire was preceded by a short introduction that explained the purpose and subject matter of the study. To be sure, those respondents have the knowledge required to give the valuable answers the questionnaire started from three qualification questions. The first qualifying question referred directly to the subject matter of the study, asking whether the respondent had any affiliation with an SNS-bc on Facebook. If respondents were members of more than one, then they were asked to choose the one for which they wanted to respond. In the second step, the respondent was asked to name the chosen brand; in the third step, respondents named the sector of the selected brand community. Summarizing, the "top of mind" method of brands and sectors was selected for the survey. The affiliation with the particular brand community (and industry) was stated by respondents through the answer to an open-ended question. This approach to asking enabled identifying a "top of mind" choice. It has been assumed that "top of mind" provides the brands with the strongest affective commitment.
The questionnaire began with general questions and then moved to detailed questions that required answers that were more precise. The study also included control variables (CVs), which were measured by using a nominal scale: "nationality" (codified: Italian-1, Polish-2) and "industry" (codified: 1-6) (see "Appendix 2" for further details). They were imputed into the model as dummy variables.
The research study was preceded by a pilot test that was conducted with 38 people, making it possible to eliminate or improve any statements that respondents seemed to find difficult to understand (Hair et al. 2010). Intentionally, in the questions about the personal brand, social media context was stressed to ensure that we captured it when measuring this phenomenon.
Data collection took place electronically, mainly through the social networking portal Facebook using a targeting function of postpromotion. A convenience method of sampling was used because it was felt to reduce the risk of the sample being unrepresentative because of the low frequency and small sample size. The data were collected from November 2016 to March 2017. There were no missing data due to the "required completion answer" constraint; however, some incomplete cases were noted. All participants signed an informed consent form before completing the questionnaire. Thus, the gathered sample size was 815 cases, reduced to 712 cases after invalid (too low SD) or incomplete questionnaire elimination (see "Appendix 2"). The data analysis was conducted through the structural equation modelling approach after a positive normality sampling assessment. The Kaiser-Meyer-Olkin (KMO) test and Bartlett's test of sphericity were provided to evaluate the factorability to apply to the model. The KMO measure of sampling adequacy was 0.933, and the significance of Bartlett's test of sphericity was less than 0.001; these results could be assessed as very good (Kaiser 1974). The cumulative percentage of total variances extracted by factors was 81%, which was positive (Hinkin 1998;Hair et al. 2010).
For the theoretical model presented in Fig. 1, a measurement model and, later, a structural confirmatory factor analysis (CFA) model were developed to ensure that the scales performed appropriately. The evaluation of the model quality was initially conducted based on construct measurements consistency tests such as the average of variance extracted (AVE), composite reliability (CR) and Cronbach's alpha. AVE exceeded 0.53 for all constructs, which was acceptable. Hair et al. (2010) and Fornell and Larcker (1981) suggested that an AVE of 0.5 or higher indicates adequate convergence of the used scales. Cronbach's alpha test was used to confirm the consistency of the construct measurement model. The alpha coefficient was greater than 0.77 for all constructs, which was adequate (Bagozzi and Yi 1988;Hair et al. 2017, pp. 112). The CR was greater than 0.77 for all loadings, which was more than the required minimum of 0.7 (Hair et al. 2010), indicating internal consistency. "Appendix 1" presents more details connected with used scales and their reliabilities. For satisfactory discriminant validity, the square root of each construct's AVE exceeded the correction between any pair of distinct constructs. The results supported the discriminant validity of the measurement model. Table 1 presents further details.
The model was then estimated and assessed. Estimation was performed via the maximum likelihood method (ML). The evaluation of the model quality was conducted based on tests such as root mean square error of approximation (RMSEA), the minimum discrepancy (CMIN/DF), comparative fit index (CFI) and Hoelter's sample size test, with the use of SPSS AMOS 23 software. Table 2 presents a comparison of the obtained results for two models: Model A, run with CVs, and Model B, run without CVs (Aguinis and Vandenberg 2014). In both cases, the general results were similar, but Model B fit the data better. Including CVs generally reduces the statistical power (Carlson and Wu 2012). The results of both models are presented and discussed following the procedure suggested by Becker et al. (2016). Table 2 includes all of the results of tests applied in the evaluation of the presented model, together with their reference values and sources. Based on these results, the model was considered a good fit in relation to the data. A model reliability level of 3.74 can be viewed as positive, with the reference ≤ 5. Based on the approximation average error RMSEA, the model fit to the data, at 0.062, also met the reference values (Steiger and Lind 1980;Byrne 2016). Measurements of the goodness of fit were close to 1, which confirmed the above-mentioned quality. Hoelter's coefficient exceeded 200, which also corroborated the above statements (Hoelter 1983). Figure 2 shows the results of the general model estimation, with the statistical significance (p value < 0.001) of all imputed CVs (Bernerth and Aguinis 2016;Carlson and Wu 2012). Table 2 presents a comparison of the results for two models: Model A, run with CV usage, and Model B, run without CV usage. The general results, namely the models' quality and hypothesis verification, were similar, and all hypotheses for both models have been confirmed except H7 about the positive "utility" influence on CsnBI. The inclusion of CV (Model A) strengthened the community identification (CI) on behavioural brand loyalty effect and weakened the effect of CI on personal branding. According to the brand-oriented motives, the most significant direct effect on CsnBI has been noted for the "hedonic" function of the particular brand SNS, although the path coefficient for this function was low and nearly exceeded 0,1, which, according to Hair et al. (2010), is unremarkable. Table 2 presents a verification of the hypotheses based on their measurements for Model A, run with CVs, and Model B, run without CVs (Becker et al. 2016).

Results
The goodness of fit of the models was comparable in both cases. The inclusion of CVs led to the strongest path coefficients between community identification, behavioural loyalty and personal brand for Model B rather than for Model A. The differences between the models were negligible (Becker et al. 2016), but including the CVs increased the level of statistical significance of the brand distinctiveness effect on CsnBI.
To deeply analyse the findings, this research verified squared multiple correlations (R2) for all dependent variables. R2 value summarizes the proportion of variance in a dependent variable explainable by the collective set of the predictors. Based on that set, the CsnBI construct is explained in this model in 90%, "community identification" in 63%, "personal brand" construct in 64%, "attitudinal loyalty" in 40% and "behavioural loyalty" in 20%. In other words, the presented structure of relations does not explain "attitudinal loyalty" or "behavioural loyalty" and "personal brand", suggesting that, regarding the proposed model, the personal brand of the user is the strongest output of community identification rather than loyalty towards a commercial brand. Moreover, the 90% level of R2 achieved for CsnBi suggests that consumer social network brand identification is perfectly explained by the set of selected predictors of the SNS-bc: social interaction, self-expression, brand distinctiveness and hedonic function.

Discussion and implications
This study addresses several gaps providing both theoretical and managerial contributions. Considering the goals of the research, three main contributions can be provided.

Motives
The first aim of this research was to understand the effect of both relationship-driven and brand content-driven factors that lead consumers to feel strongly connected with other consumers of an online community in the social media context (i.e. consumer-other consumers identification) and to deeply identify themselves with a brand (i.e. consumer-brand identification).
We found that the main motives influencing consumer-other consumers identification were social interaction and self-expression. While both of these motives affected the ties with other consumers, the effect of social interaction was less strong than for self-expression. That is, self-expression was the main motive to feel connected with other consumers in an online community, and this result supports the existing body of knowledge regarding the drivers of community identification, which highlighted the importance of selfexpression in social media (e.g. Orehek and Human 2017;Teichmann et al. 2015).
Interestingly, in contrast to existing findings of the previous research (see, for instance, Stokburger-Sauer et al. 2012), the influences of brand content-oriented motives such as "brand distinctiveness" were found to be statistically Consumers' performance Brand performance/outcomes Self-oriented motives Fig. 2 SEM model, hypothesis verification. ns-not significant, *p < 0.05, **p < 0.01, ***p < 0.001 significant on the lower than others factors level. This might indicate that in a virtual world, the distinctiveness of a brand is not as important as personal and relational motives such as self-expression or social interaction. Similarly, hedonic functions influenced consumer-brand identification significantly but weakly. To summarize, the leading motive to enhance CsnBI was the users' self-expression, which belongs to selforiented gratification. Hence, the first theoretical contribution of this study is to extend the body of knowledge regarding the main drivers and motives that lead consumers to join social media, to have relationships with consumers and the brand. The most important ones are mainly related to selfexpression needs and relationship-driven motives rather than brand-related content or utilitarian motives. This extends the previous research adopting the UGT framework as this research sheds more light on the motives that lead users to choose and actively use social media (Claffey and Brady 2017;Jahn and Kunz 2012). In detail, this study confirms that providing an online context that allows consumers to build their personal brand and to exhibit their presence is a remarkable gratification that leads users to identify with an online community and feel connected with its members and the hosting brand.

Community relationships and identification
The second contribution of the study is to explore the different effects of consumer-other consumers identification compared to consumer-brand identification on the overall brand community identification. This contributes to the brand relationship and brand community management literature in an important way. Despite the three central relationships that influence consumer behaviour in brand communities, namely customer-to-brand, customer-to-customer and customer-to-the community (Chang et al. 2020), most of the studies have focused on the former with a lack of attention on the customer-to-customer relationship .
This study investigated community identification and the two sides of its building process: one related to the customer-to-brand relationship and one related to the consumers being part of the community (customer-tocustomer relationship). The results found that the effect of the consumer-other consumers aspect in brand community identification is higher than the effect of consumer-brand identification. In other words, social factors are crucial for transforming a brand audience into a brand community on the social network, rather than the user-brand relationship. The previous research pointed out that both consumer-brand relationships and consumer-other consumers relationships influence the identification of the consumer within a brand community online (Hsu et al. 2015;Luo et al. 2016). However, this research-combining in the same study, the two sides of community identification-demonstrated that the consumer-other consumers identification has a more substantial effect on brand loyalty outcomes than the direct consumer-brand identification does. These findings empirically demonstrated and emphasized that the social relationships between users were the essence of the social media's existence and success and provided justification for online brand community managers to create a breeding ground that foster consumer-driven conversations and social interactions.
Such theoretical contribution addresses managerial activities, suggesting the community manager (the person who moderates the SNS-bc) to create an idea that justifies the feeling of being committed, integrated and cooperating with the other participants of the community to make this network space the place for consumer interaction. Özbölük and Dursun (2017) proved that although the brand community members unite around a common goal, they are different in their interest and commitment to the brand and the community, they differ in the roles they play, and they do not constitute a community, which shares a common commitment.

Attitudinal and behavioural loyalty in social media
Third, this study found that the role of the community in both brand loyalty and personal brand creation is significant. Considering company performances, which is brand loyalty, the study found a deficit of behavioural loyalty in relation to attitudinal loyalty, suggesting that social media create stronger brand equity (measuring the state of mind) than brand value (measuring final purchasing). Namely, the hedonic and symbolic character of a brand is more meaningful than functional for consumer behaviour, especially due to the positive association between narcissism and social media usage, visible through a customer's personal brand expectations.
In the case of the brand loyalty output expected by a hosting company (Dessart et al. 2015), the findings of this research contribute to the theoretical debate regarding the relationship between social media and brand loyalty creation (Coelho et al. 2018), suggesting that community identification strongly affects attitudinal loyalty rather than behavioural loyalty. There are several reasons for this result. First, there is probably some distance between the social network as a communication channel and the selling channels-the majority of goods selling takes place in reality. Second, to build one's image with the use of a brand in the real world, an individual has to own a branded product; that is, an individual needs to buy a branded product to achieve the visible benefit from the particular brand's image. Possession and use are key factors in the process of self-identification with the use of the brand image (Belk 1988(Belk , 2013. In a virtual world, possession of a branded product is not necessary for the successful use of its image. Thus, it is reasonable to envisage that in this context, the symbolic character of a brand is more meaningful than functional for consumer behaviour, as supported by the findings of this research. This investigation is the first research to distinguish between the effects of community identification on attitudinal loyalty and behavioural loyalty in the social media context. Another theoretical contribution of this study is not only within the debate of loyalty creation via social media but also about how and why such outcome is generated through the gratifications and benefits deriving from social media usage. This is related to the previous research applying UGT where several studies have asked for a more in-depth analysis of how the active usage of social media and derived benefits can generate brand firm positive outcomes (Corrada et al. 2020;Sundar and Limperos 2013). This study enriches the previous research about how active behaviour is helpful in building purchase intention and repurchase intention and, in particular, focuses on loyalty creation via the active usage of social media and the identification users have with brand communities, confirming the validity of UGT paradigm applied to social media.

Personal brand in social media
One of the most interesting and up-to-date contributions of this study is the more potent effect community identification has on consumers' personal brand compared to brands' performance (loyalty). While these two relationships have been separately investigated in the previous literature, this study is the first to compare them in one research and to find that high community identification is a significant driver for personal brand building. Such results extend previous studies which pointed out that self-concept is fundamental to customers' relations with brands and other customers in social media (e.g. Hwang and Kandampully 2012;Labrecque et al. 2011).
In addition, this is consistent with UGT paradigm and extends previous studies applying it. This paradigm assumes that users are goal-directed in their behaviour and are aware of their needs. If applied to social media, the key drivers and pursued needs are purposive value, self-discovery, entertainment value, social enhancement and maintaining interpersonal connectivity (Cheung et al. 2011).
The previous research on social media adopting UGT found that brand/firms gaining positive company performances such as an increase of "likes" or high purchase intention is a result of an active consumers' usage of social media and the perceived benefits/gratifications derived from its usage (e.g. Sundar and Limperos 2013;Corrada et al. 2020). What this research found and extended from the body of knowledge adopting UGT is that personal brand creation is an outcome from the gratification and usage of this media, which is rarely obtained from other media usage. In particular, for social media compared to other media, the active role of users leads to co-creation for brands but also for the self, that is, for the user who actively joins and adopt online brand pages.
This study showed that one of the vital users' interests is their own personal brand creation. In other words, considering that an excellent personal brand reputation is the desired effect of community identification, the deliberate creation of a personal brand's reputation by social media users determines the values created by the community that is hosting the social media. This finding is consistent with the previous research, which found a positive association between narcissism and social media usage (e.g. Carpenter 2012). Hence, this study is also relevant and contributes to the open debate related to personal branding literature and how personal brand building is a key driver for social media success (Kucharska 2017). Hence, the opportunity of building personal branding for users is a very important benefit and a result that can be translated into an important asset for brands too, as a result of the customer engagement and the active attachment to online brand communities.
Based on that, while companies' main benefit from using social media is the sustainability of relationships with customers to create loyalty (De Vries and Carlson 2014), customers expected performance is to create their own strong personal brand and increase the social presence.
Based on the full picture of consumer-community and brand relation presented in the empirical model, this study suggests what social media managers can do to engage user's most into brand's communities: to support users to build and show their personal brands and social presence in brand pages.
As aimed, the present investigation provided many meaningful insights that help to understand the mechanism and, subsequently, to act reasonably. Key managerial implications are summarized as follows: • Companies are very often too focussed on their own performances, forgetting about customers' aims. The present study proved that those consumers' personal brands, rather than only company brands, are focal points of each community. The only means of creating loyalty towards brands is to allow consumers to build and show their personal brands via brand social pages. Based on the UGT, companies should arrange ad hoc social media pages to allow consumers to express themselves and their image to other customers via different tools, from images and video sharing, gamification and other self-representation tools. • "Industry" significantly influences behavioural loyalty.
In other words, companies should tailor their actions very thoroughly, bearing in mind particular brand and industry characteristics. It may not be the best option for community success to pay the external agency for adopt-ing the same tools for all their clients (from different sectors) without a deep understanding of the essence of consumer-brand relations in the particular industry. For example, being very sensitive, the financial industry is probably not as good as the fashion industry at creating the user's brands (show up). We mentioned fashion. The findings as presented suggest fashion are overused by consumers to create their image and do not lead to behavioural loyalty. This insight is meaningful. The authors hope to inspire companies and other studies to investigate this industry in more depth to find the method of how to adapt to this situation and transform it. • Personal brand and "nationality". The presented findings proved that there is a different perception of personal branding feeling, taking into account national cultures.
International companies should take this perception into account when creating their tools.

Limitations and future research
While this research has presented a significant contribution to the understanding of brand community identification drivers and outputs, it has limitations, mainly concerned with methodological issues. First, because of the convenience sampling, a majority of the study participants were aged between 18 and 34. Thus, future research should include an extended age range of respondents. Moreover, the sample represented only two European populations: Italian and Polish. To have a full picture of the European population, representation of other nations would be required. Moreover, not only the nationality but also probably the age, occupation, gender, "the moment of life" and "personal aims", summarizing not only demography but also psychographic characteristics, will influence effective SNS-bc management, further encouraging more in-depth investigations. Another issue related to the sample selection is in relation to respondents who were social media users who had joined some brand pages on social media. This virtual relationship with the brand did not necessarily reflect a "real" consumption of the brand products in real life; there could be a gap between users' identification with the brand community and their actual behaviour in real life. Therefore, future research should link the virtual world with the real world.
Another limitation is related to the social network adopted for the study, that is, Facebook. The results only consider benefits participants perceive from following and being fans of brands within this platform. It would be interesting for future research to compare results with those obtained from other social network platforms, such as Twitter, Instagram or others. Regarding personal branding, this research explores the role community identification has on building personal branding. However, more research is needed to understand the antecedents that help users to build their personal brand via social media presence.

Conclusion
This study has been the first to relate community identification with both attitudinal and behavioural loyalty. In addition, it links community identification with personal branding, which can be considered one potential outcome expected by users from social media usage. The central value of this study is the full picture of consumers' motives to identify with the community and the outcomes of this identification. Based on all the presented relation structures introduced in the study, the findings suggest that social media managers should engage the users in communities to support their personal brands, possibly also helping brands to strengthen brand loyalty.
Funding Open access funding provided by Università degli Studi di Verona within the CRUI-CARE Agreement. Funding was provided by Fondazione degli Studi Universitari di Vicenza (Grant No. BIT-Business Innovation and Digital Trasformation @ Vicenza).

Compliance with ethical standards
Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.