Introduction

The ever-evolving social media platforms have become powerful tools to build and maintain customer relationships and brand loyalty. Most brands nowadays operate accounts on multiple social media platforms, such as Facebook, Instagram, YouTube, Twitter, and Snapchat, to reach different segments of customers and engage with customers in real time. As of December 2020, Nike, one of the top athletic brands in the world, has more than 124 million followers on Instagram (https://www.instagram.com/nike/). Consider the number of people a post on Nike’s Instagram account can reach, which is more than the total population in Germany (83 million people), France (67 million), or the UK (66 million) (World Bank 2021). In recent years, US adults spent an average of approximately 80 min per day on social platforms (eMarketer.com 2020). Therefore, marketers have been competing on social media platforms to cultivate customer relationships so that they can attract, convert, and retain customers.

Numerous brand studies have suggested that brand community is an effective way to deliver a better customer experience (Mascarenhas et al. 2006) and build brand loyalty (Erdoğmuş and Cicek 2012; Habibi et al. 2016; Fernandes and Moreira 2019). The concept of brand community has rapidly evolved in online platforms and brought unique aspects of brand loyalty established through social media (Habibi et al. 2016). A brand community is defined as a “specialized, non-geographically bound community, based on a structured set of social relations among admirers of a brand” (Muniz and O'Guinn 2001, p. 412). McAlexander et al. (2002) argued that brand community is “the holy grail of brand loyalty” (p.38). Therefore, since the development of the Internet, more and more online brand communities were created to facilitate the business-to-customer (B2C) and customer-to-customer (C2C) relationships, aiming to build brand loyalty.

Consumer brand loyalty has been one of the most studied areas in the brand management field. Jacoby and Chesnut (1978) conceptually defined that consumer brand loyalty should not be limited to repeated purchase behavior. Instead, the belief, affection, and intentional dimensions of consumer loyalty must be examined to grasp all aspects of the concept. Later, Oliver (1999) suggested brand loyalty as a “deeply held commitment to re-buy or re-patronize a preferred product or service consistently in the future, despite situational influences and marketing efforts having the potential to cause switching behavior” (p. 34). Additionally, numerous brand studies have suggested that brand community is an effective way to deliver a better customer experience (Mascarenhas et al. 2006) and build brand loyalty (Erdoğmuş and Cicek 2012; Habibi et al. 2016; Fernandes and Moreira 2019).

In the past two decades, along with the development of social media, researchers have been conducting empirical research on how to build customer–brand relationship through e-loyalty (Gommans et al. 2001) on social media (Laroche et al. 2013), among Gen Y consumers (Lazarevic 2012), and in emerging markets (Nguyen et al. 2011). Recently, scholars have argued that consumer engagement plays an important role in building consumer loyalty on social media (Helme-Guizon and Magnoni 2019; Li et al. 2020). Helme-Guizon and Magnoni (2019) stated that consumer-brand engagement is a key component in inducing brand loyalty through social media platforms. Consumer–consumer interactions, a similar concept of brand community, have substantial impacts on brand trust and affection. Similarly, Li et al. (2020) scrutinized the relationships between customer engagement, brand attachment, and brand loyalty. They indicated that customer engagement does not directly influence brand loyalty; the critical first step is to build strong emotional connections and attachments in developing customer–brand relationships. The aforementioned studies synthesized the findings of customer–brand relationships, the types of brand community participation, and the motivations and consequences of participating in a brand community.

Brand management scholars also conducted review studies to synthesize and aggregate the findings from these empirical studies. For instance, Khamitov et al. (2019) conducted a meta-analysis on brand relationships (i.e., brand attachment, brand love, self-brand connection, brand identification, and brand trust) and their impacts on brand loyalty. Hook et al. (2018) examined the antecedents and consequences in studies of consumer participation in brand communities. Although these review studies provided detailed antecedents and outcomes of brand community, they were not focused on online social media platforms. There is still a lack of framework-based systematic review focusing on the scholarship of online brand communities, specifically concerning how to build consumer loyalty on social media platforms.

Without a systematic review to answer who, where, what, and how scholarship was established concerning brand loyalty on social media platforms, it is impossible to establish a comprehensive theoretical framework and inform future research directions (Paul et al. 2021). Therefore, the objectives of this study are twofold. Our first objective is to conduct a field review of literature on brand management to map out the scholarship on building consumer loyalty via social media platforms. Our second objective is to identify possible research gaps/issues in the current literature and suggest future research directions in brand loyalty, brand community, and social media marketing.

Specifically, there are five research questions we intend to answer in this study: (1) what is the magnitude of the research in the use of social media to build brand loyalty (amount, journals, years, geographic areas, etc.); (2) what theories are utilized/proposed to explain the nature and process of using social media to build brand loyalty; (3) what are the antecedents and consequences of brand loyalty on social media; (4) how was brand loyalty conceptualized and measured in the literature; (5) what research gaps and possible future research directions exist? We then describe the research methodology, the publication activities, the theory themes applied in the empirical studies, the antecedents and consequences tested in the research models, and the measurements of brand loyalty. Next, we present the issues identified in this review and offer directions for future research. Finally, we discuss the limitations of the present study and offer our conclusion.

Methodology

Systematic review process

A systematic literature review has been suggested as a scientific method of secondary research to inform and strengthen the theoretical foundation of scholarship (Childs 2017; Paul and Criado 2020). Moreover, scholars have suggested that systematic literature reviews can help to integrate extant knowledge to provide a state-of-the-art understanding, to identify the extant knowledge gaps, and help to find future research directions (Hulland and Houston 2020; Paul and Criado 2020). Moher et al. (2009) suggested a four-stage protocol (i.e., identification, screening, eligibility, and inclusion) to solicit articles for systematic reviews and meta-analyses. Following this procedure, we collected the articles using two major sources: library databases and Google Scholar search. Figure 1 illustrates the systematic review method used in this study. The article identification process is from February to April 2020. First, we collected the articles by using keywords such as “loyalty,” “social media” or “social network,” “consumer relationship” or “customer relationship,” and “brand” from databases such as EBSCO, PsyInfo, Hospitality and Tourism Complete, Emerald Journals, and ABI/Inform. We narrowed the publication years from 2005 to 2020, since we would like to focus on more recent scholarship developments in this review. The publication type was set to “all” because we wished to include all types of the publication for the purpose of inclusivity. The search resulted in 404 papers in the initial list. To satisfy our screening criteria, articles must: (1) be written in English, (2) have full-text access, (3) not be duplicates, (4) be conducted in the context of social media platforms, (5) have measured brand loyalty in consumer levels, and (6) have used multiple items to measure brand loyalty. After the first screening, 29 articles remained for further analysis.

Fig. 1
figure 1

Systematic literature review flowchart

Full size image

Next, we performed a search on Google Scholar using the keywords “loyalty” and "social media." Google Scholar was chosen for the following reasons. First, it is the world largest search engine for scholarly works (Gusenbauer 2019). Second, it is freely accessible to the public and included various types of scholarly works such as proceedings, theses/dissertations, and patents (Lim and Weissmann 2021). We downloaded the search results from Google Scholar using bibliography research software named “Publish or Perish” (https://harzing.com/resources/publish-or-perish/windows). The researchers found that using the Published or Perish software to download Google Scholar search results had two significant benefits on paper collection. First, the list included a column of the order of the Google Scholar search results, ranked in Google’s algorithm (GSrank) by the article’s relevance to the keywords. The second benefit was that the search results included the citation numbers (how many times the paper was cited) of each paper. Therefore, researchers could use the number of citations to confirm that no widely cited/well-known articles were missing from the data collection (Harzing 2020). The current research included articles with at least one citation per year (citation/years since published = 1).

Using the aforementioned criteria, we screened the top GSrank 150 articles (equivalent to 15 pages of Google Scholar search results), resulting in 62 articles that passed the same screening criteria used for the database papers. Among the 62 articles, five papers overlapped with the articles collected from the databases. After the screening stage, a total of 86 papers remained for further analysis.

In the eligibility stage, all the 86 papers were examined for a second time to verify the eligibility of the papers for the following criteria: (1) assessed one or more brand relationships utilizing a multi-item scale, (2) included a measurement identified as customer brand loyalty, (3) entailed empirical consumer-level reactions, and (4) conducted in the context of social media platforms.

In the final inclusion stage, the researchers conducted a content analysis to extract information from these empirical papers. We adopted the integrated approach to examine the scholarship in the domain of brand loyalty on social media (Lim and Weissmann 2021). Paul et al. (2017) introduced the Theories, Context, and Methods (TCM) framework to organize the findings of review results, whereas Paul and Benito (2018) introduced the Antecedents, Decisions, and Outcomes (ADO) framework to organize the concepts involved in a knowledge domain. That is, the articles were coded using a coding scheme guided by the TCM and ADO frameworks. The researcher extracted: (1) bibliographic information (e.g., year, types of journals, citation information), (2) research context and methodology information (e.g., sample size, social platforms, participant characteristics, data collection methods), (3) theories, antecedents and consequences (outcomes) of brand loyalty, and (4) the measurements of the focal decision (i.e., brand loyalty). Each paper was reviewed by two of the three researchers. If there were a coding discrepancy, the third researcher would review the article and discuss with the two coders in the research meetings to reach a consensus (Braun and Clarke 2006). Utilizing this integrated approach, the TCM framework answers the research questions: (1) what is the magnitude of the research in using social media to build brand loyalty? (2) what theories were utilized/proposed to explain the nature and process of using social media to build brand loyalty? The ADO framework guided this review to answer what the antecedents and consequences of the brand loyalty are, and how brand loyalty was conceptualized and measured. Figure 4 illustrates an overview of the integrated TCM and ADO frameworks for systematic review of building brand loyalty on social media. In the following sections, we report the outcomes of the content analysis by this systematic review study. All the articles used in the final review are included in the reference list.

Publication activity

The 86 empirical research papers were published from the year 2009 to April 2020 (Fig. 2) in 67 unique outlets, including peer-reviewed journals (N = 78), conference proceedings (N = 4), master thesis (N = 1), and Social Science Research Network (SSRN, N = 1) (see a list of publication outlets in “Appendix A”). The journal types include marketing (N = 15), management (N = 14), e-commerce (N = 8), retail (N = 3), hospitality, tourism, and sports (N = 7), and others (N = 13) (Fig. 3). The various publication outlets in the literature indicate that there is widespread interest in this research stream (Fig. 4).

Fig. 2
figure 2

Publication amount from 2009 to 2020

Full size image
Fig. 3
figure 3

Frequency of occurrence of articles by journal types (please see detail journal list in “Appendix A”)

Full size image
Fig. 4
figure 4

Integrated TCM and ADO frameworks for systematic review of brand loyalty on social media

Full size image

Altogether, there were 33,400 participants surveyed in these 86 papers. The majority of the papers were from Asia (44%), Europe (20%), and the USA (15%). Ten percent of the studies collected data from more than one area, and 11% were from other regions of the world. Two papers did not report the location where data were collected (Table 1). Regarding the participants’ characteristics, 76% of the research used non-student samples, while 20% of the research collected data from the student samples. The rest were either from Amazon Mturk or from a mixed sample.

Table 1 The geographic areas and sample characteristics
Full size table

In order to evaluate the studies that had more impacts on influencing this stream of study, a citation analysis approach was used (Kim and McMillan 2008). By examining the top-cited papers, we can identify the key sources that influence the field of study. Among the twenty most cited articles, four of them are from Computers in Human Behavior, three are from the International Journal of Information Management, two are from Journal of Business Research, and two are from Procedia-Social and Behavioral Sciences. The most cited paper was published in the International Journal of Information Management and had a total of 788 citations and the highest cites/year rate (112.57 per year) as well (Mosavi and Kenarehfard 2013). This result suggested that although there is a widespread interest in building brand loyalty on social media, there are limited journal outlets that publish highly cited papers in this research field.

Results

What theories are utilized/proposed to explain the nature and process of using social media to build brand loyalty?

There are five major theory themes discovered in the literature review: (1) consumer self-identity theories, 2) brand community-related theories, 3) theories related to consumer decision making, 4) theories that focused on social media platforms as a communication medium, and 5) relationship-based theories. Below we will discuss the five major theory themes in more detail. Table 2 provides a list of the five theory themes and the articles that applied those theories.

Table 2 Major theory themes applied in the brand loyalty literature
Full size table

Consumer self-identity theories

This group of theories describes the processes and consumers’ motivations when engaging in a B2C relationship. There are 13 papers (out of the 86 papers) applying self-identity theories as the theoretical framework in their studies. Social identity theory (Tajfel and Turner 1979, 1986; Dutton and Dukerich 1991), self-perception theory (Bem 1972), self-expansion theory (Aron et al. 2005), symbolic self-completion theory (Wicklund and Gollwitzer 1981), and congruity theory (Lee and Jeong 2014) were utilized to portray how social media may facilitate consumer loyalty through the process of self-identification, recognizing the brand images to enhance self-identity, and social comparison.

Relationship-based theories

This group of theories focuses on the nature of B2C (business to customer) relationships and relationship marketing. Twelve papers have utilized this stream of theories to explain why social media can be utilized to build brand loyalty. There are also some relationship marketing theories named: interpersonal-relationship-marketing theory, relationship-marketing theory (Grönroos 2009), and commitment-trust theory of relationship marketing (Morgan and Hunt 1994) (Table 2). The theories that describe the nature of the relationship include social exchange theory (Blau, 1964), relationship quality theory (Crosby et al. 1990), parasocial interaction theory (Horton and Wohl 1956), contagion theory (Burt 1987), expectation confirmation theory (Oliver 1980), promising theory (Calonius 2006), gift-giving theory (Sherry 1983), and construal level theory (Liberman and Trope 1998).

Consumer decision-making theories

Several consumer decision-making theories were also applied in this stream of research, including the consumer decision-making model (Ajzen and Fishbein 1980), the theory of planned behavior (Ajzen 1991), the theory of reasoned action (Eid 2011), the technology acceptance model (Davis 1989), the social–psychological model of goal-directed behavior (Perugini and Bagozzi 2001), the mean-end chain theory (Gutman 1982), the incentive motivation theory (Ellingsen and Johannesson 2008), and consumer-style inventory, which describe how different types of consumers make purchase decisions (Sprotles and Kendall 1986). There are 10 studies (out of the 86) that applied consumer decision-making theories as a theoretical framework in their studies.

Brand community-related theories

There are six papers that applied brand community-related theories as a theoretical framework in their studies. This group of theories describes the dynamics of customer engagement in the brand community and its influence on brand equity. The theories applied in this group are brand community theory (Habibi et al. 2016), consumer-brand-engagement model (Fernandes and Moreira 2019), brand equity theory (Hanaysha 2016), and brand co-creation (Kamboj et al. 2018).

Communication-related theories

Some researchers viewed social media as an innovative communication medium and investigated how brands and consumers adopted social media and use it for communication. The theories applied in this theme include the stimulus-organism-response model (Mehrabian and Russell 1974; Jacoby 2002), signaling theory (Taj 2016; Kirmani and Rao 2000), media richness theory (Daft and Lengel 1986), and social presence theory (Goffman 2008). Five papers used communication-related theories as a theoretical framework in their studies.

What are the antecedents and consequences of brand loyalty on social media?

After examining all the articles, categories were developed regarding the focal area of each respective antecedent and consequence of brand loyalty in the context of social media. Eight types of antecedents and five groups of outcomes were found and discussed below. Figure 5 illustrates the antecedents and outcomes of brand loyalty tested in the literature.

Fig. 5
figure 5

The antecedents and consequences (outcomes) of brand loyalty

Full size image

Antecedents

Eight broad categories of antecedents were found and were termed in order of overall prominence among the articles: brand characteristics, consumer characteristics, social media characteristics, consumer-brand relationship and engagement, social media marketing activities, social media engagement, online brand community engagement, and perceived value and risks. These categories were decided based on the focus of the constructs and as interpreted by the original author/s provides a summary of previous work on the antecedents of brand loyalty. The findings suggest that previous studies have study antecedents of brand loyalty from the perspectives of the providers of social marketing activities (brands), recipients of social marketing activities (consumers), and the characteristics of the social media marketing itself. These eight categories are interconnected, and each category of antecedents presents useful insights into how to build brand loyalty through social media marketing activities.

Brand characteristics. Previous studies reveal that traits associated with a brand have a strong influence on consumer's loyalty to the brand. Nowadays, businesses have joined social media to increase brand awareness, strengthen brand identity, and acquire more customers. Brand characteristics-related antecedents include brand personality, brand image, brand equity, brand identification, and brand equity dimensions (brand awareness, brand associations, and perceived quality) (see Table 3). In studies investigating the dimensions of brand equity, brand loyalty was treated as both a dimension and an outcome of brand equity (Severi et al. 2014). A few studies have specifically examined the effects of retailers’ reputation on consumer-retailer loyalty (Rapp et al. 2013). The findings of those studies reveal that a favorable brand image or a good reputation can help a company gain a competitive advantage and increase consumer loyalty in the social media context.

Table 3 Antecedents of brand loyalty and their relationship to brand loyalty
Full size table

Consumer characteristics. Today, consumers are involved in various online activities, ranging from consuming content to participating in discussions, sharing knowledge, and contributing to other consumers’ activities. Like brand characteristics, individual differences and consumer characteristics (e.g., personality, need for affect, attachment styles, and motivations) could also influence consumers’ social media use and their relationships with a brand. Consumer-related characteristics examined in the prior studies include brand consciousness, online community participation, and social media usage (see Table 3). Although most of the traits were found to have a positive impact on brand loyalty, some may undermine brand loyalty. For example, Ismail’s study (2017) suggested that social media marketing activities contribute to brand consciousness, and in turn, brand loyalty will exist when the consumer is highly conscious of the brand. However, in the same vein, social media marketing activities may also boost value consciousness, and value consciousness and brand loyalty are negatively related (Ismail 2017).

Social media characteristics. Although branded websites are still the most popular sites for online brand interactions, most online brand interactions now occur on social networking sites rather than branded sites. Social media has changed how brand content is created, distributed, and consumed, transferring the power to shape brand images from marketers to consumers' online connections and content (Tsai and Men 2013). A category unique to brand loyalty in the context of social media is social media characteristics. Social media has become a hub for companies to promote goods and services by providing consumers with relevant and accurate information, and reducing their efforts to search for information (Laroche et al. 2013). This category, in contrast to the others, is concerned with the information and design of the social media sites or pages, including entertainment value, social media communication style, and visual appeal and image (see Table 3). This category is the least frequently used throughout the sample, implying that these social media content-related features are not as important as the other categories of antecedents. However, these features can still influence brand loyalty and should not be ignored.

Consumer–brand relationship and engagement. Branding literature indicated that marketers could enhance their brand loyalty by strengthening consumer–brand relationships (Aaker 1996; Fournier 1998). The most examined antecedent category of brand loyalty was the consumer–brand relationship constructs, such as brand attachment, brand experience, brand trust, social website trust, and customer satisfaction (see Table 3). Many of those constructs are coexisting and influencing one another. Most of the prior research has confirmed a positive relationship between these factors and brand loyalty, suggesting that the customers in a good relationship with a brand are more likely to respond positively to its social media marketing efforts and engage in loyalty behaviors to the brand.

Social media marketing activities. Social media marketing is defined as “a broad category of advertising spending, including advertising using a social network, virtual worlds, user-generated product reviews, blogger endorsement, RSS feeds of content and social news sites, podcasts, games, and consumer-generated advertising” (Tuten 2008). The constructs examined to understand the impact of social media marketing efforts include electronic word of mouth (E-WOM), social media marketing activities, social media communication, and social networking practices (see Table 3). Studies have confirmed that if a consumer responds favorably toward a company’s advertisements and promotions through social media, then a relationship will begin to develop between the consumer and the brand. Consequently, a strong and favorable consumer–brand relationship in social media would lead to brand loyalty (Ismail 2017).

Social media and online brand community engagement. Social media is understood as platforms of virtual communities where individuals with the same interests interact with each other. Therefore, it provides marketers with remarkable opportunities to reach consumers in their social communities and build more personal relationships with them (Kelly et al. 2010). Social media platforms such as Facebook, Twitter, and YouTube have provided a new way for customer engagement and relationship management, and social media engagement has become a vital part of any marketing strategy (Hennig-Thurau et al. 2010). Social media engagement-related antecedents focus on the interaction between customers and the brand in social media platforms, including active/passive social media engagement and social interactions (see Table 3). Studies have confirmed that higher consumer engagement with a brand/company through social media would increase the possibility of more customer interactions and result in higher brand loyalty (Laroche et al. 2012). Digital customer engagement and interaction is a vital part of a business strategy that contributes to the overall brand experience. Oliver (1999) postulates that brand communities and social bonding are the focal forces that drive brand loyalty. Online brand community-related antecedents include brand community on social media, online community commitment, and online community trust (see Table 3). Prior research shows that participation in virtual brand communities and benefits gained from online brand communities have a positive influence on brand commitment and loyalty (Raïes and Gavard-Perret 2011; Zheng et al. 2015).

Perceived value and risks. An individual considers both benefits and costs before making a rational decision about an act. Social media use is believed to have both benefits (i.e., social connectivity, social involvement, information attainment, and entertainment) and risks (i.e., social risk, time, psychological risks, and privacy concern) for consumers. This category of antecedents includes perceived value, perceived usefulness, perceived security, perceived costs, and perceived risk (see Table 3). In terms of values and benefits, social media enable social interaction and information exchange among social network members and assist them in online buying and selling of products and services (Kapoor et al. 2018). Researchers have found that when consumers perceive a value that extends beyond the functional value in their interaction with a brand on social media, the added value is generated, which leads to brand loyalty. Prior research has shown that consumers also associate some costs and risks with social media use and participation. According to the prior research, those risk factors have a negative impact on brand loyalty (Ha and Pan 2018; Zhang and Li 2019).

Outcomes

Customer–brand loyalty is generally considered the ultimate desirable marketing-based outcome from strategic marketing activities (Taylor et al. 2004). Therefore, compared to antecedents, fewer studies have investigated the further outcomes of brand loyalty. Concerning the consequences of brand loyalty in the social media setting, five categories were discovered. These were, in order of prominence: WOM or E-WOM (n = 6), repurchase or repeat-purchase intention (n = 4), purchase intention (n = 2), branding co-creation (n = 1), and brand equity (n = 1) (for details, please refer to Table 4). Among them, branding co-creation is a relatively new concept of marketing, referring to the processes by which a company and its customers jointly participate in value creation (Prahalad and Ramaswamy 2004). Kamboj et al.’s study (2018) has suggested that online brand communities resulted in brand loyalty, which, in turn, lead to co-creation with customers. Other studies have also confirmed that brand loyalty established through social media marketing activities is positively related to future consumer purchase behavior/responses.

Table 4 Consequences of brand loyalty
Full size table

Dual nature variables

Interestingly but not surprisingly, this review has discussed a couple of dual nature variables: E-WOM and brand equity. They were viewed as antecedents of brand loyalty in some studies (Severi et al. 2014; Sánchez-Casado et al. 2018) and were also studied as a consequence of brand loyalty in other studies (Rialti et al. 2017; Severi and Ling 2013) in the context of social media. The same dual nature variables were also observed in Hook et al.’s (2018) review on brand community.

How was brand loyalty conceptualized and measured in the literature?

Among the 86 articles, 27 articles did not provide the measurement items of loyalty used in the paper. The measurement items of loyalty in the rest of 57 articles can be summarized by the following categories: (1) increased interaction with brands, the example question is “intention to get interactive with brands through social media” (Erdoğmuş and Cicek 2012); (2) intention to repurchase, such as “I intend to keep purchasing products from the brand” (Hsieh and Wei 2017; Zhang et al. 2016); (3) purchase intention, which indicates consumers will purchase from the brand in the future (Habibi et al. 2016; Jahn and Kunz 2014; Popp et al. 2016; Rialti et al. 2017); (4) attachment, which refers to the liking or emotional attachment of brands, such as “I would feel emotionally connected toward the brands I follow online (Nisar and Whitehead 2016); (5) positive word-of-mouth, the example is “I recommend this company’s products/services to friends and relatives” (Martins and Patricio, 2018); (6) satisfaction, such as “I am satisfied with X brand with every visit” (Godey et al. 2016); (7) willingness to pay more for brands, such as “I would be willing to pay a premium for this brand over other brands” (Hsieh and Wei 2017); (8) favor the brand over others, the example question is “I usually use this product/brand as my first choice in comparison with the other product/brand” (Severi et al. 2014); (9) Trust toward brands (Erdoğmuş and Cicek 2012; Hossain and Sakib 2016; Qiutong and Rahman 2019); (10) Commitment, such as “I consider myself to be loyal to the brand” (Kamboj et al. 2018; Laroche et al. 2012) and “I am committed to purchasing from the brand” (Frernandes and Moreira 2019; Kamboj et al. 2018; Zheng et al. 2015); (11) Unwillingness to switch the brand, the common theme is consumers will not consider other brands if the brand is not available (Chahal and Rani 2017; Clark and Melancon 2013; Parihar et al. 2019; Severi et al. 2014).

To consolidate these item categories, we applied Oliver’s (1999) four-phase loyalty, as the framework incorporates the temporal dimension of brand loyalty, which is highly relevant in building consumer relationships on social media (Curran et al. 2010). Oliver (1999) defined the two constructs of loyalty: attitudinal and behavioral loyalty. The attitudinal loyalty follows the cognition-affective-conation pattern (Oliver 1999). First, cognitive loyalty is based upon the brand preference because of prior information and experiences with respect to product attributes. In this stage, consumers’ loyalty is largely influenced by their evaluations comparing different brands (Evanschitzky and Wunderlich 2006). Thus, we categorized the loyalty measurements of “Favor the brand over others” and “Trust towards brand” as the cognitive loyalty. Second, affective loyalty indicates the emotional attachment, liking, and satisfaction toward the brand. Thus, we categorized “attachment” and “satisfaction” as the affective loyalty. Third, the conative loyalty refers to behavioral intention, such as purchase and rebuying intention. As such, we listed akin motivations, which included “Increased interaction with brands,” “Intention to repurchase,” “Purchase intention,” “Positive WOM,” and “Commitment.” The last dimension is action loyalty, which indicates behavioral intentions are converted to actual actions (Oliver 1999). In this dimension, consumers will overcome obstacles (i.e., psychological and financial barriers to switch brands) and demonstrate continuous purchasing behavior toward the brand (Han et al. 2011; Oliver 1999). Therefore, we listed “willingness to pay more” and “unwillingness to switch brand” as the action loyalty. Table 5 shows the categories of loyalty measurement and the examples of survey questions.

Table 5 Measurement items of loyalty
Full size table

Identified research gaps and future research directions

The findings of this framework-based systematic review revealed several gaps in the current literature. This section provides a detailed account of the issues that need to be addressed in the existing literature regarding the impact of social media on consumer–brand loyalty. In the following sections, we provide identified gaps and future research questions concerning theoretical development, antecedents and consequences, and measurements of brand loyalty.

Future research directions on theoretical development

Despite the fact that social media platforms have enjoyed explosive growth in the past two decades, there is not a dominant theory used in collected literature. Scholars in the brand management field adopted a great number of theories from disciplines such as psychology, communication, and sociology to untangle the relationships between the consumers, brands, and brand community. Some gaps in the theory building should be addressed by future researchers.

First, different from the traditional brand community, social media communities have unique ways of communication that are drastically different from the traditional brand community. However, there is no specific theory/model emerged to describe or address this nature yet. Therefore, there is a need to develop an overarching theoretical framework that consists of major concepts in the social media community to prescribe the unique characteristics of social media communications. Second, we noticed that all the theories in the brand-community-related theories were published within the past two decades. Therefore, this shows that the scholarship in the field of brand community studies (and theory development) is still in its early development.

To capture the dynamic nature of brand loyalty, it is imperative to recognize the multi-dimensional/phase nature of building consumer brand loyalty. In Fig. 5, we illustrate the dimensions of brand loyalty and its antecedents and consequences as a framework to invite future research for further examination and testing. Based on the aforementioned research gaps, we propose the following research questions.

RQ1

Can we use theory-development-based reviews to examine the development of each theory’s themes in order to develop an overarching theory explaining the dynamics of building consumer loyalty through social-media communities?

RQ2

For the community related theories, what research milestones demonstrate significant progress in the theory development?

RQ3

What dimensions/ phases of brand loyalty contribute the most to the long-term business competitive advantages? Can we use meta-analytical reviews to identify the essential dimensions or phases of consumer loyalty?

Future research directions on antecedents and consequences of brand loyalty

Based on our review of antecedents and consequences of brand loyalty, we have identified several research gaps that need to be filled by conducting further research in this area. First, this review’s findings show that the 86 articles cover the major geographical regions in the world, but over half of these studies focus on Asian and European contexts, and only a few studies have paid attention to the Pacific and African regions. Due to this, a need arises for future research to investigate whether these Asian and European context findings can be applied to other geographical contexts. As one of the few cross-country and cross-cultural studies, Peña-García et al. (2018) compared online shoppers in Colombia and Spain, finding that cultural and economic differences significantly affected the customer loyalty formation process in online environments. Therefore, more research is needed to investigate whether the antecedents and consequences of brand loyalty on social media differ in different socioeconomic and cultural contexts. Second, almost all the consequences found by this review are related to the benefits of brand loyalty for brands, such as brand purchase intention and brand equity. Very few studies have examined the benefits consumers receive from their loyalty towards a brand. Prior research suggests several consumer-related consequences could occur from brand loyalty, such as personal attachments, financial and non-financial benefits, and helping consumers achieve their lifestyle goals (Nam et al. 2011). Therefore, future research should focus on consumer-related consequences of brand loyalty through social media. Third, another issue that emerged from this review was dual nature variables, which are those variables that can act as both antecedents and consequences of brand loyalty. These variables present an issue of confusion currently in the literature on brand loyalty. Future research should investigate these dual nature variables, provide better definitions for these variables, and examine their “true” relationship with brand loyalty in social media. These imperatives may provide a valuable starting point for future research questions such as:

RQ 4

Do the antecedents and consequences of brand loyalty in the social media environment differ across nations, and why?

RQ 5

What are the benefits of social media engagement with brands for customers?

RQ 6

What is the true relationship between E-WOM and customer brand loyalty in the social media environment?

Future research directions on measuring brand loyalty

Consistent with prior research, we found that there is no single-widely accepted measure of loyalty (Mellens et al. 1996; Rundle‐Thiele and Bennett 2001). Previous literature has identified the association of cognitive loyalty with consumers’ product evaluations through constructs such as perceived value, perceived benefits, and positive experience, which involves functional and psychological aspects toward the brand (Evanschitzky and Wunderlich 2006; Han et al. 2011). From our findings, most studies utilized a brand as the “first choice” compared with other brands based upon previous purchasing experiences. However, other product attributes and evaluations were not measured in terms of cognitive loyalty. Thus, our findings indicate that the consistently favorable attitudes toward the brand (e.g., the brand is consumers’ first choice), as well as meeting consumers’ expectations (e.g., trust toward the brand’s performance), are two major components of measuring the cognitive loyalty. In terms of affective loyalty, only a few studies included affective components (attachment, n = 6; satisfaction, n = 2). The affective dimensions involved with the brand loyalty seems largely omitted from the studies. Nevertheless, in many studies, the constructs of satisfaction, enjoyment, brand love, and brand attachment are considered “antecedences” that positively influence consumers’ brand loyalty.

The items often used to measure conative loyalty are related to positive word-of-mouth and commitment. Oliver (1999) defined cognitive loyalty as a “brand-specific commitment to repurchase” (p.35). Thus, it is unsurprising that repurchase intention, willingness to recommend, and intent to purchase in the future were included. There are only two studies that have included “increased interaction” with brands (Erdoğmuş and Cicek 2012; Qiutong and Rahman 2019). Since social media is often used by brands to enhance customers’ brand engagement and loyalty (Shanahan et al. 2019), future research in scrutinizing the impact of social media marketing and brand loyalty should consider incorporating an outcome of increasing interaction with brands while measuring the cognitive loyalty.

RQ 7

How to distinguish loyalty from trust and commitment? Does loyalty contain a combined concept of the two?

RQ 8

Would other behavioral measurements such as increased interactions and time spent on brand’s content be strong indicators of action loyalty?

RQ 9

Can brand loyalty be measured differently from loyalty to online brand communities?

Limitation

Although the authors used a systematic approach to review this literature conducted in the past 20 years regarding building consumer loyalty on social media, readers should interpret this paper’s findings with the following limitations in mind. First, the present study only focuses on the empirical studies that used multiple items to measure brand loyalty in this review article. Therefore, this study does not include those studies that use behavior measurement of engagement, such as clickstream data or actual purchase behaviors. As clickstream data have been gradually made available for academic discovery, a literature review of the social media impacts on consumer engagement behaviors using clickstream data is warranted. Secondly, we did not include qualitative studies such as content analyses of social media platforms or consumer interviews. Such literature may provide a more detailed illustration of how to build compelling content on social media to facilitate consumer loyalty. Thirdly, the authors recognized that different types of markets (e.g., stable versus high-switching tendency) may require different types of loyalty measures for predicting future levels of loyalty (Rundle‐Thiele and Bennett 2001). Therefore, future studies may want to investigate how different dimensions of loyalty measures can predict better consumption behaviors among different types of brands (e.g., durable products, experiential products, service products, etc.).

Conclusions

A well-written systematic review should inform scholars of what we know, how we know it, and where the next steps for this academic research stream lie (Lim et al. 2021). In this study, we examined a total of 404 papers and found 86 empirical studies that investigated the building of brand loyalty on social media platforms. There are several contributions this review intends to achieve. First, we consolidate the theories applied in 86 papers into five major clusters, namely consumer self-identity and consumption style theories, brand community-related theories, decision-making theories, communication-related theories, and relationship-based theories. Second, we identified eight major antecedents and five consequences of brand loyalty in the context of social media communities. Third, since consumer loyalty is a multi-dimensional construct, there has been inconsistency and confusion on how to measure it. We applied Oliver’s (1999) four-phase loyalty as the framework to consolidate the eleven categories of measurements of consumer loyalty found in the 86 articles. The identified variables from these empirical studies can serve as a springboard for future research directions and theory development. Researchers who intend to conduct consumer loyalty studies may find the examples of the measurement items summarized in this study useful for their research. As researchers can connect university library databases with Google Scholar now (Stanford University 2021), we recommend scholars who want to conduct review studies to adopt Paul et al.’s SPAR-4-SLR protocol, to avoid waste research resources in the review process (Paul et al. 2021). We also proposed future research questions that researchers who want to study consumer brand loyalty can consider answering in their future research.