Customer Brand Engagement on Online Social Media
Customer Brand Engagement on online social media by faniizaldi in Browse > Personal Growth > Self-Improvement....
Department of Business Administration
Master of Science in Marketing
CUSTOMER BRAND ENGAGEMENT ON ONLINE SOCIAL MEDIA PLATFORMS A Conceptual Model and Empirical Analysis
Master Thesis Author: Justina Malciute Advisor: Polymeros Chrysochou
Total number of characters: 104,269
Aarhus University Business and Social Sciences August 2012
Abstract A great interest in the concept of customer engagement has emerged along with the rise of online social media during the past few years. Marketing practitioners were the first ones attempting to define and understand the potential outcomes of customer engagement. However, due to a lack of scholarly interest and empirical support, the nature of customer engagement has remained rather vague and its presupposed capability to enhance customer relationships still uncertain. The aim of this study is to bridge this gap by proposing a conceptual model of customer brand engagement in the context of online social media platforms and conducting an empirical analysis. Drawing on the overview of academic literature and the results of a quantitative online consumer study, the paper delivers a thorough investigation of the concept and offers empirical evidence of its impact on the ultimate business performance. The most important findings of this study suggest that both customer brand relationship related factors and online social media platform related factors can influence the level of customer engagement, which in turn will influence the level of behavioral loyalty and the spread of word-of-mouth communication. Thus, this paper is an important contribution to academic marketing literature in the field of customer engagement, which still remains mostly conceptual or qualitative, and provides useful managerial insights for marketing practitioners. Keywords: customer engagement, brands, social media, customer relationships, brand loyalty, word-of-mouth.
Table of contents
1. Introduction.................................................................................................................. 1 2. Literature review .......................................................................................................... 3 2.1 Conceptual foundations....................................................................................... 3 2.2 Engagement conceptualizations in social science and management literature ... 5 2.3 Engagement conceptualizations in the marketing literature ............................... 6 2.4 Conceptual relationships ................................................................................... 10 2.5 Engagement in the online social media context ................................................ 14 2.6 Problem statement ............................................................................................. 16 2.7 A conceptual model of customer brand engagement on online social media platforms ................................................................................................................. 17 3. Methodology............................................................................................................... 20 3.1 Data collection .................................................................................................. 20 3.2 Measurement of constructs ............................................................................... 23 3.3 Statistical analysis ............................................................................................. 28 4. Results ........................................................................................................................ 30 4.1 Descriptive analysis .......................................................................................... 30 4.2 Measurement reliability and validity ................................................................ 33 4.3 Model estimation results ................................................................................... 37 4.4 Moderation effects ............................................................................................ 38 5. Discussion and implications ....................................................................................... 41 5.1 Implications for marketing theory ..................................................................... 42 5.2 Managerial implications .................................................................................... 43 5.3 Limitations and future research ......................................................................... 45 6. Conclusion .................................................................................................................. 46 References ...................................................................................................................... 48 Appendix 1: Online Questionnaire ................................................................................. 52 Appendix 2: Top Facebook Pages, Worldwide, Food & Drink Brands ......................... 62
List of figures Figure 1. Conceptual model of customer engagement behavior .................................... 14 Figure 2. Conceptual model of customer brand engagement on online social media platforms ......................................................................................................................... 19 Figure 3: Fan distribution based on engagement level (N1=112) .................................. 31
List of tables Table 1: Characteristics of the respondents (N1=112, N2=307) .................................... 23 Table 2: Construct measurement items, sources and scale reliabilities .......................... 26 Table 3: Means, standard deviations and results of t-test for equality of means (N1=112, N2=307) .......................................................................................................................... 31 Table 4: Means, standard deviations and results of t-test for equality of means in behavioral brand loyalty of high and low engaged fans (N1a=56, N1b=56) ................. 32 Table 5: Reliability and validity measures for first-order latent constructs (N1=112) .. 33 Table 6: Average variance extracted and squared correlations between first-order latent constructs (N1=112) ....................................................................................................... 35 Table 7: Reliability and validity measures for second-order latent construct of customer brand relationship related antecedents (N1=112) ........................................................... 36 Table 8: Estimated weights and variance inflation factors for formative dimensions of second-order latent construct of online social media platform related antecedents (N1=112) ........................................................................................................................ 36 Table 9: Results and direct effects of the structural path model (N1=112) .................... 37 Table 10: Results of the two-stage PLS approach for estimating moderating effects (N1=112) ........................................................................................................................ 40
1. Introduction “Engage or die” is the new marketing catchphrase, which emerged as a result of the rise of social media in the past few years (Nelson-Field & Taylor, 2012). The practitioners from various industries caught on to it and the topic quickly became of great interest. Numerous business conferences, seminars, discussion forums, blogs, commentaries and white papers were suddenly talking about the concept of customer engagement, which did not really exist in the marketing literature before (Brodie, Hollebeek, Jurić, & Ilić, 2011a). The rules of engagement are new to the marketers and require some major changes in the conventional marketing thinking. It is no longer a monologue dictated by the firm through a commercial, print ad or a corporate website. The emergence of new media provides businesses with an opportunity to start a two-way digital conversation with the audiences and makes it almost effortless for an individual customer to talk back and also talk to each other (Deighton & Kornfeld, 2009). The new media channels such as YouTube, Wikipedia, Facebook, Twitter or MySpace gave a voice to the customers and made it possible for them to create and easily share their own web content. In other words, each individual has now the opportunity to become a media producer, an author, a reviewer, or engage in many other kinds of behaviors that can be consumed by others on the Internet. Thus, instead of generally being the ones to talk brands have now become the ones mostly talked about. The businesses gradually came to realization that they have to change their way of looking at the customer, and the concept of engagement appeared to be the key to success. The rationale behind this assertion is the prevailing conception of customer engagement as a way to create deeper and more lasting customer brand relationships (Kumar et al., 2010). And even though the traditional media still plays the major role in reaching the customer, the companies are increasingly using the new social media channels for managing their customer relationships. Research showed, that social media has emerged as a valuable tool widely employed by businesses and even 54% of executives of consumer goods companies participating in a recently conducted survey said that social media was central to their effort to engage consumers in 2011(WARC, 2012a). Hence, even though no single theory exists 1
on how customer engagement on social media works, marketers have been actively pioneering the field. Almost every brand today has an established profile on the mainstream social media platforms such as Facebook, Twitter or Google+. Others have also turned to more novel platforms such as Instagram, Pinterest or Foursquare. There are multiple different ways and tools that businesses can use in order to engage their customers. However, despite all the effort the levels of customer engagement resulting on the platforms of social media suggest that the conventional marketing knowledge lacks the ability to explain and guide the marketers throughout the process. One recent practitioner study of the most popular brands on Facebook has discovered that less than 5% of brands were able to attract repeated fan visits to their page within a 30 day period, meaning that under one in 20 fans in a month chose to return to the brand page more than once (WARC, 2012b). On the other hand, the proportion of Facebook fans who not only visit the fan page but also engage with it was found to be even lower. Only 1% of customers observed in another study were found to actually engage with the brand after initially becoming a fan on Facebook (Creamer, 2012). Hence, given the entire struggle that businesses are going through trying to engage their customers, the inevitable question arises – is it worth it? Some of the biggest brand owners such as Coca-Cola, Unilever and Ford who already managed to establish a large fan base are still attempting to define the potential return on investment from using Facebook and expect that it will take at least a couple more years until the value of fans is established (WARC, 2012c). Thus, the brands are willing to take a leap of faith building on the core premise of social media paradigm, which suggests that brands need to engage their customers in order to sustain growth. Yet, the link between the effects of engagement and business performance remains tenuous and fails to explain the return in real terms (Nelson-Field & Taylor, 2012). Not surprisingly, the concept of engagement on social media platforms has also received criticism and is sometimes even referred to as an air of the early-dot-com hype (Baker, 2009), given that its effectiveness and consequences to the brand are still largely uncertain. The buzz of social media along with the dilemma of the newly emerged concept of customer engagement among the practitioners has also started attracting the interest of marketing scholars. The Marketing Science Institute has listed customer engagement as one of the research priorities for the period of 2010-2012 recognizing the lack of conceptual 2
frameworks and methods for understanding this concept (MSI, 2010). Hence, making use of the new media opportunities requires a deeper knowledge of how customers engage with the different types of media and what it ultimately means for the brand. This study attempts to develop a conceptual model of customer brand engagement on online social media platforms by reviewing the existing marketing literature concerning the concept and subsequently refining it through empirical analysis in order to help marketers better understand how the process of customer engagement works in this increasingly complex landscape of social media.
2. Literature review While the notion of engagement is not new in the literature of various academic disciplines, it has only emerged in the field of academic marketing relatively recently. Before 2005 there were very few academic articles in the field of marketing which have mentioned the term “engagement” (Brodie et al., 2011a). Since then the term has gained popularity. However, despite the significant practitioner interest evolved during the last decade, there have only been a few systematic scholarly attempts to define the concept, its distinctiveness from the more traditional relational concepts like participation or involvement, and, finally, the conceptual roots of customer engagement.
2.1 Conceptual foundations A few underlying logic perspectives were identified in the academic literature exploring the conceptual foundations of customer engagement. First of all, Brodie et al. (2011a) suggest that the theoretical roots of customer engagement can be examined by drawing on the service-dominant (S-D) logic and the relationship marketing theory. The S-D logic is a framework that conceptualizes business exchange by addressing service as the main purpose, and explains how the different network actors (firms, customers and other stakeholders) can co-create value while interacting with each other (Karpen, Bove, & Lukas, 2012). The term “service” here is referring to “the process of using one‟s resources for the benefit of another entity” (Vargo & Lusch, 2008). The logic implies that the cocreation of superior value is replacing the more traditional notion of value provision, 3
meaning that creating superior value in cooperation with the customer becomes a source of competitive advantage for the firms. To date, a set of 10 foundational S-D logic premises have been established building on marketing relationships characterized by customers‟ interactive service experience (Vargo & Lusch, 2008): 1. 2. 3. 4. 5.
Service is the fundamental basis of exchange. Indirect exchange masks the fundamental basis of exchange. Goods are a distribution mechanism for service provision. Operant resources are the fundamental source of competitive advantage. All economics are service economics.
6. The customer is always a co-creator of value. 7. The enterprise cannot deliver value, but only offer value propositions. 8. A service-centered view is inherently customer oriented and relational. 9. All social and economic actors are resource integrators. 10. Value is always uniquely and phenomenologically determined by the beneficiary. Four of these underlying S-D logic premises (numbers 6, 8, 9 and 10) have been found as of particular relevance in explaining the conceptual roots of customer engagement (Brodie et al., 2011a). Together the four premises imply that value is not something embedded in the product, but the benefit that the customer gets out of using the product instead. Thus, the nature of value is highly contextual and subject to experiences (Karpen et al., 2012). Moreover, value that is realized through market exchange always involves a combination of resources and, therefore, cannot be created unilaterally, which makes the customer a co-creator of value (Vargo & Lusch, 2008). Naturally, the interactive nature of the co-creation process leads to viewing the firm and the customer in a relational context and, since the benefit is always determined by the customer, it is inherently customer oriented. Finally, the value is created within the networks where the firms and individuals are motivated to interact in order to integrate their specialized resources and create more complex services (Vargo & Lusch, 2008). These four S-D logic premises build a solid conceptual foundation for the concept of customer engagement. In particular, it is also suggested that the customer experiences of the co-creative and interactive nature taking place in complex relational environments may actually be viewed as the act of “engaging” (Brodie et al., 2011a). 4
Another perspective of exploring the conceptual foundations of customer engagement draws on the so called broadened relationship marketing domain (Brodie et al., 2011a; Brodie et al., 2011b; Hollebeek, 2011a). Relationship marketing refers to “all marketing activities directed toward establishing, developing, and maintaining successful relational exchange” (Morgan & Hunt, 1994), which are critical to the firms in order to build valuedriven interactive long-term relationships with their existing as well as potential customers and organizational networks and facilitate the processes of value co-creation (Brodie, Ilic, Juric, & Hollebeek, 2011b). While the S-D logic and relationship marketing perspectives introduce the notion of the customer behavior being focused on interactive and co-creative experiences in the complex relational networks, Hollebeek (2011a) also draws on the social exchange theory to explain the rationale behind the customers‟ motivation of contributing to the superior value creation. The social exchange theory functions under the premise that one party will do a favor to another party because of being motivated by expected future return. Therefore, it would also suggest that a customer experiencing a benefit from a brand relationship is expected to respond with positive thoughts, feelings and behaviors (L. Hollebeek, 2011a). As a result, all three foundational perspectives of customer engagement build on the interactive nature of exchange between the value creating network actors.
2.2 Engagement conceptualizations in social science and management literature The concept of engagement has been used in various disciplines including organizational behavior, psychology, sociology and political science. Different studies have been exploring various sub-forms of engagement (e.g. civic engagement, social engagement, student engagement, engagement of nation states, employee engagement, stakeholder engagement), which led to a variety of approaches to interpreting the concept (Brodie et al., 2011a). Literature within organizational behavior describes engagement as physically, emotionally or cognitively expressed “task behaviors that promote connections to work and others”, which motivate the employees and encourage personal development (Bowden, 2009). The concept of social engagement in the field of social psychology has been defined 5
as “a sense of initiative, involvement and adequate response to social stimuli, participating in social activities and interacting with others”, whereas student engagement in the field of educational psychology has been conceptualized as “students‟ academic investment, motivation and commitment to their institution, their perceived psychological connection, comfort and sense of belonging towards their institution” (L. D. Hollebeek, 2011b). An overview of the diversity of engagement conceptualizations across the different academic disciplines reveals few important observations. First of all, engagement can be viewed as a process that can be characterized by “specific interactions and/or experiences between a focal engagement subject (e.g., student; customer) and object (e.g. course/module; brand, product, or organization, respectively)” (Brodie et al., 2011b). Second, most of the reviewed conceptualizations present engagement as a multidimensional concept comprising behavioral (actions), cognitive (thoughts) and emotional (feelings) dimensions (L. Hollebeek, 2011a). Even though there is still a relatively large number of researchers, who view engagement from the unidimensional perspective, the focus remains on the three mentioned dimensions with the behavioral focus being the dominant one (Brodie et al., 2011a). According to the Oxford Dictionary the verb “to engage” means to employ or hire, to bind by a contract, to hold fast, and to take part in something (van Doorn et al., 2010). All these meanings point to the behavioral aspect of engagement, however, the unidimensional perspective lags behind in expressing the wider scope of the concept (Brodie et al., 2011a). Furthermore, Hollebeek (2011a) also points out that despite of looking into engagement from a wide range of disciplines, all the different definitions of the term reveal favorable expressions towards the concept and its highly interactive nature. The next section will explore engagement research in the practitioner and academic marketing literature.
2.3 Engagement conceptualizations in the marketing literature The exploration of available marketing literature reveals the emergence of several engagement sub-forms, such as “customer engagement”, “customer engagement behaviors”, “consumer engagement”, “customer brand engagement” as well as the more general conceptualizations of simply the “engagement” itself (L. Hollebeek, 2011a). 6
Bowden (2009) presents customer engagement as a sequential psychological process that customers move through to become loyal towards a brand. This process is suggested to model the mechanisms by which loyalty may be developed and maintained for two different types of customers – new and existing. Bowden (2009) is also discussing the distinction between customer engagement and the more traditional marketing constructs such as involvement, commitment and loyalty. It is in fact suggested that customer engagement process helps to examine the dynamic relationships between these constructs and further the understanding of how they drive the development of customer loyalty. Customer engagement has also been explored as a new perspective in the field of customer management (Verhoef, Reinartz, & Krafft, 2010). It has been highlighted that the emerging concept of customer engagement is highly important in the increasingly networked society. Building on the research of van Doorn et al. (2010), Verhoef et al. (2010) consider customer engagement as behavioral manifestations towards a focal object (e.g. a brand or a firm), other than purchase, resulting from motivational drivers. The concept of customer engagement behaviors implies that van Doorn et al. (2010) are focusing on the behavioral aspects of the relationship between the customer and the firm. Some other authors have also suggested that customer engagement includes a continuum of behaviors ranging from pure voice (complaining, recommendation, word-of-mouth) to pure exit (reduced or discontinued consumption) (Hirschman, 1970). All the customer engagement behaviors are proposed to comprise five dimensions: valence (positive or negative), form and modality, scope (temporal and geographic), nature of impact and, finally, customer goals. Moreover, van Doorn et al. (2010) establish a conceptual model suggesting that customer engagement behaviors are affected by customer characteristics, firm initiatives and the contextual environment. In addition, they also present a number of consequences that customer engagement behaviors bring to the firm, the society and the customer itself. Despite the customer management research mostly being focused on the transactional side of the customer-firm relationship, the non-transactional forms of behavior have also gained their share of attention recently. Verhoef et al. (2010) acknowledge the importance of the impact of word-of-mouth and co-creation in particular. It has been recognized that ignoring the non-transactional behavior manifestations may have detrimental effects to the firm because of potentially wrong valuation of the customers 7
(Kumar et al., 2010). The paper of Kumar et al. (2010) introduces a new metric for customer valuation, where they include both the value from transactional and the nontransactional behaviors and, therefore, disagree with the view of van Doorn et al. (2010). Hollebeek (2011b) presents the concept of customer brand engagement and defines it as “the level of an individual customer‟s motivational, brand-related and context-dependent state of mind characterized by specific levels of cognitive, emotional and behavioral activity in direct brand interactions”, where the focus lies on the interactions between a specific subject (the customer) and the focal object (brand). The cognitive activity refers to the level of engrossment or concentration towards a brand, whereas the emotional and behavioral activities reflect the level of an individual‟s pride or inspiration and the level of energy expressed while interacting with the brand, respectively (L. D. Hollebeek, 2011b). Just like Bowden (2009), Hollebeek (2011b) also suggests that customer brand engagement contributes to developing customer loyalty by focusing on conceptualizing the positively valenced expressions of customer brand engagement. In her other works Hollebeek (2011a) further explores the concept of customer brand engagement and, by utilizing qualitative research methods, identifies the key themes of customer engagement behavior: immersion, passion and activation. This implies that the level of customer‟s brand-related concentration (immersion), positive affect (passion) and the level of energy put in particular brand interactions (activation) together represent just how much the customer is prepared to exert cognitive, emotional and behavioral investments while interacting with the focal brand (L. Hollebeek, 2011a). Mollen and Wilson (2010) elaborate on the concept of engagement from the perspective of online consumer experience. Building on the findings from e-learning and online marketing literature, the authors suggest that a consumer‟s experiential response to a website or some other computer-mediated entity comprises three experiential states including perceived interactivity, telepresence and engagement. In particular, engagement is defined as “a cognitive and affective commitment to an active relationship with the brand as personified by the website or other computer-mediated entities designed to communicate brand value” and is suggested to comprise the dimensions of active, sustained, cognitive
processing, attainment of instrumental value (relevance and utility), and experiential value (emotional congruence) (Mollen & Wilson, 2010). Another conceptualization addressed in the literature is the “brand engagement in selfconcept” (Sprott, Czellar, & Spangenberg, 2009). The construct suggests that consumers vary in their tendency to possess brand related schemas, meaning that differences exist in consumers‟ tendency to engage brands in their self-concepts and, therefore, also in their brand-related behaviors. Sprott et al. (2009) develop a scale to measure the self-brand connections in individuals. However, the concept has been criticized for failing to fully capture the interactive nature of customer engagement (Brodie et al., 2011a). Engagement has also been conceptualized as a state of sustained attention, which can be characterized by full absorption and involvement as well as being fully occupied or engrossed in something (Higgins & Scholer, 2009). Higgins & Scholer (2009) also recognize that individuals can be engaged on different levels of intensity and suggest that the more a person is engaged, the more intense will be the experience of the motivational force. This means that a more engaged individual will experience the positive target more positively and the negative target more negatively in the pursuit of his goal. Thus, the authors express considerations towards both positive (e.g. attraction) and negative (e.g. repulsion) expressions of engagement. Brodie et al. (2011a) have derived the main themes prominent in the literature concerning customer engagement and developed a set of five fundamental propositions, which consequently provide the basis for the suggested general definition: “Customer engagement (CE) is (1) a psychological state that occurs by virtue of interactive, co-creative customer experiences with a focal agent/object (e.g. brand) in focal service relationships. It occurs (2) under a specific set of context-dependent conditions generating differing CE levels; and (3) exists as a dynamic, iterative process within service relationships that co-create value. CE plays (4) a central role in a nomological network governing service relationships in which other relational concepts (e.g. involvement, loyalty) are antecedents and/or consequences in iterative CE
processes. It is (5) a multidimensional concept subject to a context- and/or stakeholderspecific expression of relevant cognitive, emotional and/or behavioral dimensions.” Unlike most other reviewed conceptualizations, Brodie et al. (2011a) suggested a definition that can be applicable in a wide range of contexts. Furthermore, a generic expression of the dimensions (cognitive, emotional and behavioral) comprising the engagement concept allows for it to encompass any context-specific expressions of the customer engagement. However, this particular conceptualization has also received criticism for being too broad and exposing to the danger of confounding the behaviors, which are potentially caused by engagement, and all other behavioral indications (Malthouse & Calder, 2011). A comment on Brodie‟s et al. (2011a) conceptualization also suggests that the interactive and co-creative nature of experiences should not imply that engagement requires a high level of overt activity. Malthouse & Calder (2011) point out that engagement can arise not only from active behaviors such as e.g. blogging, but simply receiving communication can also be viewed as interactive and co-creative, as long as these experiences are immersive. Finally, Brodie‟s et al. (2011a) definition also addresses the issue of differentiating customer engagement from other relational concepts and suggests that they represent the potential antecedents and/or consequences embedded in the iterative process of service relationships.
2.4 Conceptual relationships Exploring the newly emerged concept of customer engagement may also lead to a question whether it could simply be the case of “the old wine in a new bottle” (Bowden, 2009). However, all the authors researching different sub-forms of engagement (Brodie et al., 2011a; Hollebeek, 2011a; Mollen & Wilson, 2010; Bowden, 2009) argue that this is not the case and that there is a clear distinction between engagement and other more familiar relational concepts. Mollen & Wilson (2010) suggest that involvement is an important dimension of engagement and therefore an important relational concept to discuss. Involvement has been defined as an internal state of arousal, which can be used to reflect an ongoing concern by 10
the customer towards a product based on the perceived importance and/or general interest in the purchase process (Bowden, 2009). Mollen & Wilson (2010) identify three major differences between engagement and involvement. First of all, the definition of involvement indicates that it requires a consumption object (e.g. product category). Second, involvement refers to a more passive allocation of mental resources and unlike engagement does not encompass an active relationship with the consumption object. Finally, engagement not only requires the attainment of instrumental value through relevance and utility, but also a certain degree of emotional bonding, which can be achieved through pleasant and satisfying experiences. Besides involvement, Bowden (2009) also compares the customer engagement process with and delineates the distinction from the concepts of commitment and loyalty. Commitment often encompasses some sort of psychological attachment, where a customer views a specific commitment object as the only acceptable choice alternative. Thus, commitment generally means that unlike in the case of involvement, a customer is not simply interested in an issue, but rather holds an actual attitudinal position. Loyalty is also known to comprise an attitudinal element. However, it is most often evaluated in the behavioral manner, e.g. the intention to repeat a purchase. Commitment and loyalty are often considered as highly related concepts. Nevertheless, the effects of the two may yield different behavioral outcomes. It has been discovered that due to attitudinal attachment brand-committed customers are actually less likely to switch brands than the brand-loyal customers (Bowden, 2009). Mollen & Wilson (2010) also discuss the constructs of interactivity, flow and telepresence in relation to the online brand engagement. However, these are depicted as a process, where interactivity is assumed to be an antecedent of telepresence, which consequently is an antecedent of engagement. There is no consensus about the definition of interactivity in the literature, so the authors propose their own definition, which suggests that interactivity is an “experiential phenomenon”, which describes to what degree customers perceive the communication as “two-way, controllable and responsive to their actions”. The construct of flow is viewed as a cognitive state, which asserts when individuals are so involved in an activity, that it makes them forget everything else. 11
Telepresence is related to flow, however, it extends to a psychological state of being present in a computer-mediated environment. The process of telepresence is expected to positively affect both the instrumental and the experiential value and, thus, suggested to be an antecedent of engagement. Brodie et al. (2011a) building on one of their fundamental propositions to the concept of customer engagement also suggest that it is only a part of a broader relationship structure, where the other concepts play the roles of antecedents and/or consequences. Apart from some of the constructs mentioned already, Brodie et al. (2011a) also consider and justify a number of other potential antecedents and/or consequences of customer engagement, such as participation, rapport, customer satisfaction, trust, self-brand connection, and emotional attachment. The authors have found some relational constructs such as involvement and participation to be prerequisite to drive engagement, whereas the others could act as both potential antecedents and consequences within particular dynamic service relationships. This point of view is in line with the argument of Bowden (2009) saying that new and existing brand customers will follow a different engagement process in developing loyalty. The iterative nature of the service relationships implies that different concepts will play different roles in different contexts. For instance, an exploratory analysis investigating consumer engagement in a virtual brand community has revealed that the consequences of consumer engagement in that particular case included loyalty, satisfaction, empowerment, connection, emotional bonding, trust and commitment (Brodie et al., 2011b). Furthermore, Hollebeek (2011b) has pursued defining the conceptual relationships of customer brand engagement and identified involvement and interactivity to be the antecedents required prior to the expression of a relevant customer‟s brand engagement level. Flow has also been determined to be an antecedent state, whereas the concepts of co-created value, brand experience, perceived quality, customer value and brand loyalty are suggested to represent the potential consequences of customer brand engagement. Finally, rapport, customer satisfaction, trust and commitment could act as both antecedents and/or consequences depending on whether the customer is new or existing. The concepts of interactivity,
rapport and value co-creation in particular have been noted as of high relevance in service contexts and Web 2.01 settings, which can be characterized by human interactive forms. Van Doorn et al. (2010) introducing the concept of customer engagement behaviors offer a somewhat different perspective of the potential antecedents and consequences, and present a useful theoretical framework for research in this area (see Figure 1). As already mentioned in the previous section, the authors suggest a conceptual model, which examines different types of motivational drivers and outcomes of customer engagement behaviors. The antecedents are divided into three major groups and include not only customer-based, but also firm-based and context-based factors. The model implies that not only customer related factors, such as attitudes, goals, resources and perceptions, but also the characteristics of the brand and the firm together with the different aspects of contextual environment can have just as much impact on customer engagement behaviors. Though, some of these factors may not necessarily elicit a direct effect. The model also indicates that the factors can interact with each other and moderate the effect of other particular factors on customer engagement behaviors. The consequences considered in the model include the effects on the customer, the firm and other constituents (e.g. the customers of other products and brands). To sum up, the literature reviewed explores different sub-forms of engagement and offers a variety of conceptualizations. Yet, even though the topic has received considerable attention among the practitioners (Cheung, Lee, & Jin, 2011), the lack of consensus in the academic literature suggests that the concept of customer engagement is still understood in a rather unsystematic way.
“Web 2.0 is a collection of open-source, interactive and user controlled online applications expanding the experiences, knowledge and market power of the users as participants in business and social processes.” (Constantinides & Fountain, 2008) 13
Figure 1. Conceptual model of customer engagement behavior ANTECEDENTS CONSEQUENCES
Satisfaction Trust/commitment Identity Consumption goals Resources Perceived costs/benefits
Firm-Based Brand characteristics Firm reputation Firm size/diversification Firm information usage and processes Industry
CUSTOMER ENGAGEMENT BEHAVIOR
Valence Form/modality Scope Nature of impact Customer goals
Cognitive Attitudinal Emotional Physical/Time Identity Firm
Financial Reputational Regulatory Competitive Employee Product Others
Consumer welfare Economic surplus Social surplus Regulation Cross-brand Cross-customer
Context-Based Competitive factors P.E.S.T. o Political o Economic/ environmental o Social o Technological
Source: van Doorn et al. (2010)
2.5 Engagement in the online social media context Internet is an open, highly cost-effective and far reaching global network, which helps reducing or even eliminating the barriers of geography and distance (Sawhney, Verona, & Prandelli, 2005). In the physical world, businesses often face the trade-off between the 14
richness and the reach of their communication. That is, a rich dialogue with a customer requires personal interaction and physical proximity, which means that there is only a limited number of customers that the firm can communicate with in the most effective manner. Internet, however, allows the firms to overcome these constraints and reach a much larger number of customers without having to lose on the richness of the communication too much. The emergence and rise of new social media channels in the recent years enabled the customers to increasingly participate in the new forms of customer/firm interaction processes. Discussion forums, chat rooms, email, bulletin boards, blogs and social networks are just some of the tools facilitating interactive customer experiences, that may eventually also foster the development of customer engagement with the specific brands (Brodie et al., 2011b). Hollebeek (2011b) also recognizes the importance of customer engagement in the so called Web 2.0 applications, which are designed in a way that enables them to aggregate the information from their user base in order to expand their content as well as value (Wilkins, 2007). Some practitioners even refer to customer engagement as the Holy Grail in the context of online marketing (Mollen & Wilson, 2010). One of the main reasons behind the suggested importance of the concept lies in the definition of Web 2.0 and the fact that this kind of setting would not persist without the user-generated content, which in turn requires users to be engaged in the new media. Not surprisingly, this specific sub-form of engagement has also gained attention among the researchers. For instance, Cheung et al. (2011) have initiated a study exploring customer engagement in online social platforms. The authors of the research-in-progress paper have defined it as “the level of a customer‟s physical, cognitive, and emotional presence in connections with a particular online social platform”. The conceptual model developed suggests that customer engagement in an online social platform is a construct comprising vigor (level of energy and mental resilience), absorption (level of concentration and engrossment) and dedication (sense of significance, enthusiasm, inspiration, pride and challenge) towards the online social platform, which are driven by involvement and social interaction. The consequences reflected in the model exhibit the authors‟ belief that customer engagement will have a positive effect on online social platform participation and word-of-mouth communication about the platform (Cheung et al., 2011). The study by Cheung et al. (2011) is expected to 15
contribute highly to the existing knowledge about social media engagement by providing a validated measurement scale for customer engagement in online social platforms. However, the research is still in progress and no results have been published to date. Thus, even though the new media present a number of significant opportunities and challenges for both researchers and practitioners (Hennig-Thurau et al., 2010), most of the existing research is primarily conceptual or qualitative (Cheung et al., 2011).
2.6 Problem statement Academic literature highlights the importance of approaching the concept of engagement with consideration to its highly contextual nature, because “engagement, separated from its (…) context, is a contradiction that ignores deeply embedded understandings about the purpose and nature of engagement itself” (Vibert & Shields, 2003). Moreover, Brodie et al. (2011a) suggest that under different circumstances the importance of the cognitive, emotional, and behavioral customer engagement dimensions may vary. Therefore, it is likely that customer engagement in different contexts, such as online versus offline environments, would manifest in different expressions. The context of online social media has become of great interest to marketing practitioners as the new social media platforms quickly emerged as valuable tools central to their effort of customer engagement (WARC, 2012a). Despite the vast popularity of the concept among businesses, the push of engagement still misses the mark and fails to explain what it ultimately means to the brand. The behavioral measures of engagement currently available on online social media platforms such as number of fans, repeated visits or interactions with the brand page provide little information about the returns to be expected (Nelson-Field & Taylor, 2012). Hence, the lack of theory-guided empirical studies in order to better understand customer engagement with brands in the context of online social media points to a fault line between the practitioners who increasingly pursue the quest for their “Holy Grail”, and the scholars who yet mostly choose to focus their empirical research elsewhere.
Hence, the main objective of this study is to bridge this gap by conceptualizing customer brand engagement on online social media platforms and answering two important research questions: 1.
What drives the customer to engage with brands on online social media platforms?
What are the outcomes of such engagement?
Identifying and validating the antecedents and consequences of customer brand engagement in this particular context is crucial in order to further advance the knowledge in the area. According to Hollebeek (2011b), the rising practitioner interest in the concept of customer brand engagement is mostly driven by the expected benefits and its explanatory and predictive power in customer relationship outcomes, such as loyalty in particular. Since it is more cost-effective to retain the existing as opposed to winning new customers, insights into customer brand engagement on online social media platforms may help businesses to capitalize on enhancing customer relationships, retention and loyalty through the use of social media.
2.7 A conceptual model of customer brand engagement on online social media platforms The five fundamental propositions underlying the general concept of customer engagement suggested by Brodie et al. (2011a) provide suitable guidelines for framing the investigation of the nature and role of customer brand engagement on online social media platforms. These five themes were therefore applied in developing the working definition and building the conceptual model. The proposed working definition in this study is the following: The concept of customer brand engagement on online social media platforms is characterized by interactive customer experiences with the brand. It is a process of dynamic and iterative nature, which stems from the domains of S-D logic, relationship marketing and social exchange theory. Customer brand engagement on online social media platforms is the central element embedded in a broader network of other relational constructs serving as the antecedents and the consequences. The concept of 17
engagement is multidimensional and comprises the expressions of emotional, behavioral and cognitive engagement specific to this particular context. Based on this definition and the findings from the literature review, a conceptual model of customer brand engagement on online social media platforms was developed (see Figure 2). The framework portrays customer brand engagement on online social media platforms as the central element embedded in the network of other constructs, which are divided into two groups of potential antecedents and consequences. In principle, the structure of the framework relates to van Doorn‟s et al. (2010) conceptual model of customer engagement behavior. However, instead of considering three types of factors that can affect engagement, the current model is focused on customer-based antecedents and consequences only. The customer-based perspective has been chosen, since not only it represents the inevitable focus of the business, but the consequences of engagement to the customer are also suggested to have an inherent effect on the ultimate business performance (Kumar et al., 2010). Furthermore, as suggested in the working definition, the conceptual framework does not only comprise the behavioral aspect of engagement, but addresses the concept in a broader sense by including the cognitive and emotional aspects as well. The group of potential antecedents portrayed in the model includes factors related to customer brand relationship quality and online social media platforms. The customer brand relationship quality related factors are further specified as involvement, satisfaction, commitment and trust. Brodie et al. (2011a) suggest involvement to be a required antecedent of customer engagement, whereas customer satisfaction, commitment and trust in relation to the brand represent the potential attitudinal antecedents also proposed by Bowden (2009) and Hollebeek (2011b). Because of the iterative nature of customer engagement, all three attitudinal factors have been found to have the potential of acting as both antecedents and consequences. The role of the factor will vary depending on whether the customer is new or existing (L. D. Hollebeek, 2011b). The structure of the conceptual model given in Figure 2, however, implies that it was chosen and built on the premise of existing customers in particular.
Figure 2. Conceptual model of customer brand engagement on online social media platforms ANTECEDENTS CUSTOMER BRAND RELATIONSHIP RELATED
INVOLVEMENT SATISFACTION COMMITMENT TRUST
CUSTOMER BRAND ENGAGEMENT ON ONLINE SOCIAL MEDIA PLATFORMS
BEHAVIORAL BRAND LOYALTY EMOTIONAL
ONLINE SOCIAL MEDIA PLATFORM RELATED
INVOLVEMENT PARTICIPATION TELEPRESENCE EASE OF USE
Another sub-group of antecedents comprises online social media platform related factors, such as involvement, participation, telepresence and ease of use. Even though involvement has already been included to the relationship quality related factors, the latter case addresses the concept in terms of personal interest and relevance towards online social media platforms. Participation, according to Brodie et al. (2011a), is another prerequisite for customer engagement, as it determines customers‟ propensity to participate on online social media platforms. Furthermore, the concept of telepresence is included in the model, since Mollen and Wilson (2010) suggest it to be a direct antecedent of online engagement. Hollebeek (2011b) and Brodie et al. (2011a) also suggested the concept of flow, which is related to telepresence and could also be considered relevant in this specific context. However, as no commonly accepted conceptualization or consensus regarding the operationalization of flow exists in the academic literature (Mollen & Wilson, 2010), it has been decided to leave the concept out of the model. Finally, ease of use has also been added 19
to the model as a potential contextual antecedent referring to the degree to which a customer perceives using online social media platforms to be free of effort (Davis, 1989). As for the consequences, two customer-based items were selected – brand loyalty and word-of-mouth, which here refers to the intention to recommend the brand. Bowden (2009) addresses customer engagement as the superior predictor of customer loyalty as compared to other more traditional marketing constructs. On the other hand, Cheung et al. (2011) suggest that a customer willing to invest physical, cognitive and emotional effort into an online platform will also have a higher propensity to spread word-of-mouth communication about it. A customer valuation framework introduced by Kumar et al. (2010) suggests that the value of customer engagement is comprised of four dimensions: customer purchasing behavior, customer referral behavior, customer influencer behavior through customers‟ influence on other existing or prospect customers, and finally, customer knowledge behavior via feedback provided to the firm. Thus, both customer loyalty and word-ofmouth have established grounds as potential engagement consequences in the literature.
3. Methodology 3.1 Data collection In order to collect the data and test the proposed model of customer brand engagement on online social media platforms an online survey was conducted using a convenience sample of Facebook2 account holders. With 901 million active monthly users Facebook is currently world‟s largest online social network (Facebook, 2012) and a highly relevant platform for this study. Among many various online services offered by Facebook, there is also something called Facebook Pages. Facebook Pages are public profiles meant to promote brands, products, artists, web sites or organizations. Once registered Facebook users visit a Page, they are able to 'become fans' by clicking on the 'Like' button. The owners of the Page can then post informational content, which consequently will appear in the news feed of their fans. The fans can choose to react to the posts in few different ways such as „liking‟, commenting or sharing it with their own networks. In other words, 2
Facebook is a medium that can give any brand a voice and allows it to establish an active conversation with Facebook users. It has therefore been largely employed by various brands and used as a tool for customer engagement. According to the statistics, some of the largest Facebook brands in terms of number of fans belong to food and drinks‟ product category (FanPageList, 2012), which has also been chosen to be the focus of this study. Even though one recent paper about Facebook has showed that the degree of fan engagement with brands from any given category is highly similar (Nelson-Field & Taylor, 2012), it is still important to narrow it down to a single category as to assure that the antecedents and the consequences of engaging with the brands are more or less homogeneous. The data collection procedure comprised two stages - a pilot study to pretest the survey instrument and a full-scale field study. During the pretest a self-administrated online questionnaire was created on an online survey tool Qualtrics3 and distributed to a number of selected web forums. A total of 57 responses were collected. The results of the pilot test have been used for reviewing and refining the questions. The full-scale questionnaire has also been launched online and distributed using various web tools such as email, social networking platforms (e.g. Facebook) as well as various international forums. The questionnaire comprised a few basic parts. It started out with an introduction to the survey and a screening question, to make sure that only those, who have a Facebook account, participate in the survey. Further questions were related to the usage of and perceptions about Facebook, such as involvement, participation, ease of use and telepresence as well as three control variables (customer goals, resources and perceived cost/benefit). In order to get to the next part of the questionnaire the participants had to state whether they are fans of any of food or drink brands on Facebook, which then allowed to divide the total sample (N) into two major groups – 1) respondents who are fans of at least one food or drink brands on Facebook (N1); 2) respondents who are not fans of any food or drink brand on Facebook (N2). The respondents in the first group were then asked about their engagement with a certain brand of their choice on Facebook as well as the ongoing relationship with that brand and future intentions related to loyalty and recommending the brand to others. A list
of 15 most popular food and drink product brands at that very moment was provided to the respondents to choose from. The brand popularity ranks were retrieved from a social media counter application previously called Famecount4 (see Appendix 2).The respondents also had an option to enter a brand name of their own liking, in case it was not provided on the list. Since the respondents in the second group were not fans of any food or drink brands, they were simply asked to pick a brand from the same list that they liked most (also with the option of entering a brand name of their own). They were then directed straight to the questions relating to the customer brand relationship and its outcomes. The final part of the questionnaire included socio-demographic questions, such as age, gender, country of origin as well as the usage of other online social media platforms. The final survey sample (N) contained a total of 419 internet users from all over the world, who also had an account on Facebook. Almost 27% of those Facebook users have identified themselves as fans of at least a single Facebook page dedicated to a brand in the food and drinks‟ product category, meaning that N1=112 and N2=307. The total sample included respondents from various age groups ranging from teenagers to seniors, with the largest group consisting of 20-29 year olds (70% of all respondents). A chi-square test was performed on age and other demographic variables to investigate whether there are statistically significant differences between the two groups of N1 and N2. Table 1 below reports demographic characteristics of the two sub-samples along with the results of the chi-square test. The findings of the test suggest that there were no statistically significant differences between the two groups of fans and non-fans with respect to age and gender of the respondents. However, the use of other online social media platforms and the time spent on it per day were found to be related to the group of the respondent. In particular, those respondents who indicated themselves as fans of at least one food and drink brand on Facebook showed a tendency of using a higher number of various online social media platforms and were to spend more time on Facebook and other platforms per day.
From the 1st of May 2012 Famecount has changed its name to Starcount. For more information visit http://www.starcount.com/pages/starcount.
Table 1: Characteristics of the respondents (N1=112, N2=307)
Age Younger than 20 20 – 24 25 – 29 30 – 34 35 – 39 40 – 44 45 – 49 50 and older Gender Male Female Use of other online social media platforms No other 1-2 others 3-5 others 6-9 others 10 and more others Time spent on online social media platforms per day Less than 30 mins 30 mins – 1 hour 1 hour – 2 hours 2 hours – 3 hours More than 3 hours Time spent on Facebook per day Less than 30 mins 30 mins – 1 hour 1 hour – 2 hours 2 hours – 3 hours More than 3 hours
Fans Non-fans (N1) % (N2) % 2 X (7) = 4.43, ρ = 0.729 7 4 37 33 33 38 11 15 4 4 4 2 2 1 3 3 2 X (1) = 0.70, ρ = 0.401 58 53 42 47 2 X (4) = 10.51, ρ = 0.033 8 9 32 43 35 35 23 12 2 1 2 X (4) = 15.92, ρ = 0.003 12 26 22 25 33 26 16 14 17 8 2 X (4) = 13.97, ρ = 0.007 19 36 29 27 27 23 14 9 11 6
3.2 Measurement of constructs The survey instrument comprised of 62 items measuring the constructs mentioned in the model – the antecedents, the consequences, and the customer brand engagement on online social media platforms itself. There were two groups of constructs representing the potential antecedents – customer brand relationship quality related and online social media platform related. The customer 23
brand relationship quality related constructs (involvement, satisfaction, commitment and trust) have been widely discussed in academic marketing literature and the choice of scales for these constructs has therefore been based on the findings of previously published research. Brand involvement has been operationalized via five items measuring an individual‟s level of interest, importance and personal relevance in relation to the brand (Beatty & Talpade, 1994). Commitment has been measured with a six item scale valuing an ongoing relationship between the customer and the brand as well as willingness to make efforts in order to maintain it (Aaker, Fournier, & Brasel, 2008). The satisfaction scale included three items focusing on the general performance of the brand (Gustafsson, Johnson, & Roos, 2005). Finally, the construct of trust has been measured with four items relating to an individual‟s perceptions and beliefs regarding the safety and security of interacting with the brand (Chaudhuri & Holbrook, 2001). The suggested antecedents related to online social media platform were involvement, participation, ease of use and telepresence. Involvement in online social media platform has been measured with the same five item scale adapted from the paper by Beatty & Talpade (1994). The construct of participation in an online social media platform has been approached as the frequency and the intensity of participation as suggested by van Doorn et al. (2010), and measured with three self-constructed items. The ease of use scale has been adapted from a research paper by Davis (1989) and included six items. Even though telepresence has been discussed in the literature and defined as the psychological state of “being there” in the computed-mediated environment (Mollen & Wilson, 2010), there is no actual measuring instrument developed for telepresence in the online social media platform context yet. Therefore, a set of four relevant items from an originally eight item scale by Kim & Biocca (1997) meant to measure telepresence in the context of television has been adapted and used in this survey. Customer brand engagement on online social media platforms has been split into three dimensions – behavioral, emotional and cognitive. The emotional and cognitive engagement scales have been used as suggested by Cheung et al. (2011), where both constructs are measured with six items each. The behavioral dimension, however, only included two relevant items of those suggested by Cheung et al. (2011) and has been 24
supplemented with seven other self-constructed items referring to the frequency of the different forms of behavioral engagement. Nelson-Field & Taylor (2012) suggest that in social media, and particularly on Facebook, engagement takes the form of all kinds of direct interaction with the fan page. The inclusion of seven additional Facebook specific items was also based on this premise. Thus, the self-constructed items refer to the frequency of various interactions with a particular fan page, such as visiting the page, noticing, reading, „liking‟, commenting and sharing its contents as well as creating and posting contents on the fan page yourself. The response format chosen for these seven items has been a seven point frequency scale (1=”Never”, 2=”Almost never”, 3=”Rarely”, 4=”Sometimes”, 5=”Often”, 6=”Almost all the time”, 7=”All the time”). The response format used for the rest of the items in the questionnaire was a seven point Likert scale anchored by 1=”Strongly disagree”, 7=”Strongly agree”. The consequences of customer brand engagement on online social media platforms have been measured in terms of behavioral brand loyalty and word-of-mouth. The scale for behavioral brand loyalty contained two items relating to future purchase intentions (Chaudhuri & Holbrook, 2001). Word-of-mouth, which can also be defined as the intention to recommend the brand to others, has been measured with three items suggested by Zeithaml, Berry & Parasuraman (1996). In addition to the 62 mentioned items, there were also three control variables included in the questionnaire and measured by two self-constructed items each. These were goals, resources, and the perceived cost/benefit of interacting with the brand pages on Facebook specifically. These control variables have been included in the survey as the literature suggests that they can also be expected to influence how customers engage with brands (van Doorn et al., 2010). The two specific goals accounted for in the questionnaire were: 1) maximizing the consumption benefits (e.g. interacting with the brand on Facebook out of interest); 2) maximizing the relational benefits (e.g. becoming a member of a brand community). The resource items referred to the time available for browsing on Facebook fan pages and the effort that it takes. Finally, the perceived cost/benefit items were focusing on the respondents‟ perceived levels of enjoyment while browsing on Facebook fan pages and its value in comparison to the time and effort spent on it. A summary of all the 25
mentioned questionnaire items including the sources of reference and the resulting Cronbach‟s alpha for each scales are displayed in Table 2 below. Table 2: Construct measurement items, sources and scale reliabilities Measure/Source Items Antecedents Customer brand relationship quality related Involvement 1.In general I have a strong interest in [BN]5 (Beatty & Talpade, 2.[BN] is very important to me 1994) 3.[BN] matters a lot to me 4.I get bored when other people talk to me about [BN]*6 5.[BN] is relevant to me Satisfaction 6.Overall I am satisfied with [BN] (Gustafsson et al., 7.[BN] exceeds my expectations 2005) 8.The performance of [BN] is very close to the ideal brand in the product category Commitment 9.I am very loyal to [BN] (Aaker, Fournier, 10.I am willing to make small sacrifices in order to keep & Brasel, 2008) using the products of [BN] 11.I would be willing to postpone my purchase if the products of [BN] were temporarily unavailable 12.I would stick with [BN] even if it would let me down once or twice 13.I am so happy with [BN] that I no longer feel the need to watch out for other alternatives 14.I am likely to be using [BN] one year from now Trust 15.I trust [BN] (Chaudhuri & 16.I rely on [BN] Holbrook, 17.[BN] is an honest brand 2001) 18.[BN] is safe to use Online social media platform related Involvement 19.In general, I have a strong interest in Facebook (Beatty & Talpade, 20.Facebook is very important to me 1994) 21.Facebook matters a lot to me 22.I get bored when other people talk to me about Facebook* 23.Facebook is relevant to me Participation 24.I consider myself an active user of Facebook (Self-constructed) 25.I log on to Facebook everyday 26.I spend long periods of time on Facebook
The abbreviation BN stands for brand name, as different respondents have answered the questions with a different brand name in mind. 6 The items marked with “*” were reverse scored.
Ease of use 27.Learning to use Facebook is/was easy for me (Davis, 1989) 28.It is easy to get Facebook to do what I want it to do 29.It is clear and understandable how to use Facebook 30.Facebook is flexible to interact with 31.It is easy to become skillful at using Facebook 32.In general, I find Facebook easy to use Telepresence While browsing on Facebook.. (Kim & Biocca, 33… I feel like my mind is in a different world created 1997) by Facebook 34… I forget about the “real world” around me 35… I feel like my mind is more present in the “Facebook world” than the “real world” 36.After I am done browsing on Facebook, I feel like my mind comes back to the “real world” Customer brand engagement on online social media platforms Behavioral How often do you... (Self-constructed) 37.…visit the Facebook FP7 of [BN]? 38…notice the posts by [BN] in your news feed? 39…read posts by [BN]? 40…‟like‟ posts by [BN]? 41…comment on posts by [BN]? 42…share posts by [BN] with your friends? 43…post on the Facebook FP of [BN] yourself? (Cheung, Lee, & 44.I can continue browsing on the Facebook FP of [BN] Jin, 2011) for long periods at a time 45.I devote a lot of energy to the Facebook FP of [BN] Emotional 46.I am enthusiastic about the Facebook FP of [BN] (Cheung, Lee, & 47.The Facebook FP of [BN] inspires me Jin, 2011) 48.I find the Facebook FP of [BN] full of meaning and purpose 49.I am excited when browsing on and interacting with the Facebook FP of [BN] 50.I am interested in the Facebook FP of [BN] 51.I am proud of being a fan of [BN] Cognitive 52.Time flies when I am browsing on the Facebook FP of (Cheung, Lee, & [BN] Jin, 2011) 53.Browsing on the Facebook FP of [BN] is so absorbing that I forget about everything else 54.I am rarely distracted when browsing on the Facebook FP of [BN] 55.I am immersed in browsing on and interacting with the Facebook FP of [BN] 56.My mind is focused when browsing on the Facebook FP of [BN] 57.I pay a lot of attention to the Facebook FP of [BN]
The abbreviation FP stands for fan page.
Consequences Behavioral brand loyalty (Chaudhuri & Holbrook, 2001) Word-of-mouth (Zeithaml, Berry, & Parasuraman, 1996) Control variables Goals (Self-constructed)
Resources (Self-constructed) Perceived cost/benefit (Self-constructed)
58.I will buy [BN] the next time I buy food/drinks 59.I intend to keep purchasing [BN]
60.I say positive things about [BN] to other people 61.I often recommend [BN] to others 62.I encourage friends to buy [BN]
63.I browse on Facebook FPs because I am interested in the brands 64.I browse on Facebook FPs because I am interested in being a part of a brand community 65.I have enough time to browse on Facebook FPs 66.Browsing on Facebook FPs does not take too much effort 67.I enjoy browsing on Facebook FPs 68.I think that browsing on Facebook FPs is not worth the time and effort*
The coefficient reliability analysis revealed that all the scales consisting of more than two items exceeded the recommended Cronbach‟s alpha benchmark of 0.70 (Nunnally, 1978). However, the construct of behavioral brand loyalty measured by two items only has performed an internal consistency of 0.61, which is considered to be questionable (George & Mallery, 1999). In addition, the same happened to be the case with the three control variables that were also operationalized by two items each and did not meet the 0.70 benchmark. However, the nature of the Cronbach‟s alpha dictates that its value is determined not only by the mean of inter-item correlations, but also depends on the number of the items in the scale, which implies that the scales with fewer items will generally be expected to yield lower reliability coefficients. Therefore, the four underperforming two item scales were not eliminated and used further in the analysis.
3.3 Statistical analysis The approach applied in the data analysis of this study is called structural equation modeling, which is a powerful framework for estimating causal models and systems of simultaneous equations with measurement error. The structural model was established of seven key constructs: customer brand relationship related antecedents (CBRR), online 28
social media platform related antecedents (OSMPR), behavioral engagement (BEH), emotional engagement (EMO), cognitive engagement (COG), behavioral brand loyalty (BBL), and word-of-mouth (WOM).The construct of customer brand engagement on online social media platforms was split into three dimensions (behavioral, emotional and cognitive) in order to observe the effects of the two groups of antecedents on each of the engagement dimensions individually. Also, it was necessary to define which of the dimensions drive the selected customer consequences – behavioral brand loyalty and wordof-mouth. Thus, the three facets of behavioral, emotional and cognitive engagement, and the two potential consequences were modeled as first-order constructs and measured directly by multiple indicators. On the other hand, the suspected customer brand relationship related antecedents and the online social media platform related antecedents were modeled as second-order constructs, which were operationalized by four first-order dimensions each. That is, involvement (in the brand), trust, commitment and satisfaction served as indicators of the higher order construct referring to the customer brand relationship related antecedents, whereas participation, involvement, ease of use and telepresence were conceptualized as the dimensions of online social media platform related antecedents. Based on theoretical considerations and the types of indicators used each construct can be measured with either a reflective or a formative model. The reflective mode implies that changes in a construct are expected to be manifested in changes in all of its indicators, whereas in the formative mode a change in value of an indicator would translate into a change in the construct, regardless of the value of the other indicators (Henseler, Ringle, & Sinkovics, 2009). In this model all the first-order constructs as well as the response constructs relating to customer brand engagement and its consequences were measured with reflective items. In case of second-order constructs the second level of relationships from the individual first-order dimensions to the combined construct has to be considered as well. It was therefore decided that the dimensions of the customer brand relationship related antecedents would serve as indicators of a reflective measurement model, whereas in case of the second-order construct of online social media platform related antecedents a formative model was more adequate.
There are two types of statistical techniques for estimating structural equation models – covariance-based (e.g. LISREL) and variance-based (e.g. PLS) (Henseler et al., 2009). The method used in this study is PLS (partial least squares) path modeling, which can be viewed as a combination of principal component and multiple regression analysis. The main reasons behind choosing PLS relate to the highly favorable features of this technique (Henseler et al., 2009). PLS allows analyzing highly complex models without making the estimation problematic even when both formative and reflective measurement models are employed. Moreover, it can be used with a relatively small sample size and there are no distributional requirements. In this study only the data collected from respondents who belonged to the group of fans of at least one food or drink brand on Facebook could be used for testing the full model, which implies that the sample size equaled N1=112. Given that the model contained a total of 15 latent constructs (13 first-order constructs and 2 secondorder constructs), a sample of N1=112 was considered to be relatively small. In addition, some of the observations turned out to be skewed. Therefore, PLS was the more appropriate technique to apply in this study. Because of its flexible nature PLS path modeling is also generally suggested to be more adequate for causal modeling applications with no prior theoretical background. Thus, it goes well with the purposes of this study – developing and testing a conceptual model of customer brand engagement on online social media platforms. All data analysis were performed using a predictive analysis software SPSS and a path modeling software application SmartPLS (Ringle, Wende, & Will, 2005).
4. Results 4.1 Descriptive analysis Means and standard deviations were calculated for each of the constructs in order to compare the differences between the two groups of fans and non-fans (Table 3). The independent samples t-test revealed that the two groups showed significant differences in several aspects. With regards to customer brand relationship related antecedents, the group of fans was found to be more involved and expressed more trust in the brands than the nonfans. On the other hand, no significant differences were discovered in the levels of
satisfaction or commitment to the brands. Furthermore, the fans also showed a higher tendency of involvement, participation and telepresence in Facebook than the non-fans. Table 3: Means, standard deviations and results of t-test for equality of means (N1=112, N2=307) Fans (N1) Non-fans (N2) Construct Mean SD Mean SD Dimensions of customer brand relationship related antecedents 1. Involvement 3.85 1.38 3.09 1.48 2. Satisfaction 4.76 1.41 4.75 1.24 3. Commitment 4.09 1.42 3.94 1.36 4. Trust 4.66 1.31 4.20 1.35 Dimensions of online social media platform related antecedents 5. Involvement 4.05 1.38 3.71 1.48 6. Participation 5.31 1.41 4.98 1.62 7. Ease of use 5.49 1.04 5.33 1.07 8. Telepresence 3.06 1.50 2.67 1.42 9. Emotional engagement 2.90 1.30 n.a. n.a. 10. Behavioral engagement 2.54 1.04 n.a. n.a. 11. Cognitive engagement 2.33 1.19 n.a. n.a. 12. Behavioral brand loyalty 4.50 1.31 4.22 1.47 13. Word-of-mouth 4.63 1.55 3.92 1.67
t-value 4.77*** 0.09 1.02 3.11** 2.11* 2.06* 1.37 2.48* n.a. n.a. n.a. 1.75 3.93***
Note: n.a. = not applicable; SD = standard deviation; t-values were obtained by performing the independent samples t-test; *significant at