INTRODUCTION

Customer-brand engagement is emerging as a highly influential concept in modern marketing, with growing emphasis on the concept in both practical (MSI (Marketing Science Institute), 2014) and academic (Brodie et al, 2011) domains. Yet, customer-brand engagement is at an early stage of understanding. The seminal works of Brodie et al (2011) and Hollebeek (2011a, 2011b) provide the conceptual foundations for understanding customer-brand engagement as the customer’s brand-related psychological state and thus build a strong framework for model development.

Early conceptualisations of customer-brand engagement emphasise the relationship between the customer and the brand (Hollebeek, 2011b). Yet the role of brand management appears to be somewhat neglected in empirical development. Initial empirical investigations focus on the influence of the customer (De Vries and Carlson, 2014; Hollebeek et al, 2014; Dwivedi, 2015), while limited recent empirical development identifies the organisation’s role in customer-brand engagement (Wong and Merrilees, 2015). Thus, an opportunity arises to integrate the customer and brand perspectives in a comprehensive customer-brand engagement model, and to empirically test such a model.

This article contributes to the domain by proposing and empirically testing a firm-led platform for customer-brand engagement, as part of an integrated model. The firm-led platform drives customer-brand engagement in conjunction with customer-centred influences. This highlights ways in which the brand can actively ‘woo’ their customers.

The model identifies the strategic firm-led antecedents of brand quality and brand interactivity. In addition, the model draws on the evolving literature to identify the customer-centred influences of brand involvement (Hollebeek et al, 2014) and brand self-congruity (Sprott et al, 2009) as influential. Moreover, the model considers the consequences of customer-brand engagement, finding increased levels of both brand value and brand loyalty. The research bolsters the domain by reliably measuring the customer-brand engagement concept and empirically testing the proposed model, allowing practical application in the strategic marketing environment.

The paper initially examines the current knowledge in customer-brand engagement, by reviewing leading conceptualisations and theoretical models. A proposed model of customer-brand engagement is then presented and tested using structural equation modelling. The results are discussed, highlighting the contributions to theory and practice. The paper concludes with exciting research opportunities.

LITERATURE REVIEW

Customer-brand engagement is relatively embryonic and is predominantly led by conceptual reasoning. While knowledge is expanding, the debate continues around conceptualisation of the construct (Kumar et al, 2010); model development is minimal; and empirical testing is limited. The current conceptualisation of the customer-brand engagement construct is reviewed. Then, the literature review examines the empirical drivers of customer-brand engagement.

Customer-brand engagement construct

The customer-brand engagement field largely converges around the conceptualisation of customer-brand engagement as a psychological state, which encompass a proactive and interactive customer relationship (Brodie et al, 2011). Marketing conceptualisations are led by discussions of engagement as a process of building loyalty (Bowden, 2009b) and as customer behaviour (Van Doorn et al, 2010). Initial behavioural conceptualisations stimulated the discussion around customer-brand engagement. However, the recognition of engagement as a psychological state (Brodie et al, 2011) invigorated the domain and launched customer-brand engagement as a fundamental marketing concept in both academia and practice.

The leading view of customer-brand engagement involves ‘the level of an individual customer’s motivational, brand-related and context-dependent state of mind characterised by specific levels of cognitive, emotional and behavioural activity in direct brand interactions’ (Hollebeek, 2011a, p. 790, italics added). The engagement state is viewed as enduring and pervasive, characterised by a level of active immersion and passion (Hollebeek, 2011a). Brodie et al (2011, p. 257) distinguish the engagement concept from related constructs, stating that customer-brand engagement ‘encompasses a proactive, interactive customer relationship with a specific engagement object’. Customer-brand engagement is a psychological state that involves a customer’s pride and passion for the brand, arising from the strength of relationship between the customer and the brand (Brodie et al, 2011). The customer’s presence in their relationship with the brand is at the heart of the engagement concept (Patterson et al, 2006). The concept of engagement is therefore highly relational, but focuses specifically upon the customer’s psychological state of passion and immersion in the brand. This view of customer-brand engagement appears to have almost universal acceptance in the literature. Yet, there remains variation in the dimensionality of the concept, with two main approaches.

Hollebeek et al (2014) empirically examine customer-brand engagement, showing support as a three-dimensional construct of cognition, affect and activation. An alternative empirical view relates to engagement as formed by vigour, dedication and absorption (Dwivedi, 2015). Both perspectives have early empirical support and although the dimensionality of the concept varies, both Hollebeek et al (2014) and Dwivedi (2015) explore a consistent theme relating to the engagement between the customer and the brand, with scale items having an approximate alignment. The Hollebeek et al (2014) conceptualisation is adopted in this study as it more predominantly highlights the psychological nature of the concept and the dimensions explicitly align to the psychological characteristics of the engagement state. While adopting the conceptualisation of Hollebeek et al (2014), we make a minor adjustment. This study is anchored in the notion that customer-brand engagement is a psychological disposition, which is distinct from behavioural manifestations. An adaptation from the three dimensional conceptualisation to a two dimensional concept is required to precisely measure the engagement state, and exclude the behavioural dimension. The exclusion of the behavioural dimension supports the delineation of customer-brand engagement from the related concept of customer brand co-creation (France et al, 2015). As such, the two-component view of engagement, formed by the customer’s cognitive and affective state, is adopted in this study. The proposed conceptualisation of customer-brand engagement refines contemporary theory and provides a foundation for nomological examination in the context of a customer-brand engagement model.

Empirical drivers of customer-brand engagement

In the engagement domain, empirical model development is emerging. However, customer-brand engagement models are scant and are largely conceptual in nature. Several leading works progress knowledge in the field, identifying a range of antecedents to, and consequences of, customer-brand engagement. However, customer antecedent models are more commonly discussed, with exploration of the firm-led antecedents remaining underdeveloped.

Van Doorn et al (2010) consider the influence of both the customer and the firm in a conceptual model of customer engagement behaviour. De Vries and Carlson (2014) propose the concept of brand strength (which they define as brand involvement and brand self-congruity) as an antecedent to engagement in brand Facebook pages.

Hollebeek et al (2014) develop one of the first empirical models of customer-brand engagement, showing customer brand involvement as an antecedent. Further, Dwivedi (2015) measures product involvement and usage experience as customer antecedents of engagement, while organisational influences are measured directly to brand loyalty. Wong and Merrilees (2015) propose a managerial model of brand engagement identifying the organisation’s brand orientation as highly influential in driving brand equity and brand performance. Their model fills an important gap by considering the firm-led influence of engagement. However, the model resides in the strategic brand domain. There is scope to consider how the customer perceives the firm’s actions, to understand how brand strategy can be managed to drive customer-brand engagement. While a growing number of models are empirically measuring the role of the customer, or the organisation, there remains a need to measure the influence of both customer-centred influences and firm-led drivers of customer-brand engagement.

Summary

The current customer-brand engagement literature shows the growing knowledge of engagement between the customer and the brand. A strong focus is on understanding the customer-brand engagement construct and consideration of the drivers of the engagement state. The emphasis on firm-led antecedents remains underexplored. Generally, limited empirical measurement is evident, as is an apparent gap in empirical model development, which combines both customer-centred and firm-led drivers, to give a holistic understanding of customer-brand engagement. Overall, the early stage of knowledge remains largely conceptual, with an evident scope for increased empirical measurement.

NEW CONCEPTUALISATION OF DRIVERS AND CONSEQUENCES OF CUSTOMER-BRAND ENGAGEMENT

The development of research hypotheses forms a comprehensive model of customer-brand engagement, (shown in Figure 1). The model incorporates the influences on customer-brand engagement and explores the impact upon brand value and brand loyalty.

Figure 1
figure 1

An integrated model of customer-brand engagement.

Full size image

The integrated model of customer-brand engagement starts by identifying two important firm-led drivers, brand quality and brand interactivity. Critically, the identification of firm-led antecedents provides brand managers with a new platform from which they are able to drive engagement. This platform emphasises the role of strategic brand management for engagement success. The brand quality antecedent has a strong historical precedence as influential in building positive associations and loyalty for the brand (Cronin et al, 2000). Brand interactivity, a novel concept, is introduced into the engagement domain. While the notion of interactivity has strong influence in conceptual discussions of engagement (Patterson et al, 2006; Brodie et al, 2011; Hollebeek et al, 2014), no known model actualises the influence of brand interactivity upon customer-brand engagement.

Two customer-centred influences, brand involvement and brand self-congruity, have featured in the literature with strong support. Brand involvement is empirically linked to customer-brand engagement (De Vries and Carlson, 2014; Hollebeek et al, 2014; Dwivedi, 2015), whereas brand self-congruity has support in the related brand relationship domain (Dolich, 1969; Sirgy, 1985; De Vries and Carlson, 2014). The proposed model brings together these two variables to empirically measure their role in a holistic engagement context.

Finally, the integrated model of customer-brand engagement identifies brand value and brand loyalty as two important consequences of customer-brand engagement. The model shows a direct influence from engagement to increased perceptions of brand value and increased brand loyalty, emphasising the importance of the engagement concept in the modern marketing environment.

The model is pioneering, aggregating both the firm-led drivers and customer-centred influences of customer-brand engagement and demonstrating the influence on brand value and brand loyalty. Attention now turns to the detailed theoretical justification of each hypothesis.

FOUR DRIVERS OF CUSTOMER-BRAND ENGAGEMENT

Conceptual discussions of engagement propose a range of relevant customer-led antecedents, including brand involvement (Vivek et al, 2012) and commitment to the brand (Keh and Teo, 2001; Lacey and Morgan, 2009; Brodie et al, 2011). The engaged customer is known to have a high level of attachment to the brand (Bowden, 2009b; Zhou et al, 2012; Hung, 2014) and a positive psychological state towards the brand (Patterson et al, 2006). The customer is immersed in the brand experience (Minkiewicz et al, 2014) and has confidence, pride and passion for the brand (Kemp, 2015). This deep connection with the brand takes the customer beyond the transactional sphere (Van Doorn et al, 2010; Tregua et al, 2015).

Firm-led drivers of customer-brand engagement

Both the customer and the brand play an important role in developing and maintaining customer-brand engagement. Wong and Merrilees (2015, p. 586) state that ‘managers need to play a role in connecting to customers through the brand and not passively wait for customers to do all the work engaging with the brand. It has to be a two-sided approach’. Indeed, the relevance for management of customer-brand engagement depends on an actionable firm-led platform. There are potentially many tools available to the organisation to increase customer-brand engagement but two critical drivers are identified as influential. Specifically, perceived brand quality and brand interactivity are highlighted in the proposed model as important to initiating and facilitating customer-brand engagement. The notion of an engagement platform is broadly consistent with Baron and Warnaby (2011), Dholakia et al (2009) and Jaakkola and Alexander (2014).

The influence of brand quality on the customer-brand relationship is well established yet could benefit from inclusion in the engagement domain. Brand quality is identified as a powerful driver in marketing, influential in development of brand relationships (Cronin et al, 2000) and brand loyalty intentions (Dwivedi, 2015). Engagement literature identifies the traditional reliance on perceived quality as a predictor of customer behaviour, which is superseded by the engagement construct (Hollebeek, 2011a, 2011b). With this shift from quality to engagement, it appears that the influence of brand quality upon customer-brand engagement is less well considered in model testing. Quality is linked to the cognitive evaluations of a brand and to emotional value for the brand (Sweeney and Soutar, 2001), both of which contribute to customer-brand engagement.

No known empirical studies directly link perceived quality to customer-brand engagement. Nevertheless, the potential for such a link is guided by the well-developed notion that brand quality can be influential in building brand relationships. In order for a customer to be willing to emotionally and cognitively engage with the brand, they need to perceive a suitable level of quality. Perceived quality is the customer’s evaluation of the brands quality, and is influenced by the brands quality standards and brand position. Evaluation of the brand offering leads to a global assessment of overall quality, which may be influential in customer-brand engagement. It is expected, in Hypothesis 1, that the customer’s perception of brand quality will influence their level of customer-brand engagement:

Hypothesis 1:

  • Perceived brand quality will have a positive effect on customer-brand engagement.

While brand quality has a strong precedence in marketing, the concept of brand interactivity is more novel but potentially as relevant in the engagement domain. Communication with the brand informs the customer of the brands eagerness to interact with, and relate to, customers. Brand interactivity is conceived as the customer’s perception of the brand’s willingness and genuine desire for integration with the customer. The customer evaluates brand interactivity at two levels, namely the brand’s technical facilitation of interaction and the demonstration of a genuine desire for connectedness. The customer’s perception of brand interactivity becomes significant in the engagement domain, where interaction is identified as being central to customer-brand engagement (Jee and Lee, 2002; Brodie et al, 2011; Lawrence et al, 2013; De Vries and Carlson, 2014; Hollebeek et al, 2014). Customers are increasingly involved in interactions with the brand (Merz et al, 2009), and this increased brand interaction has positive effects for the customer-brand relationship (Jee and Lee, 2002). Strong support for the influence of interactivity is implied in conceptual discussions of engagement, where interactivity is often incorporated into discussions of the engagement concept (Patterson et al, 2006; Brodie et al, 2011). In additional, related literature identifies interactivity as fundamental to customer-brand engagement (Lawrence et al, 2013; De Vries and Carlson, 2014).

Qualitative investigation shows that customers engage with brands, which listen to what they are saying and participate in two-way conversations (Shao et al, 2015). Case study analysis demonstrates that a brand which is perceived to have a high level of interactivity, presents itself as open to more personalised relationships (Sawhney et al, 2005). When the customer perceives the brand as interactive, they feel welcomed and encouraged to engage with the brand. The customer feels valued by the brand and trust is formed, strengthening the relationship (Merrilees and Fry, 2003) and building the engagement state.

While interactivity is consistently identified as important to customer-brand engagement, there is little previous attempt to measure brand interactivity as an explicit antecedent. In fact, measurement of the overall perceived interactive nature of the brand appears to be a relatively novel concept. Although there is no apparent evidence of empirical measurement of brand interactivity as an antecedent, strong conceptual foundations and theoretical reasoning provide confidence that brand interactivity could play a significant role in customer-brand engagement. It is expected, in Hypothesis 2, that the customer’s perception of brand interactivity will have a direct effect on their level of customer-brand engagement:

Hypothesis 2:

  • Brand interactivity will have a positive effect on customer-brand engagement.

Customer-centred influences of customer-brand engagement

While the literature discusses many diverse customer-centred antecedents to engagement, two fundamental antecedents are highlighted as commonly recognised in the marketing literature. Namely, brand involvement and brand self-congruity are identified as customer-centred influences on customer-brand engagement. Brand involvement is widely acknowledged as influential to customer-brand engagement (Bowden, 2009a; Hollebeek, 2011a, 2011b; Vivek et al, 2012; De Vries and Carlson, 2014; Hollebeek et al, 2014). The (more mature) employee engagement literature supports involvement as an essential characteristic of engagement (Saks, 2006). This is applied in the marketing context, where the customer’s interest and personal relevance is a required condition for engagement (Bowden, 2009a). Such involvement provokes a psychological commitment to engage with the brand (Bowden, 2009b).

Empirical measurement demonstrates the influence of brand involvement upon customer-brand engagement (Hollebeek et al, 2014), based on the three-dimensional view. A further empirical study provides partial support for the relationship (Dwivedi, 2015), though using an allied concept, product involvement instead of brand involvement. Theoretical and emerging empirical support provides a basis for the relationship. It is reasoned, in Hypothesis 3, that the involved customer possesses a high level of interest and sees personal relevance in the brand (De Vries and Carlson, 2014), which drives the customer to develop passion in the brand and to immerse themselves in their brand experiences, as engaged customers do, thus:

Hypothesis 3:

  • Brand involvement will have a positive effect on customer-brand engagement.

Empirical measurement of the proposed customer-brand engagement model has potential to provide verification of the influence of brand involvement on the psychological concept of engagement, without the influences from behavioural dimensions. In addition, the integrated model of customer-brand engagement measures the influence of brand involvement in conjunction with other relevant antecedents and consequences to provide further insight into the relative importance of brand involvement in the engagement domain.

A second customer-centred influence of importance is brand self-congruity. The influence of brand self-congruity on brand relationships has long been supported in marketing literature (Grubb and Hupp, 1968; Dolich, 1969). Customers participate and contribute in brand settings in order to express their self-identity (Lloyd and Woodside, 2013; Wirtz et al, 2013). The image an individual holds of their personality can align with the perceived personality of a brand, resulting in brand-self congruity (Sirgy, 1985). A high level of alignment between the brand and self-concept contributes to customer-brand engagement (De Vries and Carlson, 2014).

The broad link between brand self-congruity and customer-brand relationships receives widespread empirical support (Sprott et al, 2009; Bergkvist and Bech-Larsen, 2010; Ramaseshan and Stein, 2014; Wallace et al, 2014) but is yet to be fully established in the engagement context. Related literature links brand identification with concepts such as brand loyalty (Bergkvist and Bech-Larsen, 2010), while brand self-connection is identified as an antecedent to brand attachment (Hung, 2014). Sprott et al (2009) conduct empirical research demonstrating the propensity of certain consumers to engage with their favourite brands in order to express their identity. While closely related, the distinction is made that their research considers an individual’s likelihood to include brands in self-concept, rather than an assessment of a specific connection with the brand, which forms the focus of this study.

The direct relationship between brand self-congruity and customer-brand engagement is not yet conclusively tested. De Vries and Carlson (2014) show partial empirical support for congruity as a driver of engagement, while Hollebeek et al (2014) demonstrate empirical support for congruity as a consequence. The differential findings suggest ambiguity in the domain and make this a contestable area for further examination. The strong conceptual support in related studies bolsters the expectation that when the customer perceives a high level of alignment between themselves and the brand, they are more likely to develop a sense of belonging to, and engagement with, the brand and we therefore argue that congruity acts as a driver which sparks the customer’s passion and immersion in the brand. Conversely, when a customer does not identify, they are unlikely to engage emotionally and cognitively with the brand. This leads to Hypothesis 4, that customers with high brand self-congruity will present a higher level of customer-brand engagement:

Hypothesis 4:

  • Brand self-congruity will have a positive effect on customer-brand engagement.

While important insights are drawn from current customer-brand engagement models, no known models explore both the customer-centred and firm-led antecedents to determine the relative influence of the customer and the organisation as drivers of customer-brand engagement. Further, in addition to measuring the influence of antecedents, it is useful to also determine the consequences of customer-brand engagement.

Consequences of customer-brand engagement

In this developmental phase of customer-brand engagement, confusion remains around the consequences of the engagement state (Van Doorn et al, 2010). An empirical model, which examines the role of specific consequences of customer-brand engagement has potential to add significant value in the domain (MSI, 2014).

The notion that ‘the customer is always a co-creator of value’ (Vargo and Lusch, 2008, p.8) has strong conceptual support (Vargo and Lusch, 2004; Payne et al, 2008; Merz et al, 2009; Hollebeek, 2011b), suggesting that the customer has potential to directly affect assessments of brand value. Yet, few studies empirically measure this phenomenon. The interactive and involving nature of customer-brand engagement is expected to impact upon the customer’s assessment of brand value. Extant conceptual literature supports the direct influence of engagement upon value (Ramaswamy and Ozcan, 2013; France et al, 2015; Tregua et al, 2015). When a customer feels passionate and immersed in the brand, they are expected to perceive an increased level of value from the brand. The more engaged a customer is, the more likely that they will derive value from (and create value for) the brand, thus Hypothesis 5:

Hypothesis 5:

  • Customer-brand engagement will have a positive effect on brand value.

Understanding the influence of customer-brand engagement upon evaluations of the brand provides important insight. In addition, it is useful to understand the effect upon customer behavioural intentions. The empirical models of Hollebeek et al (2014) and Dwivedi (2015) both identify consequences of customer-brand engagement relating to behaviour intentions, specifically brand usage intent and loyalty intent (respectively). The influence of customer-brand engagement upon loyalty is a central theme within the literature. Conceptual model discussions support the notion that customer-brand engagement will act as a driver to customer behaviour (France et al, 2015). Several studies conceptualise the links from the highly connected and passionate customer to affective and brand loyalty (Van Doorn et al, 2010; Vivek et al, 2012; Kemp, 2015). Brand loyalty is increasingly identified as a consequence of customer-brand engagement with engaged customers showing a more active brand voice (Bijmolt et al, 2010), providing feedback to the brand (Kumar et al, 2010) and advocating the brand to others (Vivek et al, 2012). The emotional commitment and connection of the highly engaged customer is expected to be influential in their loyalty behaviour, and Hypothesis 6 therefore states that customer-brand engagement will lead to increased brand loyalty:

Hypothesis 6:

  • Customer-brand engagement will have a positive effect on brand loyalty.

The theoretical foundations for the hypotheses have been developed by drawing upon, and expanding, existing literature in the domain. Importantly, the proposed model and related hypotheses address some of the key gaps in the literature by combining customer-centred influences and firm-led drivers, and incorporating relevant consequences of customer-brand engagement for empirical examination.

METHODOLOGY

The study aims to build from the existing conceptual knowledge and emerging empirical studies in the field by developing and measuring an integrated model of customer-brand engagement and the related antecedents and consequences. A quantitative research design is used, with self-administered questionnaires for gathering data for testing.

The research questionnaire was developed with a focus on ease of use and attainment of strong completion rates. A pre-test pilot study was conducted using a convenience sample of 51 postgraduate university students. The pilot test allowed for identification of errors and preliminary scale optimisation, with final purification performed with the full sample (see Table 1 for factor loading results). The final questionnaire consisted predominantly of 5-point Likert scale responses relating to the research hypotheses. Scale items were selected based on the desired constructs of measurement and where possible established and tested scales were used, or adapted, including brand involvement (De Vries and Carlson, 2014), brand value (Cronin et al, 2000), brand engagement (Patterson et al, 2006; Schaufeli and Bakker, 2010; Hollebeek, 2011b; Dwivedi, 2015), brand quality (Cronin et al, 2000), brand interactivity (Merrilees and Fry, 2003; Labrecque, 2014), brand loyalty (Yi and Gong, 2013) and brand-self congruity (Sirgy, 1982; De Vries and Carlson, 2014).

Table 1 Measures of model constructs: Scale items and reliability
Full size table

The goal of this research is to test the hypothesised model, rather than describing characteristics of the general population or a single specific brand. As such, a non-probability convenience sampling method is a justifiable approach for this study, utilised in much of the emerging engagement research (Sprott et al, 2009; De Vries and Carlson, 2014; Hollebeek et al, 2014; Dwivedi, 2015). The questionnaire was distributed using a multiple location sampling framework, which included customers of selected small service brands (including fashion and homewares retailers, hairdressers and cafes), as well as a university sample (including a mix of undergraduate and postgraduate students and staff). This approach provided diversity in respondent characteristics, including age (18–34=49 per cent; 35–54=34 per cent; 55+=15 per cent), gender (72 per cent female) and income (<AU$50 000=57 per cent; $50 000–$100 000=26 per cent; >$100 000=11 per cent). The questionnaire was administered using paper-based and online methods. The split distribution strategy provided access to broad groups without influencing the data (Huang, 2006; Lin and Van Ryzin, 2012).

Customer respondents were asked to complete a survey for the brand they were patronising. Alternatively, the university sample was asked to participate by nominating a brand that they like and use as a customer, and then answer the same questions in relation to that brand. This strategy increased participation and completion rates by appealing to the respondent’s brand of interest. In addition, this method aimed to increase the representation of engaged customers, who are a relatively small portion of customers. For example, a Gallup study shows engagement levels to be as low as 12–15 per cent (Yu et al, 2014). Therefore, to attract a sufficient level of engagement in the sample, oversampling is necessary. While the expectation is that respondents may be more positively disposed to their selected brands than the patronised brands, the results contradict this, with the patronised brands showing a higher level of engagement (4.0) than the nominated brands (3.6). Overall results indicate a relatively neutral response, with engagement scores ranging from 1 to 5, with a mean of 3.7/5 and 12 per cent respondents scoring negatively, thus mitigating concerns for strong positive engagement bias.

Data was collected from 358 respondents and entered in SPSS, where examination of missing and erroneous data was conducted and 32 cases were removed because of majority missing data. Data was prepared for IBM AMOS, using a partial disaggregation method. Partial disaggregation provides meaningful analysis using composites to minimise random error (Dabholkar et al, 1996). The parcelling technique provides more parsimonious parameter estimates and results in better fitting models and improved interpretation (Bagozzi and Heatherton, 1994; Little et al, 2002).

DATA ANALYSIS AND RESULTS

A final complete data set of 326 was explored for suitability before analysis, exceeding requirements of sample size (Bentler and Chou, 1987; Tinsley and Tinsley, 1987). The method of survey administration was well-distributed between paper-based (62 per cent) and online (38 per cent) platforms. Interestingly, respondents selected a well-dispersed mix of brands across service (62 per cent) and product (38 per cent) categories. Results of KMO analysis (0.94) assure the data adequacy, exceeding the 0.60 requirement (Kaiser, 1970).

Reliability and validity results

It is essential to establish the reliability and validity of each scale before model testing. Therefore, each scale underwent a consistent process of factor analysis. As a focus in this study, the customer-brand engagement scale was developed to reflect the conceptualisation of engagement as a psychological state, thus excluding behavioural assessments. The new engagement scale, and all scales, loaded with a single dimension. Further, coefficient factor loadings were appropriately greater than 0.50 (Hair et al, 1998) (see Table 1). The final unidimensional scales were examined for internal reliability, using Cronbach’s α coefficient, where all coefficients exceeded the 0.70 requirement, with majority showing a very good level, >0.80 (De Vellis, 2012) (see Table 1).

Convergent validity is established by assuring all constructs maintain an average variance extracted (AVE) greater than 0.50 (MacKenzie et al, 2011; De Vellis, 2012), see Table 1. Discriminant validity is confirmed with the square root of the AVE exceeding the values of all absolute standardised correlations of the construct (Fornell and Larcker, 1981), (Table 2). Further, all correlations are significant, showing a sufficient relationship between the constructs. Results of testing are strong, lending support for the validity and reliability of all constructs within the study.

Table 2 Discriminant validity and correlations
Full size table

Model analysis results

Structural equation modelling is used for data analysis. Testing of the hypothesised measurement model is the first step, of the two-step approach, in assuring accuracy of the model (Fornell & Larcker, 1981). The measurement model produced a non-significant chi-squared result (χ2=66.59; DF=56; P=0.16). In addition, the normed chi-square (χ2/DF=1.19) falls appropriately below the cut-off of 3 (Carmines and McIver, 1983). Goodness of fit indices performed strongly (GFI=0.97, TLI=0.99 and CFI=0.99), showing acceptable fit to data (Hu and Bentler, 1999). Examination of model misspecification attained acceptable (low) levels, (SRMR=0.017, RMSEA=0.025). Thus, the measurement model elicits strong model fit. The second step in model assessment is the estimation of the structural model. The structural model elicits a significant chi-squared result (χ2=138.88; DF=63; P=0.00). However, other measures suggest acceptable fit, including a satisfactory normed chi-squared result (χ2/DF=2.20). Goodness of fit indices show strong model fit (GFI=0.94, TLI=0.97 and CFI=0.98). Acceptable (low) measures of model misspecification were also attained (SRMR=0.044 and RMSEA=0.063) and the entire confidence range for RMSEA is 0.049 to 0.077, suitably less than the established 0.08 threshold (Byrne, 2010). Results of model testing therefore establish a structural model of strong fit with the data.

Hypothesis testing

The influence of customer-centred and firm-led antecedents to customer-brand engagement is shown to be meaningful, with all 6 hypothesised relationships being supported; see Table 3. Importantly, four direct antecedents to customer-brand engagement are identified as statistically significant, showing the effect of both customer-centred influences and firm-led drivers. The number of highly significant supported hypotheses suggests that a very robust model of customer-brand engagement has been developed and demonstrated.

Table 3 Customer-brand engagement hypothesis testing
Full size table

The result of the consequences of customer-brand engagement also performed strongly in the model. Results support the influence of customer-brand engagement upon brand value, thus provide empirical evidence that brand value is heavily influenced by the customer-brand engagement phenomenon. In addition, customer-brand engagement is shown to have a direct influence on brand loyalty. Thus, while antecedents to customer-brand engagement are established, evidence of the consequences of customer-brand engagement are equally as strong.

Hypothesis testing yields strong results for the interactive customer-brand engagement model, with 77 per cent of the variance of customer-brand engagement explained by the model. Examination of hypotheses and model testing shows that customer-brand engagement is influenced by both customer-centred and firm-led constructs. Customer-centred influences (brand involvement and brand self-congruity) dominate the direct influence upon customer-brand engagement, with firm-led drivers (brand quality and brand interactivity) contributing directly to the construct but to a lesser extent.

Further results

To increase the robustness of results, four sets of additional results have been undertaken. These four areas cover (a) the direct, indirect and total effects of the antecedent variables; (b) control for demographic variables; (c) model mediation; and finally (d) multi-group comparison of product versus service brands.

First, elaborating the core results, we can detail the direct, indirect and total effects of each antecedent variable. The full results are shown in Table 4.

Table 4 Direct, indirect and total effects of antecedents of customer-brand engagement
Full size table

Second, further tests examine the sensitivity of results to demographic variables. Multiple regression was conducted with and without the demographic variables of age, gender and income. The β co-efficients were insensitive to demographic variables and in fact, were not statistically significant.

Third, using the Baron and Kenny four-step approach (Mallinckrodt et al, 2006), mediation analysis was conducted. Results indicate that engagement plays a mediation role. However, there is partial rather than full mediation, in particular quality continued to have a direct effect on value.

Finally, multi-group invariance analysis was conducted to explore the suitability of the model in both product and service contexts. This analysis compares the difference in CFI between the constrained (constraining the equality of co-efficients across both product and service brands) and unconstrained models, providing evidence of non-variance (∆CFI<0.01) (Cheung and Rensvold, 2002). Assessment of structural model invariance shows equivalence of groups at the model level (Δχ2(18)=13.81, P>0.05) (Byrne, 2004). Examination of invariance at a path level supports model invariance, with all path models reporting χ2 below the significance threshold (P=0.05). In other words, there were no substantial differences in the integrated customer-brand engagement model between product and service brands.

DISCUSSION

Results of scale and model testing provide some exciting findings for the domain. While discussions around the conceptualisation of customer-brand engagement show support for a consistent view of the concept, the dimensionality remains much less certain. Previous studies have supported different components of customer-brand engagement. A multi-dimensional construct was allowed for in this study, yet empirical results support only a single dimension. The cognitive and affective characteristics are indistinguishable as separate dimensions but strongly contribute to the engagement construct as a single dimension. This casts doubt on the number of dimensions to customer-brand engagement and becomes a contestable debate in the domain. From our perspective, there seems to be a high amount of collinearity (in these and previous results), to the extent that it is possible there is only one real dimension of customer-brand engagement. More research is needed.

The integrated model of customer-brand engagement brings into discussion two types of drivers of customer-brand engagement, the firm-led platform and the customer-centred influences. To clarify, firm-led means that the initial responsibility and initial action are taken by the firm, which then begins the dialogue with, and interpretation by, the customer. The antecedents of brand quality and brand interactivity are demonstrated as providing a platform from which brand management may influence the customer’s level of brand engagement. While two relevant antecedents are empirically demonstrated as influential, the identification of the management platform is insightful in itself. Brand managers play an important role in driving customer-brand engagement, with a potential suite of tools, which may be useful in the platform for engagement. Further, the customer-centred influences, brand involvement and brand self-congruity, are strong in their role in driving engagement. This finding certainly supports emerging models, which demonstrate the significant role of the customer.

Beyond the influence of engagement drivers, the model addresses an emerging area of interest, namely the consequences of customer-brand engagement. A growing body of literature conceptualises the active role of the customer in the value creation process. However, measurement of the impact is limited to date. Notably, the current customer-brand engagement model empirically demonstrates that the customer plays an active role in the value creation process, adding empirical support to the customer-centric forum. The hypothesised consequence of brand loyalty is demonstrated, thus reinforcing the role of engagement as central to the success of the brand and influential in customer behaviour.

The applicability of the interactive model of customer-brand engagement across both product and service brands is novel in the emerging brand engagement domain. No known studies have developed and examined a model across various brand types, yet the model shows the consistent influence of the firm-led platform, and the customer-centred influences, across both service and product brands.

Overall, the customer-brand engagement model enhances the previous understanding in the branding domain. The model supports existing studies of customer-brand engagement and extends knowledge by incorporating additional influences.

Contributions

The paper offers significant contributions to both the academic and practical domains by building on the strong conceptual base of the customer-brand engagement field. In developing a sound nomological network, assurance is provided about the conceptualisation of customer-brand engagement as an individual’s motivational state towards the brand, characterised by cognitive immersion and emotional passion for the brand [adapted from Hollebeek et al (2014)]. The results support the proposal of customer-brand engagement as a psychological state, distinct from customer behaviour, and represented by a unidimensional construct.

Beyond conceptual clarity, model development and empirical examination provide new evidence for the antecedents and consequences of customer-brand engagement. From a managerial perspective, the customer-brand engagement model identifies a firm-led platform with direct, and indirect, influences on customer-brand engagement. Managerial relevance derives from the identification of the actionable influences of brand quality and brand interactivity on customer-brand engagement. The model expressly recognises the role brands may play in driving engagement, and equips practitioners with specific tools, such as the notion of an engagement platform, to stimulate customer-brand engagement.

Importantly, brand quality is identified as a firm-led driver of significance to customer-brand engagement. This core variable seems to have been overlooked, both conceptually and empirically as a driver of customer-brand engagement. Yet, customers are inclined to engage with brands that provide a desirable level of quality. The brand plays an important role in developing the level of quality from which the customers form their evaluation. Although all variables in the customer-brand engagement model could be highlighted, the brand quality variable is noteworthy.

Brand interactivity is emphasised as a highly relevant influence on customer-brand engagement. The conceptualisation of brand interactivity, as the customer’s perception of the brand’s willingness and genuine desire for integration, provides new understanding. The customer’s perception of brand interactivity is significant to engagement in the modern marketing environment, where interaction plays an increasingly important role in the customer-brand relationship.

Previous models have empirically demonstrated the influence of customer-brand engagement on brand usage intent (Hollebeek et al, 2014), loyalty intentions (Dwivedi, 2015) and brand performance (Wong and Merrilees, 2015). Building on this strong base, the proposed model contributes by measuring the consequences of brand value and brand loyalty. Understanding the influence of engagement upon the customer’s perception of value provides real insight for brand managers and cements customer-brand engagement as a concept of central importance to brand success. In addition, the consequence of brand loyalty demonstrates real benefit for the brand and provides a link to increased brand performance. Multi-group model testing reveals the wide applicability of the potential usefulness of the model across a range of brand types.

CONCLUSIONS, LIMITATIONS AND FUTURE RESEARCH

Overall, the paper provides a new perspective in conceptualising customer-brand engagement, as well as identifying pertinent customer-centred and firm-led antecedents. The results buttress the work in the domain and provide increased managerial relevance. The final contribution lies in the empirical demonstration of the effect of customer-brand engagement on brand value perceptions and brand loyalty intentions. Yet, opportunity exists for further empirical demonstration of customer-brand engagement on actual customer behaviour.

A limitation of the research lies in the cross-sectional nature of the study, which investigates a single cultural context and a specific point in time. The study was conducted in Australia, a typical, developed Western culture. It may be interesting and useful to explore this research in other countries. Potentially, the influence of culture may have a role to play in the customer’s propensity to engage, as could economic influences and the current dynamics of the consumer-brand relationship trend.

A model is developed, which is applied across a broad brand setting and demonstrated to be relevant for both product and service brands. However, in specific brand and category types, there is potential for other variables to also be of importance. For example, the allied concept of brand relational authenticity (Ilicic and Webster, 2014) may add further insight as an antecedent to customer-brand engagement, especially in services. Fast moving goods offer special challenges and deserve dedicated attention.

Future research could use alternative research designs to target either a more purposeful or generalisable population for comparison of results. An additional area for research arises from the distinction between the psychological state of engagement and consequential behavioural manifestations developed in this article. Future research which explores this phenomenon may provide additional insight. Research could consider which specific behaviours are motivated by high levels of customer-brand engagement and what worth these behaviours have for the brand. By offering the refined conceptualisation of customer-brand engagement which is distinct to behaviour and accurately modelling antecedents and consequences of customer-brand engagement, new opportunities relating to the measurement and investigation of related variables in this area could prove bountiful.