The Relationship between Customer Brand Engagement and Brand Loyalty for Local Car Brands in Malaysia

Brand loyalty has become a primary goal for a company in today's business environment due to intense market competition. This paper aims to examine the relationship between customer brand engagement and brand loyalty for local car brands in Malaysia. Brand loyalty is vital for local car manufacturers to ensure that customers remember their brands and will not switch to competing foreign brands. Therefore, companies need to engage with customers, resulting in brand loyalty and helping companies create a loyal customer— nevertheless, research confirming the main contribution of customer brand engagement to brand loyalty is still lacking. This paper has conducted a quantitative survey using online. 279 data were collected from the user of Malaysian local car brands; Proton and Perodua, and the data then were analysed using Smart PLS. The findings of this paper have indicated that customer brand engagement positively influences brand loyalty. Hence, brand loyalty has become an essential factor in sustaining market share, and it is much more cost-effective to retain the existing customer than attracting a new customer.


Introduction
The automotive industry is one of Malaysia's most important industrial sectors (Mabkhot, Shaari, & Salleh, 2017). It contributed 4.3% to the country's gross domestic product (GDP) in 2019 despite facing a high degree of change in megatrends and disruption (MIDA, 2020), as well as a highly saturated local auto market (The Asean Post, 2019). To date, the rising competition in the market has made 47 brands competing in the market. Out of 47 brands, 45 brands are from foreign brands, and only two brands from the local car brands; Perodua and Proton. For a record, the market share contribution of local car brands has declined from 52.6% in 2012 to 48.8% in 2018. Following to the market liberalization, several new players, including automotive firms from Japan, South Korea, Thailand and Germany, entered the small and mature Malaysian market. The resulting competitive pressures led to the local car manufacturer's market share dropping to 48.8% in 2018.
Globalisation and a competitive market directed the business to cope with numbers of challenges (Tabaku & Mersini, 2015). The advancement of Web 2.0 has given the customer the power to choose due to more choices to consider. Easy access for information on the brands is not limited to the products, also to the prices and services offered. The customer also became more empowered because of Web 2.0 (Berthon et al., 2012). The shift of power from business to the customer has raised the brand competition (Cuillierier, 2016), and the companies require to look for initiatives to build and strengthen brand loyalty continuously.
Further, brand loyalty can provide a competitive advantage to the company (Winters & Ha, 2012) as well as a positive effect on company performance (Pihl, 2013) such as sales growth, cost efficiency, profitability, and brand referrals (Bennur & Jin, 2017). Given these advantages, the importance of brand loyalty has broadly been stressed in the context of fostering repeat purchases in academic research (Mabkhot, Salleh, & Shaari, 2016). However, it is becoming increasingly difficult to ignore that many companies are having difficulties in retaining their customers. Retailers are required to develop strategies to improve and manage their customer experience to differentiate their business from their competitors (Greve, 2014;Klaus, 2015) because attracting new customers may increase the marketing cost compared to retaining existing customers (Han et al., 2018). Numerous research has been conducted to examine the key determinants of brand loyalty, given the importance of brand loyalty construct in marketing and customer behaviour (Gupta, Schivinski, & Brzozowska-Wos, 2017). Thus far, it remains a challenging issue due to the intense competition despite managers' efforts emphasising customer loyalty strategies (Nyadzayo & Khajehzadeh, 2016).
Today, brand loyalty has become the practitioner's primary goal by engaging and maintaining future connections with customers caused by increasing market competition (Rather & Sharma, 2016). Notably, attracting loyal customers is essential to ensure competitive business strategies where in this competitive world, brand loyalty is vital for the firms' profitability and existence (Jørgensen, Mathisen, & Pedersen, 2016). Hence, it is crucial to emphasise brand loyalty due to its ability to improve company growth and sales volume (Bennur & Jin, 2017). However, other variables may increase the level of brand loyalty, which is worthy for further investigation (Mabkhot et al., 2016). Also, the company's biggest challenge is how to make a customer loyal to the brand. Thus, measuring a customer's loyalty, which involves repeat purchases or recommending the brand to others, is a logical step (Mishra, Kesharwani, & Das, 2016). Therefore, brand engagement is crucial in marketing activities, and it goes further than selling products, but brand engagement means the customer has to be more passionate and involved about the brand (Wong & Merrilees, 2015).
Due to the intense competition among car brands, brand loyalty has become a strategically vital factor in thriving in the market sustainably. The rising brand competition left the car manufacturers without any other option but to actively engage with the customer to ensure customer retention and brand loyalty. The engagement term can be generalised as the company's offering to its' customers to full fill the customer needs and wants. In general, car manufacturer usually engages its' customer through car showroom and service centre. The introduction of the Internet has broadened the engagement through the company's website. In recent years, engagement has been viewed as a promising concept that is expected to contribute to customer behaviour, including brand loyalty (Hollebeek, Glynn, & Brodie, 2014).
Remarkably, engagement could help companies to create a loyal customer (Kandampully, Zhang, & Bilgihan, 2015). Nevertheless, research confirming the main contribution of customer brand engagement to brand loyalty is still lacking either exploratory or quantitative studies (Calder, Malthouse, & Maslowska, 2016;Dwivedi, 2015;So, King, Sparks, & Wang, 2016). Furthermore, customer brand engagement tends to focus on a specific context and adopting a narrow view of the loyalty concept (Fernandes & Moreira, 2019). Moreover, empirical research on customer brand engagement is scant Khan, Rahman, & Fatma, 2016). Based on this rationale, this paper addresses the following question: (1) What is the relationship between customer brand engagement and brand loyalty on local car brands in Malaysia? Therefore, by addressing the gaps mentioned above, it is pertinent to investigate whether customer brand engagement results in positive behavioural outcomes. Investigating this relationship may enhance the understanding of customer brand engagement and brand loyalty (Algharabat, Rana, Alalwan, Baabdullah, & Gupta, 2019;Rather, Tehseen, & Parrey, 2018). In general, this paper is expected to contribute to the body of knowledge of brand loyalty and extend the study of customer brand engagement into other industries (Islam & Rahman, 2017).

Literature Review Brand Loyalty
Dick and Basu (1994) have defined customer loyalty as a commitment to re-purchase the product again in the future. Oliver (1999) later has suggested loyalty as a commitment to rebuy or re-patronised preferred product/services consistently in the future. Meanwhile, Morrison and Crane (2007) refer brand loyalty as an emotional bond between the customer and the brand, which will turn the customer to advocate for the brand and subsequently promote the brand to others. Therefore, brand loyalty exists when a customer's attitude and intention focus on a specific brand. Also, loyalty has been described as a two-dimensional concept, namely behavioural (e.g., repeat purchase) and attitudinal (e.g., positive attitude towards the brand) (Dick & Basu, 1994).
The behavioural viewpoint measures loyalty as a static result of a dynamic process such as repeat purchase, frequency, and actual consumption. Meanwhile, the attitudinal viewpoint describes loyalty as an affection towards a brand and specifies trust, psychological attachment, and emotional commitment (Baloglu, 2002;Bowen & Shiang-Lih, 2001;Mechinda, Serirat, & Gulid, 2009;Petrick, 2004). Accordingly, Chaudhuri and Holbrook (2001) have agreed that brand loyalty consists of behavioural loyalty and attitudinal loyalty. Behavioural loyalty is defined as a customer's constant purchase from the same brand in the future, and attitudinal loyalty refers to a customer's commitment and attitude towards the brand (Chaudhuri & Holbrook, 2001). Therefore, brand loyalty is a continuous customer preference to purchase the same brand or a specific product or category of service in the future (Muchardie, Yudiana, & Gunawan, 2016;Schiffman & Wisenblit, 2015).
The importance of loyalty is documented in the marketing literature for many decades (Chinomona, 2016). Brand loyalty can affect customer decision-making process to purchase the same product (Nyadzayo & Khajehzadeh, 2016). Even though competitors have tempted the customer to switch to their brand, the customers insist on purchasing the same brand and refuse to switch (Keller, 2009). Hence, brand loyalty has brought an advantage to a company such as reduced marketing cost, bring in new customers, and higher trade leverage (Algesheimer, Dholakia, & Herrmann, 2005). Previous researcher has agreed that brand loyalty is beneficial to the company. Hence, to achieve brand loyalty, it requires that the customer must have confidence in the brand. Subsequently, the customer will become an advocate for the brand and will consistently promote the brand to another potential customer (Morrison & Crane, 2007). Meanwhile, Rather (2017Rather ( , 2018 concurred that brand loyalty signifies a positive attitude of a customer towards a brand or offering, in addition to the repeat buying behaviour. Therefore, brand loyalty is marked as the main focus of marketing strategies, particularly in a service provider company known to have a tight and undifferentiated market (Fournier & Yao, 1997).
Drawing from the work of Dick and Basu (1994), who propose that customer loyalty is the result of psychological processes and has behavioural manifestations both attitudinal and behavioural components, this paper attempts to enhance Dick and Basu (1994) framework by adding advocacy as the additional dimension of brand loyalty. This paper also proposes defining brand loyalty as customer intention to offer positive referral (Sashi, Brynildsen, & Bilgihan, 2019), commitment, attitude, and continuous purchase from the same brand in the future (Chaudhuri & Holbrook, 2001).

Customer Brand Engagement
The concept of engagement was first conceptualised by Kahn (1990) in the psychology context and defined engagement as "task behaviour that promotes connections to work and to others" which are expressed physically, cognitively, and emotionally, hence able to increase employee motivation as well as stimulate personal development. According to Bowden (2009), other authors also have emphasised that engagement is a psychological process arising from the combination of calculative commitment, development of trust, involvement, and finally affective commitment. Generally, engagement is based on interactions and subsequent from the interaction between the engagement subject and the engagement object. The engagement subject cited in the literature includes customers or consumers, while the engagement objects may include brands, offering, organisations, and organisational activities occurring beyond purchase (Abdul-Ghani, Hyde, & Marshall, 2018;Patterson, Yu, & Ruyter, 2006;van Doorn et al., 2010). There are many sub-forms discussed in engagement for example, customer engagement (Halaszovich & Nel, 2017), consumer engagement (Brodie, Hollebeek, Jurić, & Ilić, 2011;Vivek, Beatty, & Morgan, 2012), online brand management (Calder, Malthouse, & Schaedel, 2009;Mollen & Wilson, 2010), customer engagement behaviour (van Doorn et al., 2010) or community engagement (Algesheimer et al., 2005;Dessart et al., 2015;Dessart, Veloutsou, & Morgan-Thomas, 2016).
Moreover, recently, the term customer/consumer brand engagement has developed (Hollebeek, 2011a; revealing the concept's growing importance. Mostly, marketing studies' engagement focuses on the consumer (Dwivedi, 2015;Hollebeek & Chen, 2014; or customer (Vivek et al., 2012;Vries & Carlson, 2014) as the subject of the engagement. However, the object in engagement may vary, and commonly, it was studied in the contexts of brands (Dwivedi, 2015;. Bowden (2009) described engagement as a "psychological process that models the underlying mechanism by which customer loyalty forms for new customers of a service brand as well as the mechanisms by which loyalty may be maintained for repeat purchase customers of a service brand". Meanwhile, van Doorn et al. (2010) has focused on specific engagement behaviour and define it as "customer's behavioural manifestation that has a brand or company focus, beyond purchase, resulting from motivational drivers". Kumar et al. (2010) has agreed with the definition and urged that without the presence of customer purchases from the firm, the engagement is not complete. Furthermore, Mollen and Wilson (2010) have characterised online engagement 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." Whereas Brodie et al. (2013); Dwivedi (2015); Greve (2014); Hollebeek (2011a, 2011b); and  have urged that the definition of engagement comprises cognitive, emotional, and behavioural dimensions.
Therefore, customer brand engagement is known as a multidimensional construct where researchers have identified three dimensions of customer brand engagement: cognitive, affection, and activation (Hollebeek, 2011a;. However, it is essential to note that each dimension's use may vary depending on the study's context (Brodie et al., 2011). Hollebeek (2011a) defined customer brand engagement as "the level of an individual customer' motivational, brand-related and context-dependent state of mind characterised by specific levels of cognitive, emotional and behavioural activity in direct brand interactions." Later, Hollebeek (2011b) analysed the themes of customer brand engagement and defined customer brand engagement as "the level of a customer's cognitive, emotional and behavioural investment in specific brand interactions." Next,  in their study of social media setting has defined customer brand engagement as "a customer's positively valenced brand-related cognitive, emotional and behavioural activity during or related to focal customer/brand interaction." Meanwhile, Dwivedi (2015) describes customer brand engagement as "consumers' positive, fulfilling, brand-use-related state of mind that is characterised by vigor, dedication, and absorption." Dwivedi (2015) adapted the definition of customer brand engagement and its dimension from Schaufeli, Martinez, Pinto, Salanova and Bakker (2002). Dwivedi (2015) also has urged that the definition offered was correspond to the cognitive, emotional, and behavioural aspects of customer brand engagement as identified by Hollebeek (2011bHollebeek ( , 2011a and . After reviewing previous studies of engagement in marketing literature, this paper proposed to use the term customer brand engagement and define customer brand engagement as a customer's cognitive, affective, and behavioural while communicating with the specific brand, adapted from Hollebeek (2011b).

Relationship between Customer Brand Engagement and Brand Loyalty
Brand loyalty is a customer's continuous preference to purchase the same brand in the future (Muchardie et al., 2016). Brand loyalty provides a competitive advantage to the company (Winters & Ha, 2012) and, consequently given a positive company performance (Pihl, 2013). A loyal customer is willing to re-purchase other products from the same brand, bring new customers, and create a positive public image (Tu, Wang, & Chang, 2012). Hence, customer brand engagement affects brand loyalty (Sprott, Czellar, & Spangenberg, 2009). In addition to the repeat buying behaviour, brand loyalty also signifies the customer's positive attitude towards a brand or offering . Higgins (2006) described engagement as being engaged in being involved, occupied, and interested in something. Customer builds a positive attitude almost instantly if they are engaged with a brand compared to a customer who is not engaged with the brand (Harrigan, Evers, Miles, & Daly, 2017;So et al., 2014). A customer who does not take action against the providers by complaining directly to the providers usually described as a disengaged customer (Anderson et al., 2013). Disengagement is a passive, slightly-negative physiological orientation towards the relationship which occurs due to an attribute failure (Bowden, Gabbott, & Naumann, 2015).
Preliminary research of engagement in marketing literature indicated that engaged customers might show loyalty to the focal brand (Hollebeek, 2011a). Empirical research has recommended customer brand engagement positively influences brand loyalty (Algharabat et al., 2019;France, Merrilees, & Miller, 2016;Machado, Vacas-de-Carvalho, Azar, André, & dos Santos, 2019). Some studies have also shown that active engagement exerts more significant influence on the formation of favourable attitudes toward a specific brand versus passive or non-engagement (Kang, Tang, & Fiore, 2015;Leung & Jiang, 2018). Nevertheless, research confirming the main contribution of customer brand engagement to brand loyalty still lacks either exploratory or quantitative studies (Calder et al., 2016;Dwivedi, 2015;So et al., 2016). Therefore, it is worth to validate the relationship between customer brand engagement on brand loyalty (Fernandes & Moreira, 2019). Thus, the following hypothesis link between customer brand engagement and brand loyalty is proposed:

Methodology
A quantitative, hypotheses testing study using structural equation modelling (SEM) was deemed suitable to examine the relationship between customer brand engagement and brand loyalty. Specifically, Malaysia's local car brands namely Proton and Perodua were chosen for number of reasons. First, the automotive industry represents 4.3% of the GDP in Malaysia. Second, the industry is dominated by two local brands with the highest registration number year to year. Third, the study's constructs have been previously identified in various contexts.
In terms of measurement scale, the questionnaire was developed based on existing literature. The customer brand engagement constructs adapted the scale developed by Fernandes and Moreira (2019); ; and Shim (2012). 16 items were used to measure the construct. Meanwhile, brand loyalty constructs adapted the scale used by Fullerton (2011). Gecti and Zengin (2013); Halim (2006); and Moliner-Tena, Monferrer-Tirado and Estrada-Guillén (2019). This construct was measured using 13 items. All of the items were measured using 6-point Likert Scale scales ranging from 1 (Very Strongly Disagree) to 6 (Very Strongly Agree). This scale is chosen to prevent respondents from selecting the neutral point (Garland, 1991;Ong, Lee, & Ramayah, 2018;Si & Cullen, 1998). This paper has followed the method used by Yoo, Donthu, and Lee (2000) where only respondents who knew and familiar with the brand will complete the survey. Therefore, only respondents who presently own a car for more than three years will complete the survey. This paper has used an online survey as the data collection method due to the wide geographical area can be covered in the online survey, and the survey can be completed at the respondent's convenience (Sekaran & Bougie, 2016). With the emergence of information and communication technology, the use of the Internet can help researchers have access to an enormous population cost-effectively (Hooker & Zúñiga, 2017). This paper selects Surveymonkey.com as the survey authoring package. Although most of the authoring packages on the market offer similar functionalities, this survey package is selected because it is easy to manage (e.g., logic flow, skip function); allows multiple channels distribution; and allows the researcher to analyse the demographic data in a real-time format. The target respondents are user's of Malaysian local car brands and have used the car for four years and above. As illustrated in Table 0.1, the paper's sample consists of 107 males (38.4%) and 172 females (61.6%), with a total of 279 respondents. Most of the respondents were in the age bracket of 30 -39 years old (57.0%), followed by 40 -49 years old (20.4%). Meanwhile, 103 (36.9%) of the respondents earned between RM1,501 -RM3,500 a month, followed with 79 (28.3%) respondents earned between RM3,501 -RM5,500 a month, 46 (16.5%) earned between RM5,501 -RM7,501 a month, 19 (6.8%) earned between RM7,501 -RM9,500 a month, and 17 (6.1%) earned less than RM1,500 a month.

Common Method Variance
Additionally, this paper adopted Harman's Single-Factor Test using SPSS to test the common method variance (CMV) as recommended by Podsakoff, MacKenzie, Lee and Podsakoff (2003) and . All of the items used were entered into a Principal Component Analysis (PCA) with unrotated factor solution to identify if a single factor emerges or one general factor accounts for more than 50% of the covariation (Tehseen, Sajilan, Gadar, & Ramayah, 2017)

First Stage Assessment of Measurement Model
Assessment of reflective measurement models included internal consistency, indicator reliability/factor loadings, convergent validity, and discriminant validity. Since this paper uses a repeated indicator approach and a two-stage approach, the reflective measurement model is assessed accordingly. The recommended value of 0.708 indicates that the latent variable can explain at least 50% of the indicator's variance (Hair, Hult, Ringle, & Sarstedt, 2017). All of the indicators in this paper have shown the results above the recommended value except for CBECog6 and CBEBeh5. Therefore, adopted from the suggestion by Hair, Hult, et al. (2017), two indicators were deleted.
All of the first-order constructs were internally consistent and valid, and they were sufficiently different to guarantee the model's discriminant validity. Fornell and Larcker's criterion (1981) and Heterotrait-Monotrait (HTMT) ratios between the constructs (Henseler, Ringle, & Sarstedt, 2015) were examined to verify the discrimant validity. The constructs' composite reliability is greater than 0.7, which is satisfactory for this paper (Hair et al., 2017). AVE for all constructs is >0.5, which explains more than half of its indicator variance. Therefore, assessment from factor loading, composite reliability, and AVE shows that the construct meets the reliability and convergent validity requirement at this stage. Additionally, the data also indicates that all constructs exhibit sufficient or satisfactory discriminant validity (Fornell & Larcker, 1981), where the square root of AVE (diagonal) is larger than the correlations (offdiagonal) for all reflective constructs. The HTMT indicated that discriminant validity had been established as it shows that confidence interval does not show a value of 1 on any of the constructs. These show that the constructs are genuinely distinct from one another and uncorrelated.   After obtaining the latent variable scores used to estimate the second-order constructs, the final measurement model was verified. Table 0.3,   Table 0.4, Table 0.5, and Table 0.6 below show the assessment result of the factor loading, composite reliability, and convergent validity of the construct in this paper. The loadings of all indicators are uniformly above the recommended value of 0.708 (Hair et al., 2017). Simultaneously, the composite reliability is greater than 0.7, which is satisfactory for this paper (Hair et al., 2017). And, the AVE for all constructs are >0.5 which suggest adequate convergent

Assessment of the Structural Model
The next step in conducting the CFA is evaluating the structural model after the measurement model is established. Hence, this involves examining the model's predictive capabilities and the relationships between the construct (Hair et al., 2017). Following steps recommended by Ramayah et al. (2018), the assessment of the structural model in PLS-SEM involve:-1. The inner VIF (Variance Inflator Factor) value for customer brand engagement on brand loyalty is 2.089. The inner VIF value of this paper is below 3.3 threshold value, which interprets that the collinearity issue is not a concern (Diamantopoulos & Siguaw, 2006). 2. Path coefficient assessment using bootstrapping procedures for the hypothesised relationships shows that the relationship of Customer Brand Engagement and Brand Loyalty of β = 0.490, t-value = 8.327, and p-value = 0.000 is found significant. The acceptable level of t-value is >1.96, and p-value is <0.05. 3. The value of R² for Brand Loyalty is 0.607. It suggests that the endogenous variable in this paper (brand loyalty) explain 60.7% of the variance in customer brand engagement. The R² value of 0.607 is above the 0.26 value, as suggested by Cohen (1988), which indicates a substantial model. 4. The effect size (f²) of customer brand engagement on brand loyalty is 0.292. To measure the effect size, Cohen (1988) value of 0.02, 0.15 and 0.35 represent small, medium, and large effects. Therefore, the effect size of customer brand engagement on brand loyalty is medium. 5. Hair, Hult, et al. (2017) urged that the model has predictive relevance when Q² is greater than 0 whereas the model lacks predictive relevance when Q² is less than 0. Furthermore, the guidelines for evaluating the Q² value indicate that values of 0.02, 0.15, 0.35 represent small, medium, and large relevance for a specific endogenous latent variable (Hair et al., 2017). Q² value for brand loyalty for this paper is 0.474 (more than 0), indicating that the model has sufficient predictive relevant. Presented in Table 0.7 above is the summary of the structural model of this paper. In the conceptual model, this paper has hypothesised that customer brand engagement positively influences brand loyalty (H1). Consistent with expectation, the results reveal that the customer brand engagement is a vital variable showing evidence to support H1. The result of this paper is in agreement with findings by other studies that show customer brand engagement has a positive and significant impact on brand loyalty (Algharabat et al., 2019;France et al., 2016;Hollebeek, 2011a;Rather et al., 2018;Sprott et al., 2009). Besides, brand loyalty concept is an essential issue for most companies (Gecti & Zengin, 2013). Customer builds a positive attitude almost instantly if they are engaged with a brand compared to a customer who is not engaged with the brand (Harrigan et al., 2017;So et al., 2014). Thus, the brand's value increases as much as customer loyalty to the brand (Gecti & Zengin, 2013).
Past studies of engagement have been conducted in various contexts such as social media, brand communities, tourism, and customer/employee interactions. Hence, this paper has contributed to the body of knowledge of brand loyalty, particularly for local car brands in Malaysia. The findings of this paper confirmed that customer brand engagement has a positive and significant impact on brand loyalty. Also, the dimension of customer brand engagement was expanded consistent with Hollebeek (2011a) comments. The author has pointed out that the concept of engagement has been interpreted differently not only from academic approaches and methods but also from differences in opinion over its nature. This is due to the lack of empirical research on customer brand engagement Khan et al., 2016). Following to that, this paper incorporated additional items to the customer brand engagement dimensions, adapted from a previous researcher.

Conclusion
Brand loyalty has become an essential factor in sustaining market share. It is also difficult to ignore that the companies are having difficulties in retaining their customer. Due to that, it is much more cost-effective to retain the existing customer as compared to attracting a new customer. Hence, this paper has shed some insight into the relationship between the emerging concept of customer brand engagement and brand loyalty for local car brands in Malaysia. Consequently, this paper contributes to the ongoing research dialogue and discussion on brand loyalty (Mabkhot et al., 2016) and customer brand engagement (Fernandes & Moreira, 2019). It is in line with the study done by another researcher where the researcher recommended to extend and validate the study of engagement and brand loyalty into other industries (Algharabat et al., 2019;Islam & Rahman, 2017;Rather et al., 2018). Furthermore, this paper also responds to the Marketing Science Institute (2018, 2020) call for further research in these areas to investigate customer brand engagement and brand loyalty.
In this highly competitive market, managers are tested on how to retain profitable customers from switching to another brand. Therefore, managers can consider improving the understanding of customer brand engagement. Also, identifying the customers' needs and wants, which will later impact brand loyalty, by way of positive word-of-mouth that causes their relatives, colleagues and others to purchase the product from the same brand. The company should work to understand which types of engagement have the most significant impact on the growing customer value of engagers, as well as the customer value of those engaged by other customers.
Today, all major brands engage their stakeholders online and offline, whether through brand communities, fan clubs, or social media groups. Therefore, the company should prioritise relevant engagement for their businesses. As the learning from the experiments, a prioritisation and categorisation will be possible to improve strategic decisions on how to ensure brand loyalty as well as the best engagement method. For instance, brand communities have gained importance as a branding tool for marketers because they offer values difficult to capture elsewhere. Meanwhile, social commerce websites' reveals the latest information about the company's products and services; hence, it will encourage customers to further engage with the brand.
Furthermore, social media can facilitate an innovative channel for branding at the core of its customer engagement (e.g., Zheng et al. (2015)), consequently shifting the business practice and service. Also, social media such as Facebook and Instagram are among the most useful platforms to engage customers with brands at each stage of the customers' journey, which, in turn, allows for co-creation of value for the brand within customer networks. Alternatively, a car company can also create its car club under the company's management so that its members could discuss and share their experience in driving the car.

Limitation & Future Research
Being limited by time and cost, the data were collected online only and self-administered. Hence, respondents answered the survey at different times and places at their convenience. Besides, the findings are based on the demographic profile of the respondents. Thus, this paper's implication and conclusion may not be able to generalise beyond any other group of individuals. Therefore, the results cannot be generalised, as they reflect the selected age group and income of customers in the context of local car brands in Malaysia.
Given the limitations of this paper, several issues could be addressed in future research. First, since this paper solely used an online platform to collect the data, future studies might interest in collecting the data offline so that the questionnaire could be answered in front of the interviewer, and any difficulty or misunderstanding of the questions could be asked spontaneously. Second, the findings are based on the demographic profile of the respondents. Thus, future research may want to investigate other sets of demographic profiles such as comparative analysis between the younger and older customers and other individual characteristics, such as experience with the brand. Third, this paper may not provide full insight of the relationship between customer brand engagement and brand loyalty as in cross-sectional studies, various segments of the population are sampled, and data collected at a single moment (Zikmund, Carr, & Griffin, 2013). Hence, to better assess causality, future research could assess the current model using an experimental or longitudinal design as it involves a survey of respondents at a different time allowing analysis of response continuity and changes over time (Zikmund et al., 2013).