Driving Relationship Marketing Tactics towards Customer Loyalty Building: The Mediating Effect of Commitment and Trust

One of the main targets in relationship marketing is to ensure continued retention of profitable customers by building on and sustaining a stable long-term relationship that will subsequently result in customer loyalty behavior. This study focuses on relationship marketing tactics (RMT) dimensions of brand reputation and trust and how it influences customer loyalty, as well as examining the mediating effects of relationship quality (RQ) of commitment and trust within the context of a competitive mobile telecommunication service industry. Using quota sampling ap proach, a total of 382 survey questionnaires’ usable feedback were collected from targeted respondents for data analyses. A quantitative analytical method was applied via SPPS version 20.0 and a PLS-SEM software procedure using Smart PLS version 3.0 with bootstrapping analysis to test the developed hypotheses. Hypothetical results revealed differential effects of RMT on customer loyalty. Brand reputation significantly influenced both commitment and trust. Alternative attractiveness also significantly affected commitment and trust. The mediating effects of RQ further revealed the continued importance of commitment and trust as key factors in developing a successful customer loyalty. Several theoretical and managerial implications are further discussed.


Introduction
In marketing, service providers' continuous need to better understand their customers has ascended as a result of numerous factors such as aggressive market competition, dynamic and complex environment, consumers' increased expectation, and rapid innovation in the service industry (Sheth, 2017;Kazemifar & Shayesteh, 2015). Nowadays, service providers are likely to explore more about their customers in an attempt to retain them and create customer loyalty. This in turn, will further increase their business profits. Since the concept of relationship that a firm's brand reputation is projected by the firm in the mind of its customers based on their experience, word of mouth and advertising. Other studies have also discovered brand reputation as a related factor that helps to build customer loyalty, trust and commitment from the customers with their existing service providers (Segupta et. al, 2015;Ryan & Cassidy, 2018;Foroudi, 2019). However, these studies did not contrast the relative impact of brand reputation with other relationship marketing tactics; even though they have provided empirical evidence linking brand reputation or reputation-related beliefs to attitude toward switching (Peng & Wang, 2007). Therefore, a strong brand reputation is supposed to meet customer's expectation and present more benefits to the customer, as a result leading to customer trust and commitment. Following that, this study attempts to measure the influence of brand reputation on the Malaysian mobile telecommunication sector by proposing the following hypotheses: H1a: Brand Reputation has a positive significant influence on Commitment in the mobile telecommunication industry. H1b: Brand Reputation has a positive significant influence on Trust in the mobile telecommunication industry.

Relationship between Alternative Attractiveness and Relationship Quality
Mannan, Mohiuddin, Chowdhury & Sarker (2017), claimed that the higher the customers were alerted about the alternative attractiveness of competing service providers, the higher the likelihood that they will choose not to be loyal to their current service providers due to the availability of other options. This argument is widely held by many researchers (Lee, Ou, & Choi, 2021;Bashir, 2011;Yen & Horng, 2010;Bansal et al, 2005). Nevertheless, each researcher pointed out different views of the relationship between alternative attractiveness and customer loyalty. Lee et al (2021); Sharma & Patterson (2000) stated that a weak alternative attractiveness offering can affect the likelihood of a customer to either stay loyal or to switch their service providers. Customers may decide to end the relationship and turn to a new service provider if they recognize the alternative to be attractive due to the availability of superior service, the proximity of location, the availability of a full range service, lower price perception or the promise of high financial returns (trustworthy). A different view by Yen & Horng, (2010), insisted that alternative attractiveness is significant and negatively impacts the customer involvement in long-term relationship and relationship commitment. The authors stated that customers are committed to the service providers and tend to preserve existing relationships only if they are unconscious of alternative attractiveness or do not perceive the alternatives to be more attractive than what has current service provider's offer. Bansal et.al (2005) has suggested that alternative attractiveness can be utilized as a "pull" factor which can attract customers to stay as they are conscious of available alternatives from that existing service provider, thus, developing a customer loyalty. Therefore, this study attempts to examine the influence of alternative attractiveness, whether it can be a factor to pull them to switch to another service provider in the Malaysian mobile telecommunication sector by proposing the following hypotheses: H2a: Alternative Attractiveness has a positive significant influence on Commitment in the mobile telecommunication industry.
H2b: Alternative Attractiveness has a positive significant influence on Trust in the mobile telecommunication industry.

Relationship between Relationship Quality and Customer Loyalty
Commitment is a central concept in relationship marketing which leads to customer loyalty (Sheth, 2017;Rahman & Ramli, 2016). Morgan & Hunt (1994) have emphasized that commitment is an important component of a successful relationship because it gives a rise to mediate behaviour. Furthermore, commitment assists in building a long-term relationship between service providers and customers. Morgan & Hunt (1994) considered commitment as one of the two most important factors (another factor is trust) in determining loyalty of this relationship and approach it as a synonym of customer loyalty. A commitment based on emotions has a positive impact on customer intentions (Reydet & Carsana, 2017), and on positive verbal communication (Hennig-Thurau et al., 2002). Moreover, commitment encourages and fosters customer collaboration and loyalty (Morgan & Hunt, 1994). Similar to trust, commitment is one of the most significant variables that aid to evaluate relationship strength level and is a useful element of loyalty measuring (Akrout andNagy, 2018, Sheth, 2017;Rahman & Ramli, 2016). Trust is one of the major components of customer loyalty, to which they may buy more, accept higher prices, and develop a positive word-of-mouth communication (Akrout and Nagy, 2018;Aydin & Ozer, 2005). Building trust between service providers and consumers will encourage them to assume that the risk which may occur in the business transactions are borne by both parties (Ahn, Shamin & Park, 2021). If the service provider is able to keep their promises, reliable, and concern for their customers, this can then influence the customer's loyalty to that particular service provider (Anderson & Karlstrom, 2014). The development of trust is considered to be a critical result of establishing a long-term successful relationship between all parties involved. This is due to the presence of service-based industries, as customers tend to act and make a purchase decision depending on their previous consuming experience (Ahn et al, 2021). Furthermore, customer's trust also plays a significant role in building long term relationship and achieving customer loyalty (Akrout & Nagy, 2018;Auruskeviciene et.al, 2010). Therefore, this study attempts to measure the impact of commitment and trust to customer loyalty in the Malaysian mobile telecommunication sector by proposing the following hypothesis: H3a: Commitment has a positive significant impact on Customer Loyalty in the mobile telecommunication industry H3b: Trust has a positive significant impact on Customer Loyalty in the mobile telecommunication industry

Relationship Quality as a Mediators between RMT's and Customer Loyalty
It has been proposed that relationship quality could be evaluated by looking at different dimensions such as trust, communication, and interdependence (Themelin et al., 2020, Tough et al, 2018. Morgan and Hunt (1994) stated that those commitment and trust components are the basic constructs for measuring loyalty. The literature on relationship quality suggests that the quality of the relationship between the parties involved is an important determinant of loyalty (Leverin & Liljander 2006). However, there is no consensus regarding the right relationship quality dimensions, and little empirical evidence regarding the nature and extent of the overall impact of relationship quality on service quality. Relationship quality dimensions may come from a variety of factors (Themelin et al., 2020;Tough et al., 2018). Akrout & Nagy (2018) and Bianci & Abu Saleh (2020) stated that trust, satisfaction, and commitment are indispensable elements to form a relationship quality dimension. The findings revealed that the level of trust and commitment are the most common attributes of relationship quality. High relationship quality may also result in customers' reliance on the integrity and future performance of their service provider (Nadeem et.al. 2020;Bianci & Abu Saleh, 2020). In this study, two variables comprising commitment and trust were proposed as relationship quality (RQ). Morgan & Hunt (1994) has described the importance of both factors in determining customer loyalty. Commitment and trust provide values for the customers, and act as important factors towards achieving loyalty (Bianci & Abu Saleh, 2020;Putit & Adullah, 2019;Jesri et.al, 2013). This study attempts to measure the impact of commitment and trust as mediating factor between brand reputation and alternative attractiveness to customer loyalty in the Malaysian mobile telecommunication sector by proposing the following hypotheses: H4a: Commitment mediates between Brand Reputation and Customer Loyalty H4b: Commitment mediates between Alternative Attractiveness and Customer Loyalty H4c: Trust mediates between Brand Reputation and Customer Loyalty H4d: Trust mediates between Alternative Attractiveness and Customer Loyalty

Conceptual Framework
Theories are used with the intention of explaining the relationship between RMT, relationship quality and customer loyalty. This study has examined a few theories of customer loyalty in order to understand their relationship with RMT's. Thus, this study proposes a conceptual framework on customer loyalty in the mobile communication context based on several past theoretical reviews. The overall framework provided the basis for the research on the RMT, relationship quality and customer loyalty as depicted in figure 1.

Methodology
The respondents of this study were mobile phone users who subscribed to mobile service providers such as Maxis, Celcom, Digi, and U-Mobile in Peninsular Malaysia. These respondents were the main users of mobile networks numbering 010, 011, 012, 013, 014, 016, 017, and 019 respectively. Both segments of post-paid and prepaid users were covered except for fixed home lines. Table 1 has shown the population's distribution of sample size and the determination of sample size appropriate for this study using quota sampling technique. A total of 400 samples were chosen for primary data collection based on the determination of sample size set by Krejcie, & Morgan (1970). Four main cities in Peninsular Malaysia that included Kuala Lumpur, Shah Alam, Melaka, and Penang were selected since they revealed the highest mobile phone rate users compared to other cities (MCMC, 2014).
The survey questionnaire's instrument details involved targeted respondents' demographic profile, and item scales of brand reputation, alternative attractiveness, commitment, trust, and customer loyalty. A set of self-administered questionnaires was developed consisting of 27 items. This set of questionnaires comprised seven (6) sections: Section A: Demographic profile, Section B: Brand reputation, Section C: Alternative Attractiveness, Section D: Commitment, Section E: Trust and lastly Section F: Customer loyalty. A multi-items scale was engaged in order to measure the variables, which was adopted and adapted from the previous literature as presented in Table  2.  (2005) In the demographic profile section, this study has followed recommendations set by Malaysian Communication and Multimedia Commission Hand Phone User Survey (MCMC 2014, pp 7), to which the data were collected by both core set and trends set. The core set includes nationality (only Malaysian citizen), sex (male and female), ethnicity (Malay, Chinese, Indian, and others), age group, level of education, and income level. Meanwhile, for the trend set, this study has included the length of usage of the mobile services. Each item in the questionnaire was measured using five point Likert-scale ranging from 1 = "Strongly disagree" to 5 = "Strongly agree" to measure the questionnaire as it is simple to construct, likely to produce a highly reliable scale and easy to read and complete for participants (Malhotra, 2009). A pilot study was initiated to check on the item reliability, with Cronbach's Alpha value above 0.70. According to Churhill & Iacobucci (2005), the instruments should be reviewed to ensure an absence of confusion, offensiveness or misleading of survey questionnaire. In addition, it can assist researchers to identify mistakes on the survey questionnaires such as wording, sequence, form and layout, difficulty, and instructions for a better understanding before actual survey is carried out (Peng, 2006). Bernard (2000) suggested an observation of at least 6 to 10 respondents in a pilot study. In this study, a pilot test was done on a convenient sample of 30 mobile phone users.

Data Analysis
A total of 395 returned questionnaires were checked, decoded, and keyed into the database using statistical software SPSS Version 20 and Smart PLS 3.0 software. Data screening processes were performed by preliminary check and data cleaning procedures to search out for legibility, consistency, competency, and ambiguity of the responses. Based on the preliminary checks, 3 questionnaires were discarded because they are returned as blanks, while another 10 questionnaires were discarded with over 30% of unanswered items in the total response (e.g., not answering certain part in the questionnaire booklet). The study has performed a partial least square structural equation modelling to test the relationship of brand reputation and alternative attractiveness on relationship quality, and the relationship of relationship quality on customer loyalty as projected in hypotheses H1a, H1b, H2a, H2a, H3a, and H3b. However, since the remaining hypotheses: H4a, H4b, H4c, and H4d requires examining the effect of mediating variables, mediation analyses were conducted for testing the mediating role of commitment and trust, using the bootstrapping method of Preacher and Hayes (2008). For this study, 5000 bootstrap resamples were chosen with 95% confidence interval (CI) of the indirect effects by following recommendation from Preacher & Hayes (2008).

Findings
A total of 382 usable survey questionnaires were carried out for further statistical analysis. The demographic profile of respondents' data was identified by performing the frequencies analysis as presented in Table 3.

Measurement and Structural Model
In assessing the measurement model, both convergent validity and discriminant validity have been used. The convergent validity of the measurement is usually ascertained by reviewing the loadings, average variance extracted and also the composite reliability (Gholami et al., 2013). The result here shows that 5 items were deleted due to low loadings. The loadings should be greater than 0.7 to ensure validity of the item. A total of 22-item then measured for composite reliability where all the values were greater than 0.7, and the AVE values were also greater than 0.5 as shown in Table 4 below. The discriminant validity of the measures criteria has been used in comparing the correlations between constructs and the square root of the average variance extracted for that construct. Table 5 below shows that all the values on the diagonals were greater than the corresponding row and column values indicating that the measures were discriminant.

Relationship between RMT's And Relationship Quality (Hypotheses H1a, H1b, H2a, And H2b)
Structural modelling procedure was executed to test the relationship between brand reputation and alternative attractiveness with commitment (outcome variable) (Hypotheses H1a & H2a). The results show that R² = 0.482, indicating that a total of 48.2% variance of commitment is explained by brand reputation and alternative attractiveness. Both predictors contributed a significant impact on commitment, where the strongest predictor is brand reputation (β=0.382, p<0.05) followed by alternative attractiveness (β=0.380, p<0.05) respectively. Meanwhile, the relationship between brand reputation and alternative attractiveness with the trust (outcome variable) (Hypotheses H1b & H2b) shows the value of R² = 0.508, hence indicating that almost 50.8% variability of trust as a dependent variable, can be explained by its predictors, that is, brand reputation and alternative attractiveness. Both predictors contribute a significant impact on trust, and that the strongest predictor comes from alternative attractiveness (β=0.53, p<0.05) followed by brand reputation (β=0.327, p<0.05) respectively. Therefore, H1a, H1b, H2a, and H2b are supported in this study.

Relationship between Relationship Quality and Customer Loyalty (Hypotheses H3a and H3b)
In order to analyse the relationship between the relationship quality (comprising of commitment and trust) with the dependent variable, customer loyalty, R² value appears to be 0.523. This indicates that a total of 52.3% variability of customer loyalty is explained by both commitment and trust. The result also establishes that commitment and trust contribute significant impacts on the dependent variable, customer loyalty with commitment appearing to be the strongest predictor (β=0.443, p<0.05) followed by trust (β=0.330, p<0.05) respectively. In this equation, trust appears to be the strongest influence on customer loyalty compared to commitment with a beta value of 0.464. Table 6 below shows the hypothesis testing of H1, H2 and H3.  (H4a, H4b, H4c, And H4d) The relationship between brand reputation and customer loyalty with the presence of commitment as mediating variable (hypothesis H4a), produced B = .1129, CI = .0601 to .1715, meanwhile, the relationship between Brand Reputation and Customer Loyalty, with the presence of Trust as mediating variable (hypotheses 4c) shown B = .1963, CI = .1316 to .2680. This result has confirmed the mediating role of relationship quality in the relationship between Brand Reputation and Customer Loyalty since zero falls outside of the interval (Strout & Bolger, 2002;Preacher & Hayes, 2008;Preacher & Hayes, 2009;Blaauw, 2012). Since path c' shown that Brand Reputation was significantly related with Customer Loyalty B = .3101, p < .001, therefore the relationship suggests a partial mediation (Baron & Kenny, 1986;Strout & Bolger, 2002;Preacher & Hayes, 2004;2008). Accordingly, hypotheses H4a and H4c are supported in this study. Bootstrapping result in Table 7 shown the relationship for brand reputation. Since zero does not occur between lower boundary confidence interval and upper boundary confidence interval, the indirect effect for this mediation can be assumed as significant (Strout & Bolger, 2002;Preacher & Hayes, 2008;Preacher & Hayes, 2009;Blaauw, 2012). Furthermore, the results from path c' shows that alternative attractiveness was significantly related with Customer Loyalty B = .2644, p < .001, suggesting a partial mediation in the relationship (Strout & Bolger, 2002;Preacher & Hayes, 2004;2008). Therefore, hypotheses H4b and H4d are supported in this study. Meanwhile, Table 8 shows the results of bootstrapping analysis for alternative attractiveness

Discussion
The theoretical mechanisms used to explain the conceptual framework and research hypotheses are derived from two different sources. First, the principle of Push-Pull-Mooring Migration Model by Bansal et.al, (2005) was explored to explain the effects of the relationship marketing tactics on customer loyalty. Second, the theoretical concept of Commitment-Trust Relationship Marketing Theory by Morgan & Hunt (1994) was used to explain the impact of proposed mediating variables in the relationship between the RMT and customer loyalty. The study focuses on customers' perception towards their service providers' RMT, which may affect their intention to consider alternative service providers and whether to switch or remain loyal to the existing ones.
A major finding of this study is the empirical identification of potential relationship marketing tactics (RMT) that enable Malaysian mobile service providers to either develop or strengthen their tactics in an attempt to retain their customers. The results indicate that brand reputation and alternative attractiveness have significantly affected the customers' decision towards selecting a mobile service provider. Besides, this finding also provides an early warning for the concerned service providers to further improve on their products and services' brand image, as well as provide better add-on valued services in ensuring a continued customer retention and subsequently customer loyalty. Further findings reveal that brand reputation is the most influential factor to be considered by the mobile service providers. Most of the mobile users have a tendency to stay loyal with a provider who offers more attractive features such as better service package, lower price, wider mobile coverage, and an excellent after-sales services, and consistent updates from their service providers. Failure to provide such attractive product features will lead the customers to look elsewhere for other alternative offers by competitors, thus resulting in switching behaviour. This study also discerns the importance of relationship quality on customer loyalty. Commitment and trust are found to be significantly associated with customer loyalty. This is harmonious with the findings from the previous research (e.g. Akrout & Nagy, 2018;Sheth, 2017;Ghasemi et.al, 2010), which suggest that the higher the level of commitment and trust perceived by the customers, the higher the level of customer loyalty achieved by the service providers. In addition, results from the mediation analyses also successfully demonstrate that relationship quality has mediating effects within the relationship between RMT's and customer loyalty.

Conclusion
This study has identified some limitations that need to be investigated in future research. Firstly, this study was developed to extend a theoretical model which is originally based from Bansal et.al. (2005). However, this study did not focus on all of variables derived. Future studies could apply other variables in the model to more meaningful manners. For example, various factors from 'Push Effect' needs to be introduced as the relationship marketing tactics (RMT) as suggested by Bansal et.al, (2005). Secondly, this study only analysed the relationship between the relationship marketing tactics (RMT) and relationship quality (RQ) on customer loyalty; given the relationship quality are Commitment and Trust. Future study could explore other dimensions of RQ such as satisfaction and communication.