Self-congruence theory: Factors affecting brand loyalty in fast-moving consumer goods industry

This study aims to formulate a conceptual model for analyzing several variables on customer engagement and brand loyalty, especially in the fast-moving consumer goods (FMCG) industry. The self-congruence theory used in this study will go through the value of congruity, brand attractiveness, and customer-brand identification. The study uses structural equation modeling with samples of 105 respondents. The result is value congruence has a positive effect on brand attractiveness and customer brand identification. Customer engagement has a positive impact on customer brand identification. Brand attractiveness has a positive impact on customer brand identification and brand loyalty. Also, customer brand identification has a positive effect on brand loyalty.


Introduction
Research from Nielsen found a change in customer preferences in making purchasing decisions lately. The Nielsen Global Customer Loyalty Research Q1 2019 found that 38% of Indonesians like to try new brands, specifically for customers in the fruit juice category, 47% of customers indicated that they would still change the brand of fruit juice they would buy 2 . In the fruit juice and juice category, Buavita is the most popular fruit juice and juice drink brand in Indonesia based on the Top Brand Index from 2018 to 2020. Based on the 2018-2020 Top Brand Index data, Buavita always excels with 37.1%, 43.3%, and 31.6% 3 . Furthermore, the Home Tester Club survey on May 5, 2019, with 373 respondents, showed Buavita to be the most chosen brand by customers based on considerations in terms of original and natural, taste does not contain artificial preservatives, choice of flavors, low sugar, and the best price. However, based on these data, only 8% of customers have loyalty to a fruit juice brand 4 .
Low brand loyalty can undoubtedly be a threat to business continuity. Brand loyalty is one way to increase market share and company sustainability in the industry (Yoo & Bai, 2013) According to Chaudhuri & Holbrook (2001), brand loyalty is critical to success and achievement marketing strategy. Therefore, many businesses compose various loyalty programs to acquire new customers and retain existing customers. Brand loyalty consists of a behavioral component and an attitude component (Jacoby & Kyner, 1973). Behavioral loyalty refers to repeat purchases, while attitudinal loyalty refers to consumer attitudes towards a preferred brand (Cossío-Silva et al., 2016).
There are many ways to create and develop customer loyalty to a brand to survive. A significant way to trigger, build, or grow brand loyalty is through customer engagement (Islam et al., 2017). According to social exchange theory, customer engagement includes various things: reciprocal interactions between customers and brands (Hollebeek, 2011). According to Kumar & Pansari, (2015), customer engagement can be observed from the existence of purchases and provide recommendations on brands, share brand experience, and improve brand products, especially in terms of quality through contribution to the brand community provide feedback on the brand.
Previous researchers found that customer engagement can affect brand loyalty. Customer engagement is a relationship between consumers with other consumers, companies, and certain brands that support increased brand loyalty (Brodie et al., 2011). A positive relationship with the brand and its followers can increase consumer loyalty. According to Dwivedi (2015), engagement has a 2 See more at Insights -NielsenIQ 3 See more at Top Brand Index With Complete Category | Top Brand Award (topbrand-award.com) 4 See more at Home Tester Club significant impact on loyalty intentions and can also increase consumers' value, quality, and satisfaction. Currently, customers engagement is a factor that drives brand loyalty (Harrigan et al., 2017).
From various existing marketing theories, this research uses a selfcongruity theory to describe how customer engagement can create brand loyalty because self-congruity can describe consumer behavior (Hosany, 2016). Through self-congruence, we can see whether consumers like a brand. Landon (1974) highlighted the role of self-concept in consumer behavior; research on consumer behavior has explained various constructs, including attitudes, preferences, choices, loyalty, and so on) in terms of the relationship between one's self-image and one's perceived image to a particular product or service. This idea became known as the theory of self-congruity. The self-congruence theory defines psychological processes and outcomes. Customers compare their perceptions of brand personality or image with their actual, ideal, social, and ideal social self-concept (Sirgy, 2018).
Studies related to the use of self-congruence theory are found in the hospitality sector (Sop & Kozak, 2019), tourism Huang et al., 2017) and the restaurant sector (Han et al., 2020). Based on our best knowledge, almost no research is known to simultaneously investigate self-congruence on the variables of customer engagement, value congruence, brand attractiveness, customer brand identification, and brand loyalty regarding the fast-moving consumer goods (FMCG) industry. Therefore, this study validates the selfcongruence theory in the FMCG or retail industry with particular reference to fruit juice packaged drinks consumers. The self-congruity theory used in this study will go through several mediating variables, namely value congruence, brand attractiveness, and customer-brand identification.
This study aims to formulate a conceptual model that can analyze several variables on customer engagement and brand loyalty and conduct empirical testing of other factors that affect brand loyalty. This study is expected to enrich the previous findings by Bergel et al. (2019), who have found the psychological process of how customer engagement affects brand loyalty; thus this research builds a new model for marketing in various business fields, especially for the FMCG industry.

Method
The population of this study is Buavita customers in all regions of Indonesia aged 17-35 years. The sampling technique used is a random sampling; the selected respondents are Buavita's Instagram followers. Purposive sampling is also done by contacting Buavita's followers on Instagram, assuming they are Buavita consumers who have bought Buavita at least two times as respondents who meet the requirements. The research was conducted from mid-March 2021 to the end of August 2021. This study uses SEM with the consideration that SEM is an appropriate analytical tool used for multivariate analysis in social research that uses latent variables (variables that cannot be measured directly). We used an online questioner for collecting data through Google Forms to 245 people, of which 137 people filled in, but 29 of them were incomplete. They, therefore, could not be used in the study, so this study used 108 respondents and proceeded data by AMOS 22 software.  The measurement scale of the questionnaire in this study is a Likert Scale of 1 -10, where a value of 1 means strongly disagree, and a value of 10 means strongly agree. Furthermore, this study uses two independents (exogenous) variables, namely customer engagement and value congruence, and three dependent (endogenous) variables, namely brand attractiveness, customer brand identification, and brand loyalty. The following is the definition of each research variable and the indicators used in this study, as seen in Table 1. Also, the research framework can see in Fig. 1.

Figure 1. Research's framework
Hypothesis based on research's framework as see in Fig.1:  H1. Value congruence has a positive influence on brand attractiveness. H2. Value congruence has a positive influence on customer brand identification. H3. Customer engagement has a positive influence on customer brand identification. H4. Brand attractiveness has a positive influence on customer brand identification. H5. Brand attractiveness has a positive influence on brand loyalty. H6. Customer Brand Identification has a positive influence on brand loyalty.

Empirical Result
The first step in performing a structural equation model (SEM) analysis is to create a measurement model to ensure the research data is valid and reliable. Based on the data processing results, standardized loading (λ) obtained all variables measurement has a value above 0.5, all AVEs have a value above 0.50, and all CR has a value above 0.70, that all variables are valid and reliable. The standardized loading (λ), CR, and AVE values of each variable are shown in Table 2.
The goodness of fitness in this measurement model is expected to meet the cut-off value based on Garson (2012). The outputs are shown in Table 3. Research hypothesis testing, this test was conducted to test the relationship between latent variables by looking at the value of c.r (Critical Ratio) > standard value 1.96 (significance level 5%) in the regression weight table from the AMOS output. The outputs are shown in Table 4. The hypothesis test between value congruence and brand attractiveness shows the results of CR 2.372, which means that hypothesis 1 is accepted where value congruence has a positive effect on brand attractiveness; this is by previous research fromElbedweihy et al., (2016). The hypothesis test between value congruence and customer brand identification shows the CR 2.176, which means that hypothesis 2 is accepted where value congruence has a positive effect on customer brand identification. This result is like research from Büyükdağ & Kitapci, (2021) and Han et al., (2020). Then hypothesis test between customer engagement and customer brand identification shows the results of CR 2.610, which means that hypothesis 3 is accepted where customer engagement has a positive effect on customer brand identification as research by Fujita et al., (2020). Meanwhile, the hypothesis test between brand attractiveness and customer brand identification shows the results of CR 2.377, which means that hypothesis 4 is accepted where brand attractiveness has a positive effect on customer brand identification as research byThen hypothesis test between brand attractiveness and brand loyalty shows the results of 2.598, which means that hypothesis 5 is accepted where brand attractiveness has a positive effect on brand loyalty, as research from Elbedweihy et al., (2016). The last hypothesis test between customer brand identification and brand loyalty shows the results of CR 3.586, which means that hypothesis 6 is accepted where customer brand identification has a positive effect on brand loyalty as research by Rather et al., (2019) and Han et al., (2020).

Conclusions
This research attempts to bridge the research gap and answer the research problems contained in the introduction above. The problem of this research is how to create excellent customer engagement in the juice industry to increase brand loyalty. Based on the results of research carried out using the object of the Buavita, it can be seen that it has a direct or indirect effect.
The relationship between value congruence and brand loyalty can go through brand attractiveness. The brand should have compatibility and similarity in value with consumers' values so that this suitability will affect the attractiveness of the brand and make customers loyal to the brand. The relationship between value congruence and brand loyalty can also be through customer brand identification. A brand needs to be perceived in unity with consumers; if consumers feel identical to a brand, it will affect customer loyalty. The relationship between customer engagement and brand loyalty can go through customer brand identification; when there is an interaction between consumers and brands and consumers feel identical to the brand, consumer loyalty to the brand can increase.
The relationship between brand attractiveness and brand loyalty can go through customer brand identification; when consumers have a good evaluation of brand characteristics, consumers will feel that consumer values can increase brand loyalty.