“The role of social media marketing (SMM) in building frozen food brand loyalty”

Social media platforms help businesses connect, communicate, and access information, boosting brand loyalty and awareness. This study aims to determine the relationship between social media marketing, brand engagement, and brand trust toward customer experience and brand loyalty. The focus is on the importance of the role of social media marketing for frozen food brands. The paper uses a descriptive research design and a quantitative approach where data were collected by distributing online questionnaires among frozen food consumers through Google Forms. The selected 250 respondents were located in big cities in Indonesia, such as Jakarta, Bogor, Depok, Tangerang


INTRODUCTION
The domain of communication technology has encountered a swift development in Indonesia from 2016 to 2020.This evolution is primarily attributed to the extensive implementation of digital technology and internet services.The quick advancement of information and communication technology (ICT), particularly in the telecommunications industry, clearly demonstrates Indonesia's progress.According to Kemp (2021), 73.7% of the total population in Indonesia uses the internet.The swift advancement of ICT has resulted in impressive transformations in diverse areas, such as marketing.
Social media has grown as a platform for interaction, communication, and information access.Indonesia, a developing nation, has embraced ICTs as a platform to establish direct relationships with various parties from various populations in various places, particularly within Indonesia.Active social media users are 61.8%,equivalent to 170 million of the total population in Indonesia.Instagram is a highly effective digital marketing platform due to its robust features that support businesses.These features include Instagram Feed, Instagram Ads, Instagram Live, Instagram Stories, Instagram Shopping, and Instagram Guide, all of which can be used for digital marketing (Kemp, 2021).
Social media portals inform customers about brands, encouraging them to do the same and helping the business become more well-known.Once consumers are aware of a particular brand, they will positively affect how people perceive its level of quality.People are inclined to procure a product and stick with a brand when they perceive it to be high caliber.Companies may invest in brand awareness to get the best results if they wish to develop new products or enter a market because it is thought to positively affect brand loyalty.
Providing comprehensive information through social media channels facilitates consumer recognition, selection, and purchase of products per their individual requirements.Conversely, inadequate or overly intricate information communicated through these platforms results in reduced consumer interest and heightened uncertainty in the purchasing process.For frozen food products featuring diverse product lines, social media must serve as a reference for consumers by offering accessible, detailed, and clear descriptions through videos, photos, and written content (Vermeir & Verbeke, 2006).

LITERATURE REVIEW AND HYPOTHESES
Frozen food is gaining popularity due to its convenience, ease of preparation, and lack of impact on taste and flavor (Dabadi & Gurung, 2020).The consumption of frozen seafood at the national level has attained 3.9 million tons.In comparison, frozen poultry has reached 1.9 million tons, meat 500,000 tons, processed dairy products 1.2 million tons, and fruits and vegetables 2.4 million tons.It is anticipated that there will be an increase in national consumption to 13.24 million tons by 2023 due to the Covid-19 outbreak, with fish consumption accounting for 45%, chicken for 22%, meat for 8%, processed dairy products, and fruits and vegetables having their own respective share (Asosiasi Rantai Pendingin Indonesia, 2019).This increase is attributed to public awareness of food supplies and restrictions on malls and traditional market operations to prevent outbreaks.Hence, frozen food is the primary food chosen by customers as a new trending product in this era.Business operators must market their products to become more known to potential consumers.With the development of technology, the marketing trend has shifted to digital marketing.The presence of digital marketing trends also affects many companies that already have social media accounts.Digital marketing leverages the function of social media, known as social media marketing.Social media platforms now provide ease in sharing information and facilitate communication.Social media presence provides an original way for companies to correspond with consumers, thus creating good opportunities (Kapoor et al., 2018).
Social media usage is crucial in the food industry because it enables businesses to reach a vast audience and share their brand message (Vasquez-Reyes et al., 2023).Companies should manage their social media presence carefully, especially when communicating marketing messages on social media.Social media marketing is carried out to initiate memory, awareness, action, and accession for a product, brand, person, business, or other entity by seeking the use of tackles from the social web such as blogging, microblogging, social networking, social bookmarking, and content sharing directly or indirectly (Rehman et al., 2022).Social media marketing uses a business's website, service, or product to conduct marketing through online media channels.Utilizing a vast community network to facilitate communication among individuals is a key feature of this approach, fostering interactivity and encouraging engagement.
Furthermore, integrating a broad community network to improve communication and foster involvement is a critical component of this tactic, which possesses significant potential in successfully advertising a business's goods and services.Social media marketing uses a cultural context that includes the virtual world of a social community, social opinion-sharing sites, social networks, and social news to find the purpose of communication (Tuten & Solomon, 2018).Social media marketing enables the diffusion of interaction, information, and content sharing (Chang et al., 2015).Consumer reactions to a brand may originate from attitudinal responses, encompassing their level of satisfaction and perception, or from their behavioral responses, including purchase in-tention, word-of-mouth, and loyalty (Kim & Ko, 2012).As such, social media marketing establishes memory, awareness, action, and recognition for a given product, brand, or business by utilizing tools such as microblogging, blogging, and content sharing.Furthermore, it disseminates services via online media channels and employs cultural contexts, such as virtual communities and news, to facilitate communication and interaction.
Social media marketing initiatives could be identified as a highly efficacious marketing communication strategy that endeavors to capture consumer perceptions and comprehension of social media marketing activities with 5 dimensions, specifically entertainment, interaction, trendiness, customization, and word-of-mouth ( Brand loyalty is a steadfast commitment to consistently repurchase a product or service in the future, resulting in recurrent brand purchases.Despite the possibility of changing, one's behavior is influenced by environmental factors, marketing efforts, and other factors.Consumers opt for brand loyalty when purchasing a brand over other brands in a product category (Lee et al., 2019).Brand loyalty encompasses the likelihood of consumers switching to another trademark, mainly when there are changes in the material used or the price of the product.Augmented brand loyalty also reduces indications of susceptibility among consumer groups to switch to competitors' products (Aaker, 2009).Companies can reduce marketing financing, attract new consumers, increase trade, and respond to threats from competitors by leverag-ing brand loyalty (Marliawati & Cahyaningdyah, 2020).Consequently, brand loyalty is influenced by the accumulated results over time regarding consumer satisfaction with the brand.
Brand loyalty is an aspect that is influenced by the perception of a product's quality and can be established by creating innovative products that are comparable or superior to competing products at lower or optimal prices (Wardhana & Yulia, 2021).Additionally, brand loyalty can serve as a benchmark for both consumer loyalty and dependence on a brand, with loyalty serving as a fundamental component of the product mix as a marketing center (Gabrielli et al., 2013).Consequently, loyalty attitudes are crucial in the underlying factors that drive repeat purchases (Tatar & Eren-Erdoğmuş, 2016).From two viewpoints, brand loyalty is described as an unwavering obligation to future brand repurchase, irrespective of circumstances (Chaudhuri & Holbrook, 2001).Based on this elucidation, attitudinal loyalty cultivates a positive impact on the relative brand value, while consumer loyalty cultivates to contribute to a high market share.
The concept of brand trust refers to the inclination of the average customer to rely on a brand's ability and expect its reliability.It encompasses customers' comfort while using a brand to fulfill their needs and wants (Delgado-Ballester & Luis Munuera-Alemán, 2001).Companies strive to create products that give customers trust, reliability, and safety, meeting their desired capabilities (Chaudhuri & Holbrook, 2001).Trust refers to implicit beliefs prohibiting individuals from taking advantage of situations, particularly with absent rules.This underscores the significance of trust in other individuals within virtual communities (Ridings et al., 2002).In the online context, individual trust has been studied concerning social media platforms.Personality traits and characteristics play a significant role in shaping the perception of an individual's trust, influencing customer behavior and intention.The relationship between social media and user personality concerning trust in a site is crucial.Trust plays a fundamental role in social media, affecting users' behavioral responses, such as their inclination to recommend the platform to others and their intention to use it in the future (Pentina et al., 2013).
Age, gender, and time spent on social media platforms influence customer trust perceptions, which can be measured by the dimensions of integrity, virtue, thoughtfulness, identification, and competence.Perception of integrity varies based on moral and ethical ideologies, along with age, gender, and frequency of social media use (Warner-Søderholm et al., 2018).The brand trust dimension assesses affective trust and relates to an emotional attachment and concern for others' well-being, while cognitive trust focuses on benefits rather than risks.Trust in social media is fostered by interaction orientation more than traditional websites, highlighting its affective nature (Calefato et al., 2015).In conclusion, brand trust is a complex and diverse concept crucial for success in the online business world.Companies can create strategies to establish and uphold customer trust by understanding the elements influencing brand trust on social media platforms.
Communication executed by corporations or conducted by users enhances the rapport with customers and augments brand credibility.The attributes of the cyber sphere constitute one of the elements propelling confidence.Investigative studies on the function of mediation in the correlation between behavioral intentions and social trading characteristics provide validation for the conviction in social commerce beliefs, particularly when engendering an advantageous online environment with respect to reputation, communication, magnitude, transaction security, word-of-mouth referrals, and quality of information (Park & Kim, 2013).As per Khong et al.'s (2013) exploration, it is assumed that one of the constituents of trust in social media platforms is consumer enfranchise or the presence of structural and psychological conditions, through user perceptions of amplifying the capacity to access, disseminate information, and effectuate transactions on social media, consumer enfranchise, and trust in social media platforms.Consumers harbor confidence in community members who subscribe to information, are reactive, and possess personal experiences with other individuals (Ridings et al., 2002).
Brand attachment is evident in consumer behavior, whereby consumers seek to reflect their own personality in every action taken.This psychological process models the mechanisms underlying the consumer repurchase process (Bowden, 2009).Another definition of brand attachment posits that it is characterized by the intensity of consumer participation and its relationship with the company's offerings or activities, which may be initiated by the company or by consumers themselves (Vivek et  According to Carbone and Haeckel (1994), experience quality is a perceptual impression that consumers form when they encounter products, services, and companies.This perception is developed by consolidating information received by the five senses.When consumers accept the offers provided by a company through emotional stimuli, they experience the customer journey before, during, and after making a purchase decision.Holbrook and Hirschman (1982) challenged the conventional information processing paradigm of consumer decision-making, which presupposes the consumer to be entirely rational.The study suggests an experiential approach to consumption that considers symbolic, hedonic, and aesthetic consumption factors from consumers' perspective.Consumer experience can serve as a distinguishing factor in competitive and less regulated markets, and, as such, any marketing strategy ought to encompass consumer experience to cultivate a favorable impression on customers.
According to Pine and Gilmore (1998), contemporary consumers are situated within an experience economy that is markedly advanced compared to previous commodity and service-based economies.The triumph of contemporary marketing is contingent upon the degree to which it can sell a stage of experience.Consumers, in turn, are exhibiting a growing immunity to messages specifically targeted toward them by companies.
Companies stage an experience when a customer pays to spend time taking in a series of memorable events.Or they engage customers and make a personal, lasting connection with them, which happens across two bipolar constructs: customer participation and connection.According to Schmitt (1999), an essential transformation has occurred in the marketing realm, whereby the traditional emphasis on product benefits and features has given way to a more comprehensive and pleasurable experience.This experience is characterized by a mutually influential relationship between cognitive thoughts and affective emotions.Due to the multidimensional nature of customer experience, multiple interpretations are possible, highlighting the subjectivity that can arise during the process.As a result, consumers' reactions, whether direct or indirect, can be subjective and impact the company.
The aim of this study is to determine the relationship between social media marketing, brand engagement, and brand trust toward customer experience and brand loyalty in frozen food.The advent of social media offers advantageous chances for a business marketing plan in an uncertain business environment and the people in Indonesia with the impetus of technical development and knowledge in influencing the sale of frozen food products.
Referring to the aim of this study and based on the comprehensive literature review, this study posits the following hypotheses: H1: Social media marketing has a positive and significant impact on brand trust.
H2: Social media marketing has a positive and significant impact on brand engagement.
H3: Brand trust has a positive and significant impact on customer experience.
H4: Brand engagement has a positive and significant impact on customer experience.
H5: Customer experience has a positive and significant impact on brand loyalty.

METHODS
A quantitative approach using a questionnaire as a research tool was applied to reach the current study's goal.Purposive sampling is used.To solve the research problem, the population was the Indonesian community who are consumers of frozen food and followers of respective social media accounts located in Jakarta, Bogor, Depok, Tangerang, and Bekasi, which is the epicenter of economic growth and development, so it has a significant chance of influencing consumer purchasing patterns through social media.This study uses an electronic questionnaire distributed via Google Forms for data collection, accessible to respondents via provided links.The data are gathered from 250 respondents.
The questionnaire examines five variables: social media marketing, brand trust, brand engagement, customer experience, and brand loyalty.A measurement interval scale was applied, specifically the Likert scale: 1 means "strongly disagree" and 5 means "strongly agree."This paper evaluates social media marketing with ten items Ebrahim (2020) generated.Nine items were taken from Huang and Guo (2021) and Konuk (2021) to measure brand trust.Nine items were adapted from Khan et al. (2020) and Verma (2021) for brand engagement.In terms of customer experience, nineteen items were taken from Klaus and Maklan (2012).Furthermore, three items of brand loyalty were taken from Ebrahim (2020).
The study employs Smart-PLS version 4.0.0 software to construct a measurement model using rigorous assessment of its reliability and validity.It subsequently scrutinizes the structural model by eliciting significant outcomes on the interrelationships among the constructs.Inferential statistical analysis comprises an inner model as a structural model and an outer model as a measurement model.In addition, statistical analysis utilizing frequencies was utilized to assess the characteristics and overall data descriptions of the sample.

RESULTS
The data collected from 250 Indonesian citizens revealed that 73.2% of the respondents were female, accounting for 183 participants.The remaining 26.8% were male, comprising 67 participants.Moreover, the survey yielded findings indicating that the demographic cohort of individuals between the ages of 17 and 25 exhibited the most significant proportion of respondents, 164 participants.Based on demographic education level, 104 were in senior high school (42%), 120 were undergraduates (48%), and 26 held graduate degrees (10%).According to the statistics from the questionnaire findings, 250 respondents are followers of frozen food Instagram pages.Table 1 shows the demographic findings.The validity test can be formed when the degree of relationship between two sources with different levels responds to the same size; convergent validity measures the relationship between constructs originating from the same phenomenon or event (Sekaran & Bougie, 2016).According to the convergent validity test performed using the loading factor value as displayed in Table 2 and Figure 1, 23 from 49 indicators measuring latent variables are valid.The AVE values serve the purpose of ascertaining the soundness of the pre-convergent test.The heuristic AVE possesses a valuation exceeding the threshold of 0.50 (Sekaran & Bougie, 2016).As demonstrated in Table 3 for AVE, the quintet of variables encompassing social media marketing activity, brand trust, brand engagement, customer experience, and brand loyalty have been officially affirmed to be valid.Cronbach's coefficient alpha and the composite's reliability (rho c ) serve the objective of the precision and coherence of the indicators' results.
Additionally, the assessment of precision serves as a crucial instrument in ascertaining the construct.In order to validate the consistency of quantification, it becomes imperative to exceed the recommended threshold of 0.7 as a general guideline (Sekaran & Bougie, 2016).Based on the data in Table 3, all five constructs are deemed reliable because the value exceeds the rule of thumb > 0.7.It is stipulated that all constructs are deemed applicable in hypothesis testing.Discriminant validity is carried out to measure the level of constructs that are considered to have a relationship and overlap each other (Hair et al., 2010).
Discriminant validity testing uses cross-loading data with a threshold limit value of > 0.70.In the heterotrait-monotrait (HTMT) test, the threshold limit value set is less than 0.85 or 0.90; when the existing value exceeds the threshold limit value, there is a problem with the discriminant validity (Hair et al., 2014).Table 4 shows that heterotrait-monotrait (HTMT) values with a set threshold limit value of less than 0.85 or 0.90 indicate that all existing variables are at values of 0.047 to 0.758, so they are conceptually different constructs because they are below the threshold values.In testing for common method bias, PLS-SEM displays the VIF result.Table 5 shows the value of the outer variance inflation factor that identified the construct indicator with a VIF value threshold of less than 3.3 and no collinearity problems.
The higher the VIF, the greater the degree of collinearity, and variance inflation factor values over five indicate a problem with critical collinearity between the indicators (Hair et al., 2014).Table 5 shows the common method bias calculation outcome using factor analysis.Goodness of fit (GOF) is a method used to measure theoretical structure in presenting reality specified in research data.The research model must be able to explain information about existing research data; this is done to determine the systematic level of a model and how the predetermined model produces a covariance matrix that is estimated and observed in indicators.Determining the value of goodness of fit refers to the value of standardized root mean residual (SRMR), normed fit index (NFI), and root mean square residual covariance (Hair et al., 2014).
The standard root mean residual (SRMR) value is declared an acceptable fit when the value is below 0.10, indicating a good match.When the value of the standardized root mean residual (SRMR) is above 0.10, it is categorized as poor fit or slightly good fit (Hair et al., 2014).The normed fit index (NFI) is a match index with a value between 0 and 1; when the model has a perfect match, the value will show the number 1 (Hair et al., 2014).Table 6 shows that the estimated value of the SRMR model is at a value of 0.083, where the value of SRMR is at the threshold limit value of < 0.12, so the value of SRMR is classified as a well-fitting model.It can also be seen that the NFI value is at 0.633, where the NFI value is expressed by the model to be classified as almost perfect because it is nearly close to 1. Q2 is used to measure the predictive relevance of the inner model.In measuring Q2, the technique used is the re-measurement of samples that pass through part of the data matrix, where measurements will be made on the model parameters and provide predictions about the estimated elements.
When the difference between the original value and the prediction value is smaller, the prediction value indicates a large Q2 value, which shows the accuracy of a model's prediction (Hair et al., 2014).A Q2 value of less than 0 insinuates that the model has no predictive relevance, while a Q2 value greater than 0 indicates that the model has predictive relevance (Hair et al., 2014).In measuring the Q2 value, measurements are made using blindfolding techniques, where three categories of values determine the predictive relevance measure; the value of 0.02 is classified as a small predictive relevance value, the value of 0.15 is classified as a medium predictive relevance value, and the value of 0.35 is classified as a large predictive relevance (Hair et al., 2014).Table 7 displays the findings of the data analysis.Hypothesis testing is a systematic step taken to see the significance level; hypotheses are accepted or rejected, where testing must be carried out to determine the relationship between the actual variables (Hair et al., 2014).In testing the hypotheses, the data seen are T-values with a threshold limit value of > 1.65 and P-values with a threshold limit value of < 0.05 so that the effect tested can be declared significant (Hair et al., 2019).The hypothesis is considered insignificant when the T-statistic value is below 1.65 and insignificant at 0.05 (Hair et al., 2014).Table 8 shows the results of hypotheses testing.This study concludes that companies that use social media marketing effectively are more like-ly to build trust with their customers.Trusted customers are more likely to buy from a brand, leading to increased sales and loyalty; customers are also more likely to recommend the brand to friends and family, effectively reaching new customers and growing the business.

CONCLUSION
The aim of the study was to evaluate the impact of social media marketing on brand loyalty through the mediating variables of brand engagement, brand trust, and customer experience.The two hypotheses indicate that social media marketing positively affects brand trust, and brand engagement positively affects customer experience.The other three hypotheses do not support the earlier research showing negative relationships.This paper concluded that social media marketing does not affect brand engagement.
Other findings indicate that brand trust does not affect customer experience and customer experience does not affect brand loyalty.In addition, this study shows that social media marketing has a negative relationship with brand engagement, and brand trust has a negative effect on customer experience, likewise customer experience on brand loyalty.Therefore, companies should use social media to connect with customers, build relationships, respond to comments and questions, and use feedback to improve products and services.
This paper has several limitations.It focuses only on the community that follows the frozen food brands.The company's history can be added as background information on the items being examined by future researchers since a study on frozen food products in Indonesia is still uncommon.This study's application can be made to other goods or services.Additionally, studies can add other variables, such as purchase intention, to demonstrate or reinforce the relationship between brand equity and brand trust and their effect on brand loyalty.These theories can be used or compared to similar sectors by scholars using other research methodologies.

Table 1 .
Description of the respondents

Table 3 .
Construct reliability and validity

Table 5 .
Common method bias

Table 6 .
Model fit

Table 8 .
Hypotheses testing This promotes brand involvement and a sense of community.Customers can become more engaged at a deeper level, and the brand's entire customer experience can be enhanced by involving them in product development or decision-making processes.The fifth hypothesis outcome shows an insignificant influence of customer experience on brand loyalty.This finding does not support Emini and Zeqiri (2021) that customer experience affects brand loyalty.To encourage brand loyalty, businesses should put a high priority on providing excellent client experiences.This entails regularly exceeding customer expectations, offering individualized encounters, providing quick and convenient service, and enhancing operations based on client input.Therefore, frozen food brands can foster strong brand loyalty by concentrating on improving customer experience.This will enhance customer retention, generate positive word-of-mouth, and sustain business growth.