Do Social Media Marketing Activities Build Long-Term Relationships? An Empirical Study of Indonesian Cosmetic Brand on Gen Z

. Social Media Marketing (SMM), which offers two-way communication between brand and customer, has been widely used for marketing activities as a communication channel. Drawing from the tenets of Stimulus-Organism-Response (SOR) framework, the present study examined how social media marketing activities (SMMA) affect brand loyalty with brand experience and relationship quality as the mediators for Indonesian cosmetic brand. Empirical evidence was collected from 300 cosmetic users who follow specific Indonesian cosmetic brand on social media and have purchased cosmetic items from that brand. The results of structural equation modelling revealed that SMMA has a beneficial influence on brand experience and relationship quality, brand experience and relationship quality are significant mediators in the relationship between SMMA and brand loyalty, SMMA has no direct influence on brand loyalty, and brand experience significantly influences relationship quality. The findings go beyond the existing literature on Social Media Marketing Activities (SMMA) by providing a holistic model for customer behavior outcome (brand loyalty) in perceiving SMMA through brand experience and relationship quality and confirming the relationship between SMMA and brand experience to relationship quality. This research suggests that managers optimize SMMA and the brand experience dimension to improve relationship quality, significantly promoting brand loyalty for building a long-term relationship .


Introduction
Positioned as the world's 4th largest country with over 272 million people, Indonesia has become a promising market for various industries, including cosmetics (Candra 2019;Briliana and Mursito 2017).With the average age of 15 to 18 years for first cosmetic usage in Indonesia and the facts of the high growth rate of Indonesian people at a young age, the growth of the cosmetic industry is also growing linearly, corresponding with the increasing growth of the cosmetic sector in Indonesia (Nurrohmah et al. 2021).In 2017, the cosmetic industry grew by 6.35%; in 2018, the cosmetic industry grew by 7.36%; in 2019, the cosmetic industry grew by 9%; and in 2020, the cosmetic industry grew by 9.39% (Media Indonesia, 2021).Competition has encouraged introducing of new products and services, lowering entry barriers and making markets more competitive.However, brands often find it more challenging to retain customer loyalty toward specific brands (Kim and Ko, 2012).In the light of technological advancements and the evolution of marketing communication from transactional to relationship-based, social media platforms have recently and widely used as communication channels for marketing activities as it offers two-way direct communication, allowing brands and customers to communicate with each other regardless of time, location, or medium (Kim and Ko 2012).Accordingly, marketers must retain meaningful customer-brand relationships and communicate value propositions on social media (Khan 2022).Previous studies considered social media marketing activities (SMMA) to offer an effective way to improve relationship equity, brand equity, and value equity (Kim and Ko 2012), brand awareness and brand image ( Ibrahim et al. 2021).Yet, a study by Wibowo et al. (2020) found relationship quality as the mediator between SMMA and brand loyalty is a strong mediator by combined SMMA and customer experience dimensions.In addition, Ibrahim et al. (2021) argued that the more effective the connection, the stronger the consumer-brand relationship, and hence the more loyal consumers to the brand.Relatedly, to determine the long-term relationships strategy, the author proposes both direct and indirect relationships with brand experience and relationship quality as the potential mediators between SMMA and brand loyalty, where SMMA and brand experience are determined dimensions of relationship quality.Therefore, the objectives of this study are to investigate the influence of SMMA and brand experience on relationship quality and to investigate the mediator factors of brand experience and relationship quality in the SMMA to brand loyalty relationship.Under these objectives, the author contributes to the SMMA, relationship quality, and brand loyalty literature.Additionally, marketers will benefit from the findings to enhance brand loyalty to build long-term relationships.

Stimulus-Organism-Response (SOR) Theory
The Stimulus-Organism-Response (SOR) theory, as defined by Mehrabian and Russell (1974), posits that external stimuli (S) evoke the cognitive and emotional internal states of consumers (O) and drive the consumers to certain behavioral response (R) in either positive or negative ways.This approach can be used and has been used frequently in previous studies for the theoretical background in social media marketing activities study, where the study by Koay et al. (2020) examined SMMA as stimuli (S), brand experience as an organism (O), and consumer-based brand equity as a response (R); and the study by Ibrahim et al. (2021) examined SMMA as stimuli (S), brand trust and loyalty as an organism (O), and revisit intention as a response (R).In parallel, this study examines perceived social media marketing activities (SMMA) as stimuli that stimulate customers' responses to brand social media.As a result of engaging in social media marketing activities (S), in this context, brand experience and relationship quality are assigned as the organism (O), referring to the emotional states and cognitive evoke by stimuli, while brand loyalty is assigned as the response (R), reflecting specific behavioral consumers response.

Social Media Marketing Activities (SMMA)
With the increased use of social media adoption within an organization, social media in marketing is characterized as a channel for establishing relationships with customers (Kelly et al. 2010).Social media marketing (SMM) is defined by Kim and Ko (2012) as a two-way conversation that seeks empathy and enforces familiar feelings with users linked with brands.In addition, Sheth and Kim (2017) argued that building an emotional connection can influence consumers' attitudes toward a brand on social media.This study incorporates SMMA by utilizing Kim and Ko's (2012) SMMA components, which are entertainment, interaction, trendiness, customization, and word of mouth (WOM) as it more relevant to the study context.SMMA's entertainment component is one of the main reason consumers adopt social media because it entertains them and it evokes good emotions, influencing brand attitudes (Sheth and Kim, 2017).The SMMA interaction components describe cyberspace for customers to talk, share, and exchange ideas, where two users can have an interactive dialogue while others can watch or join the conversation (Ariel and Avidar 2015).The trendiness components of SMMA represent the most recent information on social media (Kim and Ko 2012).The customization component of SMMA displays a brand's capacity to give personalized information to customers from diverse sources, resulting in good perceived control and customer satisfaction (Seo and Park 2018).Lastly, the WOM component of SMMA is described as an online user interaction about a brand (Khan 2022).

Brand Experience (BX)
Brand experience is defined as subjective customer response and behavioral response to brand-related stimuli part of a brand's design and identity, packaging, communications, and settings (Brakus et al. 2009).Existing research categorizes brand experience dimensions as sensory, emotional, cognitive, relational, and behavioral (Nysveen et al. 2012), and sensory, affective, intellectual, and behavioral (Brakus et al. 2009).Additionally, the online environment can evoke the brand experience since it allows for interactivity and a real-time brand experience, empowering customers to engage with the brand and other customers (Morgan-Thomas and Veloutsou 2013).According to Brakus et al. (2009), brand experience has a behavioral impact on consumer satisfaction and loyalty directly and indirectly through relationship quality (Francisco-Maffezzolli et al. 2014).

Relationship Quality (RQ)
The term "relationship quality" emerged from previous research of relationship marketing, where it is defined as a meta construct comprised of several key components reflecting the overall nature of relationships between companies and consumers (Rauyruen and

Brand Loyalty (BL)
Losing consumers in today's highly competitive environment is costly (Athanasopoulou, 2009).Previous studies argued that establishing brand loyalty can lower customer acquisition costs (Chinomona 2016), as some researchers argue that the cost of acquiring new customers is higher than the cost of retaining current customers (Kumar et al. 2011, Keisidou et al. 2013), or even up to six times higher (Chinomona 2016).As a result, scholars contend that brand loyalty has been leveraged to gain a competitive advantage (Denoue and Saykiewicz 2009, Iglesias et al. 2011, Chinomona 2016).Loyal consumers are one of a brand's competitive assets, classified as emotional creatures who are highly motivated by their desire to maintain a relationship with the brand with which they have an emotional connection (Chinomona 2016).Although researchers currently lack agreement on brand loyalty definitions and metrics, previous studies claim that brand experience and commitment are acceptable brand loyalty measures (Khabiraj and Shanmugan 2010).According to the literature and the findings of Chinomona's (2016) latest study, the brand experience can, directly and indirectly impact loyalty via relationship quality as a mediator (Francisco-Maffezzolli et al. 2018).Furthermore, previous research has discovered that relationship quality, directly and indirectly, affect loyalty (Rauyruen and Miller 2007).As a result of prior research, this study employs brand experience and relationship quality to be measured as the moderators of brand loyalty.

Brand Experience (BX) as Mediator
Based on the S-O-R approach proposed in this study, perceived social media marketing activities act as stimuli (S) that evoke brand experience (O).As previously mentioned, brand experience is described as a behavioral response evoked by brand-related stimuli that are part of a brand's design and identity, packaging, communications, and setting (Brakus et al. 2009).Additionally, previous study discovered that brand experience can be evoked through online environment (Morgan-Thomas and Veloutsou 2013).Accordingly, SMMA, an online-based environment, may stimulate (S) consumers' cognitive and emotional internal states (O), which in this context is brand experience.In parallel, studies by Khan (2022) and Zollo et al. (2020) have identified that SMMA stimulates brand experience.Thus, based on the S-O-R method and the findings of prior studies, the study proposes that perceived SMMA influence brand experience: H1: Perceived social media marketing activities (SMMA) have a positive influence on brand experience (BX) Continuing the previous approach, the S-O-R approach proposed in this study, brand experience, which operates as an organism, drives consumers towards a certain behavioral response, wherein this context is brand loyalty.According to Maheshwari et al. (2014), brand experience is the factor or driver of brand loyalty.In parallel, numerous studies have discovered that brand experience influences brand loyalty (Brakus et (2022) found no direct relationship between SMMA and brand loyalty; instead, the study discovered an indirect relationship through the mediating effect of community participation and lovemark.Yet, since studies that show a direct relationship between SMMA and brand loyalty are more frequent, the study proposed the subsequent hypothesis: H2: Perceived social media marketing activities (SMMA) have a positive influence on brand loyalty (BL)

Relationship Quality (RQ) as Mediator
According to the S-O-R theory approach provided in this study, perceived social media marketing activities act as external stimuli formed through entertainment, interaction, trendiness, personalization, and word-ofmouth (Kim and Ko 2012).The customization dimension in social media, which assisted companies in communicating their uniqueness and optimizing information provided, significantly increased customer satisfaction and commitment compared to noncustomization (Seo andPark 2018, Fung 2008).Accordingly, SMMA as the stimulus, forming through interaction and customization, can evoke customers' satisfaction and commitment; in this context, satisfaction and commitment are constructed as relationship quality.Therefore, the study proposed that perceived SMMA influences relationship quality:

Methods
This research was established using quantitative methods and carried out at DKI Jakarta, Indonesia from January to April 2022.The data sample was collected from the 4th to the 9th of February and conducted online throughout the Java area.The population of this study consists of cosmetics customers who used social media in the Java region, which comprises Jawa Barat, Jawa Tengah, Jawa Timur, Banten, DKI Jakarta, and DI Yogyakarta.According to Nurrohmah et al. (2021), in Indonesia, most of the ages for the first time using cosmetics are between the ages of 15 and 18 years, which is the adolescent stage.In parallel, the ages for initially accessing social media range from 13 to 17 years old (Statista, 2021).Furthermore, the Java area was chosen because Java island has the highest population density in Indonesia, accounting for more than 50% of the population (Dsikowitzky et al. 2019).With half of Indonesia's population residing in the Java region, the researcher expected the research findings to be generalizable to all islands or provinces in Indonesia.Accordingly, in estimating the target population, the number of cosmetic customers who use social media, the researchers calculate the total population within the Java area between 15 and 65 above, including both females and males, which is 117,323,985 people (BPS, 2020).In determining the sample, the study uses the purposive sampling method and conducted a preliminary test to select a sample of Indonesian cosmetic brands and spread it to 20 (twenty) beauty enthusiasts through an online survey.They were asked to mention 3 (three) Indonesian cosmetic companies with which they are familiar.The pre-test findings show Wardah with nine times mentioned, followed by Somethinc with eight times mentioned.However, Wardah cosmetics have been extensively studied (Samosir et al. 2016, Aisyah 2017, Amalia 2019, Salere et al. 2019); the researchers chose Somethinc as a sample of Indonesian cosmetic brands in this study.Accordingly, the sampling frame includes people who follow Somethinc on social media.Thus, the sample selection eligibility criteria were as follows: 1) follow Somethinc on social media, representing social media users in the population; and 2) purchase Somethinc's cosmetic item, representing cosmetic customers in the population.In addition, by identifying and setting the parameter by at least purchasing a Somethinc item once, the respondents are expected to have experience with the brand and plausible to have the intention to try other products of Somethinc, which aligns with the measurement indicator of brand loyalty in this study.The minimal sample size for this study is 110 respondents and this study obtained a total sample size of 300 respondents.

Data Collection
This study utilizes the data collection using primary data and collecting using an online questionnaire survey through a Google Form.The questionnaire's list of questions is adapted from previous research to meet the research objective and measured using a five-point Likert scale on each query.Likert scales are a measurement scale with five response categories ranging from 1 for "strongly disagree" to 5 for "strongly agree," requiring respondents to identify their level of agreement or disagreement with each of the questionnaire's statements (Malhotra 2015).In this context, scale 1 (one) ranges from 1.00 to 1.80; scale 2 (two) ranges from 1.81-2.60;scale 3 (three) ranges from 2.61-3.40;scale 4 (four) ranges from 3.41-4.20;and scale 5 (five) ranges from 4.21 to 5.00.(Sozen and Gufun 2019).The Likert scale options in this study can be seen in Table 1.The questionnaire contained 18 questions related to the demographic of the respondents and 18 questions linked to the studied variables as a construct in the conceptual framework model in Figure 1

Results and Discussion
The results presented in Table 3 show that all of the measurement items, except BX2, are greater than the required value of outer loading, which is greater than 0. Lastly, Table 4 shows the results of VIF value on model 3, which are 2.064, 2.387, and 2.209, respectively, suggesting that social media marketing activities, brand experience, and relationship quality variables are free of multicollinearity.Since all of the models are free of multicollinearity, indicating each independent variable on each model is not collated with each other and would not cause any instability in the subsequent regression analysis.

H3:
Perceived social media marketing activities (SMMA) have a positive influence on relationship quality (RQ) According to the S-O-R paradigm proposed in this study, relationship quality as an organism (O) drives customer response (R), which is brand loyalty.It is evidenced by several studies that found relationship quality affect brand loyalty (Hennig-Thurau et al. 2002, Rauyruen and Miller 2007, Huang et al. 2014, Adhikari and Panda 2019).Accordingly, the study proposed that relationship quality influences brand loyalty: H6: Relationship quality (RQ) has a positive influence on brand loyalty (BL)
al. 2009, Iglesias et al. 2011, Nysveen and Pedersen 2014, Chinomona 2016, Ong et al. 2018).Additionally, studies found that brand experience influences consumer satisfaction and commitment, which are the components of relationship quality proposed in this study (Human-Ramirez and Merunka 2019, Brakus et al. 2009, Das et al. 2019).Accordingly, studies have discovered that brand experience indirectly influences brand loyalty through relationship quality as a mediator (Francisco-Maffezzolli et al. 2018, Lee and Kang 2012).Several studies have found a direct link between SMMA and brand loyalty.Bilgin (2018) identified a direct relationship between SMMA and brand loyalty and an indirect relationship with the mediating variables of brand awareness and brand image.Khan (2019) established a direct relationship between SMMA and brand loyalty.Furthermore, Ibrahim et al. (2021) discovered a direct relationship between SMMA and brand loyalty and an indirect relationship with the mediating variable of brand trust.However, Fetais et al.

Table 1 .
Five-Point Likert Scale and Range

Table 2 .
Operational Model

Table 3 .
Operational Model

Table 5 .
Brakus et al. (2009)dels 1, 2, and 3 The third findings share the significant influence of the relationship quality on brand loyalty.With the coefficient value of 0.617, relationship quality is considered a strong driver variable and mediator variable of brand loyalty.The statement is also supported by the moderately low coefficient value of other types of driver and mediator variables between SMMA and brand loyalty: Bilgin (2018) identified brand awareness and the brand image as having the coefficient value of 0.321 and 0.222, Fetais et al. (2022) identified love mark and community engagement having the coefficient of 0.282 and 0.41, and Ibrahim et al. (2021) identified brand trust with the coefficient of 0.18.Compared to those mediators and drivers variables between SMMA and brand loyalty, relationship quality could be concluded as a strong mediator variable.Additional support for the finding was also concluded in a previous study byWibowo et al. (2022), where relationship quality provides a significant coefficient value of 0.749 as the mediator of SMMA and brand loyalty.As a notation, the different coefficient values obtained in this study and the previous study may differ due to different variables in constructing the relationship quality.The previous study byWibowo et al. (2022)constructs relationship quality with trust and satisfaction and is influenced by SMMA and customer experience, while this study constructs relationship quality with satisfaction and commitment and is influenced by SMMA and brand experience.Concluding from the obtained coefficient value, supported by previous studies, relationship quality is a strong mediating role and crucial role between SMMA and brand loyalty.The other driver of brand loyalty in this study, brand experience, is also considered a significant influence mediator between SMMA and brand loyalty.Compared to the other mediator and driver variables in previous studies and this study, relationship quality, with a coefficient value of 0.206, the influence of the brand on brand loyalty is considerably low.Yet, the strength of brand experience impacts relationship quality is moderately strong, with a coefficient of 0.493.The strong relation between brand experience and relationship quality has been identified and supported by the prior study byBrakus et al. (2009)that found that brand experience affects customer satisfaction, where customer satisfaction is the dimension of relationship quality.Based on the findings, the marketers are advised to optimizing social media marketing activities by paying attention to entertainment, interaction, trendiness, customization, and word of mouth indicators.In addition, marketers also suggested to manage photo quality and well-design description to promoting brand experience and relationship quality.Lastly, due to the limitation of this study, therefore, future studies are recommend to Figure out relevant indicators of SMMA and measure them with the limited sample of new customers, incorporating a broader area of Indonesia, and Examining the model with samples based on specific customer generation.