Building Community Trust and Financial Inclusion Through Online Religious Leaders and Digital Communication: Evidence from Islamic banking in Indonesia

In a modern economy, the banking sector is one of the most vital institutions that significantly impacts a nation’s economic stability and growth. The significance of banks can be measured by the proportion of banking assets relative to the total assets of financial institutions (Saunders, 2020). In Indonesia, banking dominates 78% of the total financial sector assets as of 2024, demonstrating its pivotal role in national economic dynamics (Bank Indonesia, 2024). The influence of banking extends beyond domestic borders. As Eichengreen (2015) explains, banking crises in one country can transmit financial shocks internationally through four main channels: (1) global financial markets, where declining investor confidence leads to capital flight; (2) trade relations, where international financing disruptions affect exports and imports; (3) cross-border financial exposure, especially among large global banks; and (4) perceptions of systemic risk that trigger panic in structurally similar economies. Indonesia’s banking system comprises both conventional and Islamic banking institutions (IBIs). Islamic banks operate based on Sharia principles and include Sharia Commercial Banks (BUS) and Sharia Rural Banks (BPRS), in contrast to conventional banks that use interest-based mechanisms (Bank Indonesia Regulation Number Nomor 11/15/PBI/2009 Regarding the Conversion of Conventional Banking Activities to Islamic Banking).

Bank Syariah Indonesia (BSI) is one of the largest Islamic banking institutions in Indonesia (BSI). However, the BSI’s contribution to the national financial sector remains modest. In 2023, BSI’s market share stood at 2.95% in terms of assets, 3.32% in financing, and 3.41% in third-party funds. While these figures reflect incremental growth from the previous year, they indicate that the BSI still lags behind its conventional counterparts in terms of financial intermediation capacity. The BSI’s limited presence is also reflected in its operational infrastructure. In 2023, BSI operated 1,986 branches, significantly fewer than conventional banks such as BRI, which has 7,878 branches. Furthermore, BSI maintains only one overseas branch, whereas conventional banks have broader international coverage (Bank Indonesia, 2025). This lack of outreach limits BSI's service capacity across Indonesia’s vast archipelago, especially considering that 87.2% of Indonesia's 273 million citizens are Muslims. However, only 8.5% of Muslim Indonesians used BSI services in 2023 (Otoritas Jasa Keuangan, 2023). In addition to accessibility challenges, BSI face service-quality concerns. Based on social media reviews from February to March 2025, customers frequently complained about online transaction failures, ATM errors, delayed responses from customer service, and system upgrades that reduced functionality. These issues reflect performance gaps that hinder user experience and trust.

According to Syukri (2019), Islamic bank performance can be optimized through three key dimensions: (1) strategy, including product innovation, digital transformation, and Sharia compliance; (2) strategy execution, including digital service implementation, risk management, governance, and marketing; and (3) environmental factors, such as regulations, macroeconomic trends, and cultural values (Hamdani, 2017; Masood, 2019; Omar, 2021). From a marketing perspective, understanding consumer behavior and decision-making processes is vital for Islamic banks. As Tirole (2017) points out, banks serve as intermediaries between surplus and deficit agents, both of whom are the customers. Optimal performance depends on how well banks meet the long-term needs and values of these customer categories. In the current banking landscape, service marketing and customer relationship management (CRM) are crucial components. Thus, customer service units are essential organizational functions (Lewis, 2021), and understanding customer behavior is critical, particularly for Islamic banks that adhere to Sharia principles (Bessis, 2015).

Ajzen’s (1971) Theory of Reasoned Action states that behavioral intentions are shaped by attitudes and subjective norms. In the context of Islamic banking, online religious leaders (key opinion leaders [KOLs]) play a vital role in shaping norms. Nurul (2017) asserts that religious figures influence Muslim consumers by legitimizing the use of Shariyah-compliant financial products. Eid (2015) further confirmed that religious authorities serve as reference groups that affect consumer trust and purchasing decisions. Gong et al. (2021) demonstrated that KOL features such as reputation, content compatibility, and interactivity directly and indirectly affect consumer purchase intentions through emotional and virtual engagement. Tran and Nguyen (2022) highlighted that only content compatibility and interactivity significantly influence trust and purchase intention. These differing results reveal an empirical gap regarding the impact of KOL characteristics on digital environments.

To assess this gap, this study adopts indicators of online religious leaders, including charisma, reputation, compatibility perception, and content quality, based on the models proposed by Brown et al. (2017) and Ernst et al. (2022). Charisma is particularly central in the Indonesian religious context, where figures such as Ustadz and Kyai wield considerable influence over public decision making. Online advertising is another determinant of customer decision-making. Ad credibility, religious alignment, emotional appeal, testimonials, and platform security influenced trust in Islamic banking. Empirical studies have supported this hypothesis. Nemati et al. (2022) found that advertising perception affects purchase behavior via attitude, moderated by sex. Sriram et al. (2021) reported that ad creativity and emotional appeal influence consumer behavior, while Shaouf et al. (2016) found that visual design indirectly affects purchase intention via attitude. This inconsistency suggests a research gap in the effectiveness of online advertising, particularly in Islamic banking. The BSI has used influencers and celebrities (e.g., Atta Halilintar, Ria Ricis) in its online campaigns across social media platforms such as YouTube, Instagram, and Facebook. However, negative customer feedback, such as misleading cashback offers, calls for a critical evaluation of the effectiveness of online advertising methods. Management performance also influences decision-making. Ten critical success factors, including Sharia governance, service quality, product innovation, operational efficiency, risk management, reputation, customer satisfaction, and digital capability. Empirical studies by Cardoso (2024) and Phong and Anh (2023) show that brand image and trust significantly affect customer loyalty.

However, environmental investments have mixed results. Aslam et al. (2021) suggest that environmental management enhances financial performance via environmental performance, while Li and Wu (2017) reports a decline in profitability post-implementation due to costly requirements. This contrast underscores the need to assess how management practices, especially those that influence brand image and efficiency, affect Islamic banking performance. In this study, management performance is evaluated using indicators such as liquidity, technological innovation, reputation and stability, and promotional strategies, with trust and motivation as the mediators. Trust is a fundamental factor in long-term bank-customer relationships. Trust develops through consistent service quality, transparency, and Sharia compliance. Althuwaini (2022) and Manzoor (2020) emphasized that social media marketing influences trust, which shapes purchasing behavior. Similarly, Stravinskiene et al. (2021) found that company reputation affects consumer trust and loyalty in multiple dimensions. Motivation is also key. Faiza (2023) identifies interaction quality, customer satisfaction, and religiosity as drivers of switching from conventional to Islamic banks in Oman. This study adopts religious, economic, social, and emotional motivations as mediators in shaping customers’ decisions. Perceived product value also plays an additional role. Yu and Lee (2019) found that emotional, environmental, and aesthetic values influence attitudes and behavioral intentions, while Gan and Wang (2017) highlighted utility and social value. This study extends perceived product value by incorporating justice and responsibility to align with transparency and accountability emphasized in Islamic banking operations. BSI was selected as the research object because it is the largest Islamic bank in Indonesia, formed through the merger of three state-owned Islamic banks. Its scale and strategic role offer a rich context for examining how online religious leaders, online advertising, and management performance influence customer decision making.

Understanding consumer behavior in the Indonesian banking sector is crucial for practitioners, regulators, policymakers, and banking consumers. Insights from studies on banking consumer behavior are highly valuable for banking practitioners in managing their institutions effectively to enhance performance and deliver superior services to customers. One of the main challenges faced by researchers is identifying the factors that influence customer behavior toward Bank Syariah Indonesia (BSI), particularly in the current context, where advancements in information technology significantly affect the operational performance of business institutions (Davis, 1989). Conceptually, the use of technology for transactions or customer services, as well as for promotional purposes, is conducted through Internet-based platforms, television, radio, and telephones. The rapid advancement of Internet technology has given rise to various digital media channels for promotion, including social media platforms such as YouTube, Instagram, Twitter, TikTok, websites, and e-mail. Promotional content about Islamic banking disseminated through these media serves as a stimulus that is received by consumers' senses, processed cognitively, and then evaluated. The way consumers interpret and process such information can influence their decision-making when choosing Islamic banking products in Indonesia.

This study focuses more specifically on the use of social media as a promotional tool, particularly emphasizing the role of online religious figures (referred to as Online Religious Leaders), online advertising, and bank management performance as key variables influencing customers’ decisions to use Islamic financial products. However, few studies have critically examined the effectiveness of social media in promoting Bank Syariah Indonesia's products. This study will be considered successful if it can accurately identify the influence of online religious leaders, online advertising, and Islamic bank management performance on consumer decisions when choosing Islamic banking products in Indonesia. The second issue lies in the difficulty of aligning the assumptions, theoretical concepts, and empirical findings—most of which are derived from studies conducted in foreign contexts—with the unique conditions of Islamic banking in Indonesia. These contextual differences pose challenges for the practical application of foreign-developed research models to Indonesia's socio-cultural, religious, and economic environment. Therefore, the localized adaptation and validation of theoretical models are essential for ensuring their relevance and applicability.

According to Kotler (2021), a service is any act or performance offered by one party to another that is essentially intangible and does not result in ownership of the service. Service production may or may not be tied to physical products. Parasuraman (2013) similarly defines services as all activities or benefits offered by one party to another that are fundamentally intangible and do not result in ownership. Service production may or may not be associated with physical product production. Services have several key characteristics. The first is intangibility, which means that services cannot be seen, touched, tasted, heard, or smelled before purchase. The second is inseparability, which refers to the simultaneous occurrence of service production and consumption, often involving direct interaction between service providers and customers. The third is variability, indicating that the quality of services may vary depending on who provides them, when, where, and how they are delivered. The fourth characteristic is perishability, which means that services cannot be stored for future use or sale; unused service capacity at any given time cannot be saved for future use or sales. According to Kotler (2021), service products are measured using several important dimensions to evaluate the quality, effectiveness, and customer perceptions of the services provided. One of the most important is service quality, which Kotler measures using the SERVQUAL model developed by Parasuraman, Zeithaml, and Berry. This model focuses on five dimensions: tangibles (the appearance of physical facilities, equipment, personnel, and communication materials), reliability (the ability to deliver services dependably and accurately as promised), responsiveness (the willingness to help customers and provide prompt service), assurance (the knowledge and courtesy of employees and their ability to inspire trust), and empathy (providing caring and individualized attention to customers).

Customer satisfaction is another critical measure, emphasized by Kotler (2021) as the primary indicator of the success of a service. Satisfaction is gauged by comparing customer expectations with their actual experiences. When services exceed expectations, satisfaction increases, and vice versa. Perceived value is also important, referring to the difference between the benefits that customers perceive from a service and the cost or price they pay. This includes both the functional and emotional values of the service products. This value can be measured through surveys exploring customers' perceptions of the benefits they receive relative to their expenses. Kotler (2021) highlighted customer retention as a key indicator of service performance. It reflects how well a company maintains a long-term customer base, which often indicates customer satisfaction and service quality. Similarly, trust and loyalty are crucial. Customers with high trust in a service brand are more likely to remain loyal and continue using the service. Customer loyalty is frequently measured using indicators such as the Net Promoter Score (NPS), which measures the extent to which customers are willing to recommend a service to others. Customer experience, as emphasized by Kotler (2021), is a vital aspect that must be measured because it determines the overall perception of a service. Customer experience can be assessed using direct feedback, experience surveys, and customer journey analyses. Additionally, the number and types of customer complaints received by a company can serve as indicators of issues with the services provided. Effective complaint management enhances customer satisfaction and loyalty. Through this approach, Kotler (2021) demonstrated that service product measurement is not only focused on the functional elements of the service but also includes emotional and psychological aspects that influence the customer's overall perception.

With the development of digital technology, service theory has also shifted. The E-SERVQUAL model developed by Parasuraman (2013) was designed to measure service quality in the context of electronic services. The primary goal of E-SERVQUAL is to identify the factors that influence customers’ perceptions of service quality in digital environments, such as online banking, e-commerce, and web-based services. The dimensions used to measure electronic service quality include efficiency, reliability, user needs fulfillment, privacy, and security. Some theories classified as middle-range theories within the scope of E-service include the E-SERVQUAL theory by Parasuraman et al. (2013), the electronic value perception theory by Sweeney and Soutar (2001), the theory of trust and security in electronic services by McCole (2002).

Grand theories related to consumer behavior are not directly applied in this research model but serve as foundational theories that are later developed into middle-range theories and subsequently into applied theories for use in this study. The grand theory related to consumer behavior used in this research is Lewin’s (1951) theory of social influence, which laid the foundation for the Theory of Reasoned Action developed by Ajzen (1977). Kurt Lewin introduced the concept that individual behavior is influenced by the social and psychological environment, and that group interactions create dynamics that affect individual behavior. Lewin emphasized the importance of social context in determining human behavior. The middle-range theory used in this research is the Stimulus-Organism-Response (SOR) theory by Russell (1974), which discusses social factors influencing consumer behavior and environmental stimuli that affect consumer decision-making. In the banking context, these theories explain how banks can understand stimulus factors and consumer needs, which have implications for the decision to use banking products. Applied theories in this study, which are a summary of previous research findings, include works by Sriram et al. (2021), Žák and Hasprová (2021), Gong et al. (2021), Tran and Uehara (2023), among others, as shown in the following diagram.

According to Ajzen (1971), the Theory of Reasoned Action (TRA), which serves as the main theory in this study, explains that individual behavior is influenced by beliefs about the consequences of a behavior, evaluation of those consequences, and social pressures that affect the behavior. TRA was first developed by Martin Fishbein and Icek Ajzen in 1967 and later expanded into the Theory of Planned Behavior (TPB) in the 1980s. TRA asserts that a person's behavior is shaped by their beliefs about the outcomes of that behavior. These beliefs are divided into two types: outcome beliefs, which are beliefs about the consequences of performing a specific behavior, and value beliefs, which are the individual's evaluation of the importance of those consequences. After considering their beliefs, individuals evaluate whether a behavior will result in desirable or undesirable outcomes. This evaluation reflects how they perceive the consequences of their actions. Subjective norms represent the perceived social pressure to perform or not perform a behavior. This includes normative beliefs, or the individual's perception of whether important others approve or disapprove of the behavior, and motivation to comply, which is the extent to which individuals are willing to conform to the expectations of those they value. Intention is an individual's desire to perform a specific behavior, which is influenced by behavioral beliefs, outcome evaluations, and subjective norms. This intention is the main predictor of whether an individual will engage in a behavior. The actual behavior is then predicted based on the intention, although TPB later adds perceived behavioral control to explain an individual's ability to carry out their intentions.

The TRA assumes that behavior can be predicted from the intention to perform it. This intention is formed through an evaluative process based on beliefs about the outcomes of the behavior and perceived social pressure. Beliefs provide the foundation for intention formation. If someone believes that exercise is beneficial for their health, they are more likely to engage in it. The evaluation of outcomes plays a crucial role in determining whether an individual finds the results of their behavior worthwhile. Subjective norms account for the social influence on personal goals. If people around the individual support or expect them to engage in certain behaviors, such influence can reinforce the individual's intention. Empirical studies applying TRA include Saleem et al. (2022), who examined online shopping consumers in Lahore, Pakistan, using the Technology Acceptance Model (TAM) in addition to TRA. This study identified perceived security, perceived benefits, personal innovativeness, and perceived ease of use as factors influencing online purchase intentions, mediated by consumer attitudes. Additional studies include those of Raut (2020), Ezeh (2021), and Jamshidi (2020).

The Stimulus-Organism-Response (SOR) theory introduced by Woodworth (1921) in the 20th century is a psychological theory that explains how external stimuli influence internal processes within an individual, which in turn produces specific responses. The SOR theory states that individual responses to stimuli are not solely determined by the stimuli but also by internal processes occurring within the organism. These internal processes include physical states, emotions, beliefs, and past experiences. Hull (1943) played a key role in developing SOR theory, arguing that responses are influenced by motivation and need. Woodworth’s version of the SOR model emphasizes the interaction between external stimuli, the human response from the organism, and internal processes such as perception, motivation, and physical state. The theory posits that external stimuli influence individuals' perception and interpretation of their environment and that internal motivators, such as needs, desires, and goals, shape their responses to these stimuli. Physical conditions, such as fatigue or health, also influence individual responses.

Russell (1974) later developed the SOR theory further in the context of human emotion and perception. Key elements include the stimulus, which refers to environmental inputs received by an individual, such as visual or auditory cues; the organism, which encompasses internal cognitive, emotional, and physical processes; and the response, which includes behavioral or emotional reactions influenced by the interpretation of the stimulus. SOR has been applied in several studies.

According to Nurul (2017), religious leaders play an important role in marketing Sharia products, including in the banking sector. Their influence in the online world makes the promotion of Sharia products increasingly widespread, thus impacting the decision to become a customer of Islamic banks. Tran and Uehara (2023) used Key Opinion Leader (KOL) features such as reputation, perceived congruence, and production involvement, as well as KOL content features such as content quality, aesthetic quality, and interactive content. They concluded that among the six KOL and content features, only perceived congruence and interactive content positively influenced trust, thereby strengthening respondents' intention to purchase green fashion products. However, reputation, production involvement, content quality, and aesthetic content features did not significantly affect consumer purchasing behavior. Meanwhile, Gong et al. (2021), who used respondents who were buyers of short videos on virtual platforms, explained that KOL features—including reputation, perceived congruence, and production involvement—as well as content features—including content quality, aesthetic quality, and interactive content—had both direct and indirect effects on consumer buying interest, with emotional response and virtual touch as mediating variables. Based on the above explanation, the research hypothesis is formulated as follows:

H1: Online religious leaders significantly influence the decision to become a Bank Syariah Indonesia customer.

Transparency through online advertising, religious leader support, testimonials and reviews, perceived security of the digital platform, alignment with Sharia values, educational content in online advertising, sustainability, and brand reputation can affect consumer trust in Sharia banking products, which in turn influences the decision to use Sharia banking products. Sriram et al. (2021), who examined the effect of customer characteristics on consumer purchasing behavior using the SOR approach, explained that advertising appeal—including creative characteristics, attention-grabbing details, emotional appeal, and celebrity endorsements—all affect purchasing behavior. However, Shaouf et al. (2016) concluded that there is no significant relationship between web advertising visual design (WAVD) and purchasing behavior. Therefore, the hypothesis of this study is as follows:

H2: Online advertising significantly influences the decision to become a customer of Bank Syariah Indonesia.

Important aspects influencing consumer decisions to use Islamic bank products include Sharia governance, operational efficiency, product innovation, consumer trust, customer satisfaction, bank reputation and image, and risk management. These aspects are important for creating consumer trust and satisfaction, which ultimately influence their decision to use Islamic banking products. In research conducted by Aslam et al. (2021) examined the influence of environmental management practices on financial performance, mediated by environmental performance, in companies listed on the Nikkei from 2007 to 2018. They concluded that after the implementation of environmental management practices, there was a positive influence on financial performance of the companies. Environmental performance perfectly mediated the influence of environmental management practices on financial performance. This is contrary to the research conducted by Li and Wu (2017) on companies that implemented environmental management systems listed on the Shanghai and Shenzhen Stock Exchanges. The results showed that after implementing environmental management systems, profitability, sales, and operational efficiency significantly declined owing to the requirements of the environmental management system. Based on the above explanation, the following hypothesis can be formulated.

H3: Bank management performance significantly influences the decision to become a customer of Bank Syariah Indonesia.

The performance of Islamic bank management is strongly influenced by various interrelated factors such as Sharia governance, operational efficiency, and product innovation. These aspects are important for creating consumer trust and satisfaction, which ultimately influence the decision to use Islamic banking products. Le Tan et al. (2021) concluded that trust in security risks is the most influential factor in online shopping behavior. Similarly, Manzoor (2020) concluded that social media significantly influences trust, which also affects purchasing behavior. However, the influence of social media on purchasing behavior was greater than the influence of trust on purchasing behavior. Therefore, the hypothesis of this study is as follows:

H4: Trust in product value significantly influences the decision to become a customer of Bank Syariah Indonesia.

According to Solomon (2023), one of the factors that influence purchasing decisions is psychological aspects, including motivation, perception, learning, and memory. According to Faiza (2023), who studied the factors causing customers to switch from conventional to Islamic banks in Oman using data analysis with SEM, AMOS, and SPSS tools, it was concluded that interaction quality, satisfaction, customer trust, and religion significantly influence the switch to Islamic banks, whereas result quality, price, social prestige, and physical environment do not significantly influence the decision to switch. Therefore, the following hypothesis was proposed:

H5: Motivation significantly influences the decision to become a Bank Syariah Indonesia customer.

Žák and Hasprová (2021) explored the contribution of opinion leaders in the global digital environment and concluded that influencers or opinion leaders have a major contribution to the success of product marketing by companies. Consumers and prospective customers face many product choices. Recommendations, credibility, and expertise from influencers in providing product information motivate their followers to buy the recommended product. Therefore, the following hypothesis was formulated:

H6: Online religious leaders influence the decision to become a Bank Syariah Indonesia customer when mediated by motivation.

Mansori et al. (2020), who used SEM to analyze respondents who were users of Islamic microfinance schemes in Malaysia, concluded that religiosity and support from religious leaders could influence the intention to use Islamic microfinance products. Intention is a further response to customers’ trust in Islamic microfinance institutions. The more obedient individuals are to religious leaders, the more likely they are to use Islamic financial products. This is in line with the findings of Tran and Uehara (2023), who conducted research on green fashion users using the stimulus-organism-response (SOR) approach and stated that consumer purchase intention is significantly influenced by Key Opinion Leaders (KOLs) through the trust variable. Therefore, the following hypothesis was proposed:

H7: Online religious leaders significantly influence the decision to become a Bank Syariah Indonesia customer when mediated by trust in product value.

Janschitz and Penker (2022) aimed to explain the impact of social media on motivation and intention to buy environmentally friendly cosmetic products and concluded that social media is an important factor influencing consumer attitudes, subjective norms, altruistic motivation, and egoistic motivation, as well as the impact of these variables on the antecedents of the intention to purchase environmentally friendly cosmetic products. Therefore, the hypothesis of this study is formulated as follows:

H8: Online advertising significantly influences the decision to become a Bank Syariah Indonesia customer when mediated by motivation.

Transparency through online advertising, religious leader support, testimonials and reviews, perceived security of digital platforms, alignment with Sharia values, educational content in online advertising, sustainability, and brand reputation can influence consumer trust in Islamic bank products, which in turn affects the decision to use Islamic banking products. Sriram et al. (2021) conducted an exploratory study to determine the factors of social media advertising that significantly influence purchasing behavior. They found that celebrity endorsement content factors and emotional appeal of advertisements significantly influence purchasing behavior, whereas the creative characteristics of advertisements do not significantly influence purchasing behavior. Therefore, the following hypothesis is proposed:

H9: Online advertising significantly influences the decision to become a Bank Syariah Indonesia customer when mediated by trust in product value.

Hati et al. (2021) tested the influence of product knowledge on the intention to invest in Islamic banks and concluded that risk perception has a dual role in influencing the intention to invest in Islamic banks, namely, as an independent variable that influences the intention to invest in Islamic banks and as a variable mediating the relationship between product knowledge and the intention to invest in Islamic banks. Product knowledge significantly affects perceived product quality but does not significantly affect the intention to invest in Islamic banks. Product perception is one of the factors that motivates customers to form behavioral intentions—in this case, the intention to invest in Islamic banks. Therefore, the hypothesis of this study is as follows:

H10: Management performance influences the decision to become a customer of Bank Syariah Indonesia when mediated by motivation.

The important aspects that influence consumer decisions to use Islamic bank products include (1) Shariah Governance, (2) Operational Efficiency, (3) Product Innovation, (4) Consumer Trust, (5) Customer Satisfaction, (6) Bank Reputation and Image, and (7) Risk Management. These aspects are crucial for building consumer trust and satisfaction, which ultimately influence consumers’ decisions to use Islamic banking products. Stravinskiene et al. (2021) examined the influence of corporate reputation on consumer trust among pharmacy network users in Lithuania. The results concluded that corporate reputation—measured by dimensions such as financial performance, vision and leadership, products and services, work environment, social responsibility, and emotional appeal—affects consumer trust in the company. Therefore, the hypothesis of this study is as follows:

H11: Management performance significantly influences the decision to become a customer of Bank Syariah Indonesia when it is mediated by trust in product value.

The research model that will be used by researchers related to Islamic banking is as follows (see Figure 1).

The population of this study comprises all customers of Bank Syariah Indonesia located in the Greater Jakarta area, specifically in the cities of Jakarta, Bogor, Depok, Tangerang, and Bekasi. The selection of these locations was based on demographic considerations: West Java and Jakarta represent the provinces with the largest and second largest Muslim populations in Indonesia, respectively (Kementerian Agama Republik Indonesia, 2025). These five cities are part of the Jabodetabek metropolitan region, which serves as a buffer zone to the capital and features a significantly higher population density than other provinces, thereby providing a representative sample of the target population.

The sample for this study consisted of 325 customers of Bank Syariah Indonesia residing in Jakarta, Bogor, Depok, Tangerang, and Bekasi. The determination of the sample size aligns with the analytical technique employed in this research, namely (SEM). According to Hair et al. (2010), complex SEM models require a sample size ranging from 300 to 500 to achieve sufficient statistical power and accurate estimation. Therefore, the use of 325 respondents ensured the reliability and robustness of the results obtained through SEM analysis. Prior to conducting the full-scale survey, the research instrument (questionnaire) was subjected to validity and reliability tests. Validity testing was performed to assess whether the questionnaire effectively measured the intended constructs, whereas reliability testing aimed to evaluate the consistency of the instrument in measuring the same phenomenon over repeated applications.

The sampling process in this study involved selecting both the type and number of samples that constituted the subjects or objects of analysis. Respondents were drawn from the customer base of Bank Syariah Indonesia in the Jabodetabek region, namely Jakarta, Bogor, Depok, Tangerang, and Bekasi. This region was selected because of its significant concentration of Muslim residents, ranking among the top two provinces nationally (Kementerian Agama Republik Indonesia, 2025), and its strategic role in the socio-economic landscape surrounding the capital. This study employed non-probability sampling, specifically convenience sampling, to recruit respondents. As defined by Sekaran (2013), non-probability sampling does not provide every element in the population with an equal chance of selection. The convenience sampling approach allows the researcher to deliberately select respondents who are already customers of Bank Syariah Indonesia and are readily accessible at the time of the data collection. This method was deemed appropriate because of its practical advantages and alignment with the study’s objective of capturing insights from current customers with experience with Sharia banking products and services.

The measurement model in this study uses constructs adopted and adapted from validated sources in previous empirical research. The Online Religious Leader variable is operationalized through dimensions of reputation, perceived fit, content quality, and charisma—based on Kardes et al. (2018) and further supported by KOL studies such as Tran and Uehara (2023) and Gong et al. (2021). Online Advertisement is measured through advertisement credibility, emotional appeal, and creative characteristics, adapted from Sriram et al. (2021). The Management Performance construct follows Abdullah (2018), which includes liquidity, product and technology innovation, reputation and stability, and promotional programs. Motivation was measured using four dimensions (religious, social, functional, and emotional) based on the Theory of Planned Behavior (Ajzen, 1977) and Solomon’s Consumer Behavior (2023). Finally, the Decision-Making Process was measured through five stages—need recognition, information search, evaluation of alternatives, final decision, and post-purchase—adapted from Solomon (2023). All constructs were measured using a Likert scale (see Table 1).

To ensure that respondents could clearly understand the questionnaire items, the questionnaire was pre-tested (Sekaran, 2013). This preliminary test involved a small group of respondents to evaluate whether the wording and content of the questionnaire were well understood. Moreover, the pre-test served to minimize potential bias and assess whether there were differences in characteristics between respondents who returned the questionnaire and those who did not. In this stage, late responses (those returned after the stated deadline) are used as proxies for non-response data. The comparison was conducted using an independent samples t-test. If the p-value is greater than or equal to 0.05, this indicates that there are no statistically significant differences between the early and late respondents. Thus, the responses from late participants are deemed representative of those who did not respond (Hair et al., 2010). This result implies that there is no response bias and that the data from both early and late respondents can be combined in the final analysis. After the pre-test confirmed the validity of the questionnaire in line with the research objectives, the finalized questionnaire was distributed directly to the targeted respondents. Once collected and screened for completeness, the data were analyzed using structural equation Modelling (SEM). SEM was chosen for its ability to simultaneously assess complex relationships among multiple latent constructs, making it appropriate for this study's conceptual framework.

The respondents in this study comprised individuals who were currently or had previously used products and services from Bank Syariah Indonesia (BSI). The respondent profile presents a descriptive overview of the demographic data collected, including gender, age, and area of residence. The questionnaires were distributed across five major cities in the Greater Jakarta area (Jabodetabek) via Google Forms and in-person distribution. Out of 353 total submissions, 325 respondents completed the questionnaire appropriately and were deemed eligible for inclusion in this research, while 28 were excluded due to incomplete or invalid responses.

Based on the data in Table 2, 84.61% of the respondents were male (275 individuals), and 15.39% were female (50 individuals), indicating a dominance of male participants in this study. Regarding age, the largest group of respondents (36.31%) were 25 years old or younger (118 individuals), followed by those aged 26–35 years (106 individuals, 32.62%), and those aged 36–45 years (101 individuals, 31.07%). These results suggest that most BSI users in this study were within the productive age range. In terms of geographic distribution, the responses were fairly evenly spread across the target areas. Bekasi, Bogor, and Depok each accounted for 20.62% of the respondents (67 individuals each), followed by Jakarta (64 individuals, 19.69%) and Tangerang (60 individuals, 18.45%). This distribution confirms the balanced sampling approach across the key metropolitan zones in Jabodetabek.

Based on the results of data processing using SPSS version 25, descriptive statistics were generated for each research variable, as shown in Table 3.

The findings provide insights into how respondents perceived each construct measured in the study. For the variable of online religious leaders, the maximum score was 64, and the minimum was 30, with a mean value of 56.11. When divided by the 13 items used to measure this variable, the average score per item was approximately 4.31, which fell into the "good" category on the Likert scale. This indicates that the respondents generally perceived the role and presence of online religious figures positively. Regarding the online advertising variable, the scores ranged from 19 to 45, with a mean of 38.95. Given that the variable was measured using nine items, the resulting mean score per item was 4.33. This score also lies within the "good" category, suggesting that respondents have a favorable perception of BSI's online advertising strategies. Similarly, the management performance variable had scores ranging between 29 and 63, with a mean of 55.94. With 13 items in this construct, the average item score was 4.30, again reflecting a positive perception from respondents toward the effectiveness and performance of the bank's management. The motivation variable recorded scores between 28 and 64, with a mean of 54.83. When distributed across 13 items, the average score per item was 4.22, reflecting strong and favorable motivation among BSI users. The trust in product value variable had a maximum score of 74 and a minimum of 35, with a mean score of 63.51. With 15 indicators in this variable, the average score per item was 4.23, indicating that respondents had a high level of trust in the value and integrity of BSI’s financial products. Finally, the variable representing the decision to become a customer had scores ranging from 30 to 73, with a mean of 56.30. Dividing this across 15 items gives an average score of approximately 3.75 per item, which is categorized as “good,” though slightly lower than the others. This shows that while respondents generally have a positive view of their decision to become a customer, their perception is somewhat more moderate compared with other constructs.

The first step in assessing the outer model is to examine indicator reliability through outer loading values. A high outer loading indicates a strong commonality between a construct and its indicators. According to Hair et al. (2010), the acceptable minimum threshold for outer loading is 0.70. The following table presents a summary of indicator reliability, internal consistency reliability (Cronbach's alpha and composite reliability), and convergent validity (AVE) for each construct measured in the study. See Table 4

The results show that all constructs met the criteria for indicator reliability, with outer loading values above 0.70. Among the items for the online advertising variable, the highest outer loading was observed for item IO3.3 (0.802), which suggests that this item best reflects the construct. Items related to ad credibility, emotional appeal, and creative characteristics also performed well and should be improved to achieve higher measurement consistency. For the Management Performance variable, all outer loadings fell within the range of 0.716 to 0.783, indicating an acceptable level of validity but with room for refinement. The Decision to Become a Customer variable showed strong convergent validity, with all items ranging between 0.821 and 0.860. For the Product Value construct, indicators KP3.2, KP4.1, and KP5.2 had the highest outer loadings, showing that they were highly representative and should be retained. In terms of motivation, the strongest indicators were MTA1.3, MTA2.2, MTA2.3, and MTA4.2, all with outer loading values above 0.806. These items should be prioritized for future measurements, whereas others should be refined for improved reliability. Finally, the indicators measuring Religious Leaders demonstrated adequate reliability, with outer loading values between 0.707 and 0.779. Regarding internal consistency reliability, all constructs achieved Cronbach’s alpha and composite reliability scores above 0.6, indicating that the scales used were reliable. The AVE values, which assess convergent validity, were above the threshold of 0.50 for all constructs. This means that each construct explained more than 50% of the variance of its indicators, confirming a satisfactory convergent validity. The highest AVE was observed in the Decision to Become a Customer variable (0.703), while the Islamic Bank Management variable recorded the lowest AVE (0.559), though still acceptable.

Discriminant validity testing was conducted to evaluate how distinct one construct is from another in capturing different phenomena. Researchers typically apply several methods in this assessment, such as the Fornell-Larcker criterion, cross-loading analysis, and the heterotrait-monotrait ratio (HTMT) (Hair et al., 2010). The first criterion considered for discriminant validity is the Fornell-Larcker criterion. According to this method, the square root of a construct’s AVE should be greater than the highest correlation it shares with other constructs. As shown in Table 5, all constructs met this requirement, indicating acceptable discriminant validity.

The next discriminant validity check involved a cross-loading analysis. This test requires that the outer loading of each indicator on its associated construct be higher than its loading on the other constructs. As shown in Table 4, all indicators satisfied this criterion. For instance, IO1.1 loads 0.780 on Online Advertising, which is greater than its loadings on any other construct. Similar patterns were observed across all constructs, indicating good discriminant validity.

The final criterion, the heterotrait-monotrait ratio (HTMT), measures the mean of all correlations across constructs. HTMT values should not exceed 0.9. Based on Table 6, all HTMT values fell below this threshold, confirming acceptable discriminant validity.

Since all constructs satisfy the Fornell-Larcker, cross-loading, and HTMT criteria, it can be concluded that each construct is empirically distinct from the others and successfully captures unique phenomena not represented by other constructs in the model. Therefore, all indicators met the discriminant validity test criteria.

To better understand the relationships between the studied variables and their contribution to customer decision-making in choosing Bank Syariah Indonesia, a path analysis was conducted, as shown in Table 7. This analysis aimed to examine both the direct and indirect effects of exogenous and endogenous variables within the structural equation model. By assessing the path coefficients, t-statistics, and p-values, the strength and significance of each hypothesized relationship were identified. This process also includes evaluating the explanatory power of the model through R-squared (R²) values, which indicate the extent to which independent variables can explain the variance in the dependent variables.

The results of the direct effects suggest that online advertising has a significant and positive impact on three key variables: trust in product value, motivation, and the ultimate customer decision-making. Specifically, the path coefficient from online advertising to product value trust was 0.404, with a t-statistic of 5.665, indicating a strong and statistically significant relationship. Likewise, online advertising is significantly associated with customer decisions (0.236) and motivation (0.382), reinforcing the role of digital promotional strategies in shaping consumer perceptions and behaviors. Trust in product value itself shows a substantial influence on customer decisions, with a coefficient of 0.420 and t-statistic of 7.407. This demonstrates that customer confidence in Sharia-compliance and the benefits of banking products play a central role in decision-making. Similarly, bank management performance positively affects product value trust (0.384), motivation (0.372), and directly contributes to customer decisions (0.252), underscoring the importance of operational excellence and customer satisfaction strategies.

Motivation, as a mediating factor, has a direct and significant impact on the decision to become a customer (0.409), highlighting its role as a critical psychological driver of customer loyalty. The influence of online religious leaders also proves significant across all paths: increasing product value trust (0.382), boosting motivation (0.348), and directly influencing the decision to become a customer (0.232). These findings validate the social and spiritual influence of religious figures in digital environments, particularly in contexts that align with religious values. In addition to direct influences, the analysis revealed meaningful indirect effects. Online advertising exerts a significant indirect influence on customer decisions through product value trust (0.170) and motivation (0.156). Similarly, the indirect effect of bank management is mediated by motivation (0.152) and product value trust (0.161), confirming the presence of complex, multi-pathway influences. Furthermore, online religious leaders affect customer decisions through indirect channels via both motivation (0.142) and product value trust (0.160), strengthening the argument that these figures indirectly shape consumer choices through layered, perceptual mechanisms.

The model’s explanatory power is captured by the R-squared (R²) values. Product value trust has an R² of 0.370, indicating that 37% of its variance is explained by independent variables. Customer decision shows a notably strong explanatory power with an R² of 0.796, suggesting that nearly 80% of decision-making can be accounted for by this model. Motivation is moderately explained by an R² of 0.329, reflecting its multifaceted nature.

The integration of the Theory of Reasoned Action (Ajzen, 1985) and the Stimulus-Organism-Response (SOR) model (Russell, 1974) underscores that external marketing and credibility stimuli activate internal cognitive and emotional responses, which ultimately shape behavioral outcomes. Online religious figures and digital influencers act as referent powers, shaping subjective norms that foster trust and motivation (Casidy et al., 2016; Žák & Hasprová, 2021). This aligns with the findings of Mansori et al. (2020), who emphasized the significant role of religious endorsement in enhancing the intention to use Islamic financial products. Online advertising also has a robust influence, both directly and indirectly, through motivational pathways. When designed with persuasive storytelling, emotional resonance, and Shariah-based values, digital ads can activate motivational states that push consumers toward favorable actions. These findings resonate with those of Sriram et al. (2021), who concluded that creative appeal, emotional relevance, and influencer support significantly drive consumer behavior. Furthermore, the ability of online ads to foster trust in product value mirrors the results of Nemati et al. (2022), who found that advertising perception affects purchase behavior when mediated by attitude and trust.

Similarly, management performance plays a pivotal role in fostering consumer trust and motivation, thus indirectly influencing customer choice. This study reinforces the conclusions of Aslam et al. (2021), who found that firms with stronger managerial and environmental systems enjoy higher financial performance and consumer confidence. Moreover, as noted by Stravinskiene et al. (2021), corporate reputation—built on product quality, social responsibility, and financial governance—strongly affects consumer trust and their purchase decisions. The relevance of this is particularly visible in BSI’s context of BSI, where transparent and Shariah-compliant management is seen as central to customer engagement. Collectively, these findings reaffirm the layered nature of behavioral influence on Shariah banking decisions. Motivation and trust emerge as crucial mediating constructs that channel the impact of external stimuli into actual customer actions. These results are consistent with the mediating effects described by Gong et al. (2021) and Althuwaini (2022), who highlighted the importance of influencer credibility, interactive content, and perceived value in converting intention into behavior. Therefore, BSI’s strategic focus should not only rest on visibility and promotional presence but also on enhancing perceived credibility, emotional appeal, and Shariah alignment in both product communication and managerial actions.

This study finds that several controllable external factors significantly influence customers’ decisions to become clients of Bank Syariah Indonesia (BSI). Online religious figures have a direct and significant impact on customer decisions, indicating that their recommendations and religious authority play key roles in shaping perceptions of Sharia-compliant banking. Similarly, online advertising significantly affects customer decisions by influencing perceptions, building trust, and enhancing motivation through persuasive and informative content. Management performance is also a crucial determinant that directly and indirectly influences customer decisions through its impact on motivation and trust in product value. High-quality service, operational transparency, and adherence to Sharia principles contribute to customer confidence. Trust in product value and customer motivation are key mediating variables. Both are not entirely intrinsic but can be strengthened through marketing strategies, communication, and excellent service. Hence, the decision to choose BSI’s services is the result of a dynamic interplay between controlled stimuli (advertising, management, religious influencers) and internal responses (motivation, trust).

Based on these findings, it is recommended that BSI strengthen its collaboration with credible online religious figures by incorporating their testimonials and endorsements into both marketing campaigns and educational content to build greater trust and product understanding. Additionally, BSI should improve its online advertising strategy by showcasing its achievements and managerial performance to reinforce the public’s perception of professionalism and reliability. Given that this study used a limited sample (325 respondents) and applied SEM-PLS for the analysis, future researchers are encouraged to expand the model by adding new variables, increasing the sample size, and exploring alternative analytical methods to enhance the robustness and generalizability of the findings.