This study integrates the Unified Theory of Acceptance and Use of Technology (UTAUT) ( Venkatesh et al., 2003) with Social Cognitive Theory (SCT) ( Bandura, 1986) to understand technology adoption behavior among internal users of a digital banking system. The research aims to explain how three primary constructs Performance Expectancy (PE), Effort Expectancy (EE), and Facilitating Conditions (FC) influence Behavioral Intention (BI) to use BRISPOT, and how BI affects Use Behavior (UB). Furthermore, this study incorporates Trust as a moderating variable between BI and UB to address the commonly observed intention–behavior gap in digital platform usage.

UTAUT provides a robust base to predict technology acceptance in organizational contexts. Performance Expectancy (PE) is considered the strongest predictor, referring to the perceived benefits of using the system in enhancing job performance. Effort Expectancy (EE) relates to the degree of ease associated with using the technology. Facilitating Conditions (FC) represent the extent to which users believe that an adequate organizational and technical infrastructure is available to support system use. These three constructs are theorized to influence Behavioral Intention (BI), which in turn predicts actual Use Behavior (UB).

However, the behavioral intention–use behavior linkage often fails to hold due to intervening contextual factors. In response, Trust is incorporated into the framework as a moderating variable that could strengthen or weaken this relationship ( Pavlou, 2003; Gefen et al., 2003). Trust is particularly important in financial services, where perceived risks, system integrity, and information security significantly affect user behavior.

Social Cognitive Theory (SCT) complements UTAUT by offering a lens through which individual cognition (e.g., self-efficacy) and social influence (e.g., peer modeling) contribute to behavior change. Employees with high self-efficacy and observational learning are likely to find the system easier to use and more beneficial, which directly supports PE and EE.

2.2.1 Performance expectancy and behavioral intention

Performance Expectancy (PE) is defined as the degree to which an individual believes that using the system will help achieve job-related goals ( Venkatesh et al., 2003). In BRISPOT’s context, PE reflects the system’s perceived ability to accelerate the loan process, reduce paperwork, and improve customer service delivery. Prior research consistently confirms that PE significantly influences the intention to use technology ( Alalwan et al., 2017; Dwivedi et al., 2019). H1:

Performance Expectancy has a positive effect on Behavioral Intention to use BRISPOT.

2.2.2 Effort expectancy and behavioral intention

Effort Expectancy (EE) refers to the ease of use associated with the system. If users perceive the system as easy to navigate, they are more likely to develop favorable attitudes and intentions toward its usage. For first-time users or employees less familiar with digital interfaces, perceived ease plays a critical role in adoption ( Venkatesh et al., 2003; Williams et al., 2015). H2:

Effort Expectancy has a positive effect on Behavioral Intention to use BRISPOT.

2.2.3 Facilitating conditions and behavioral intention

Facilitating Conditions (FC) encompass the availability of resources, training, and support infrastructure to enable system use. Though traditionally linked with actual usage, in this study FC is expected to influence intention, given that the presence of support systems shapes user perceptions before adoption ( Rogers, 1995; Venkatesh et al., 2012). H3:

Facilitating Conditions have a positive effect on Behavioral Intention to use BRISPOT.

2.2.4 Behavioral intention and use behavior

Behavioral Intention (BI) is a key determinant of actual technology usage. A strong intention to use a system generally results in higher engagement and frequency of use. However, this linkage is influenced by both internal (e.g., confidence) and external (e.g., trust, organizational constraints) factors ( Ajzen, 1991; Venkatesh et al., 2003). H4:

Behavioral Intention has a positive effect on Use Behavior of BRISPOT.

2.2.5 The moderating role of trust

Trust refers to the belief that the system is reliable, secure, and capable of performing tasks as intended. In the context of BRISPOT, trust influences how confidently users convert their intentions into actual usage. Trust mitigates perceived risk and enhances willingness to rely on the system ( Gefen et al., 2003; Pavlou, 2003). H5:

Trust positively moderates the relationship between Behavioral Intention and Use Behavior of BRISPOT.

Based on the Unified Theory of Acceptance and Use of Technology (UTAUT) and Social Cognitive Theory (SCT), this study develops a comprehensive theoretical model that integrates behavioral and cognitive perspectives to explain the adoption of BRISPOT. The model highlights how performance expectancy, effort expectancy, and facilitating conditions influence behavioral intention, which in turn drives use behavior. Furthermore, trust is introduced as a moderating variable to capture the contextual and psychological factors that may strengthen or weaken the intention–behavior relationship.

This integrated framework provides a nuanced understanding of internal technology adoption in a banking environment, emphasizing both system-level enablers and individual-level perceptions. By bridging the gap between intention and actual use, the model aims to generate practical insights for enhancing the effectiveness of digital transformation initiatives in the financial services sector.

The hypothesized research model is illustrated in Figure 1.

This study adopts a quantitative, explanatory research design to examine the influence of Performance Expectancy, Effort Expectancy, and Facilitating Conditions on Behavioral Intention, and the subsequent effect on Use Behavior of the BRISPOT application, with Trust as a moderating variable. The research is grounded in the Unified Theory of Acceptance and Use of Technology (UTAUT) and enriched by the Social Cognitive Theory (SCT), providing a robust framework for analyzing technology adoption behavior among internal banking professionals.

The population of this study consists of relationship managers (RMs) from Bank Rakyat Indonesia (BRI) who are assigned to handle Small and Medium Enterprise (SME) credit segments and are authorized users of the BRISPOT digital application. The unit of analysis is individual RMs.

Using the Hair et al. (2010) guideline for Partial Least Squares Structural Equation Modeling (PLS-SEM), a minimum sample size of 10 times the largest number of structural paths directed at a particular construct is recommended ( Solimun, et al., 2021). As five structural paths are tested, the minimum sample size is 50. To enhance generalizability and validity, a total of 150 valid responses were collected using purposive sampling, targeting only active SME loan officers with at least 6 months of BRISPOT experience.

Primary data were collected using a structured online questionnaire, distributed via internal email and communication platforms over a period of two weeks in early 2025. All questionnaire items were adapted from validated scales in prior studies and measured using a five-point Likert scale (1 = strongly disagree to 5 = strongly agree). The survey consisted of six main constructs: PE, EE, FC, BI, UB, and Trust.

To ensure clarity and validity, the instrument was pre-tested with 10 RMs and refined based on feedback regarding language, terminology, and structure. Extended data, including the questionnaire items and anonymized dataset, are available in the Figshare repository ( https://doi.org/10.6084/m9.figshare.30029947) ( Priyastomo et al., 2025).

Performance Expectancy (PE): Adapted from Venkatesh et al. (2003), measured using four items related to perceived usefulness of BRISPOT in job performance. •

Effort Expectancy (EE): Four items measuring perceived ease of use, drawn from UTAUT and Davis (1989).

All items were adapted and translated to Indonesian using back-translation procedures to maintain construct equivalence.

The data were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM) via SmartPLS 4.0. PLS-SEM was selected due to its suitability for predictive, theory-building research, especially when working with complex models and small-to-medium sample sizes ( Hair et al., 2019; Ubaidillah, et al., 2022).

The analysis followed a two-step approach: •

Measurement Model Evaluation: Validity and reliability were assessed using Composite Reliability (CR), Cronbach’s Alpha (CA), Average Variance Extracted (AVE), and outer loadings. Discriminant validity was evaluated using the Fornell–Larcker criterion and Heterotrait-Monotrait (HTMT) ratio.

To address common method bias (CMB), Harman’s single-factor test was employed and revealed no dominant factor. Procedural remedies included ensuring anonymity, randomization of item order, and assuring participants that there were no right or wrong answers.

All ethical protocols were followed, and informed consent was obtained from participants. No identifying information was collected, and the data were analyzed in aggregate.

To ensure the robustness of the measurement model within the PLS-SEM framework, this study conducted a comprehensive assessment of construct validity, which encompasses both convergent and discriminant validity, as well as internal consistency reliability. This assessment employed key metrics including factor loadings (FL), average variance extracted (AVE), Cronbach’s alpha (CA), and composite reliability (CR), in accordance with established methodological guidelines ( Ahmed, 2024).

As shown in Table 1, all items exhibited standardized factor loadings above the recommended threshold of 0.70, indicating that each observed variable demonstrates substantial contribution to its corresponding latent construct. Furthermore, all constructs reported AVE values exceeding 0.50, confirming satisfactory levels of convergent validity. Reliability indicators also met the acceptable criteria, with both CA and CR values surpassing the 0.70 benchmark, thereby indicating internal consistency and construct reliability ( Ahmed, 2024; Sarstedt et al., 2019).

Following the confirmation of convergent validity and internal consistency, the discriminant validity of the constructs was examined using the Fornell-Larcker criterion. The results, summarized in Table 2, demonstrate that the square root of each construct’s AVE is greater than its correlations with other constructs, thereby providing empirical evidence of discriminant validity.

To reinforce these findings, the heterotrait-monotrait ratio of correlations (HTMT) was also employed as an additional criterion, which is considered a more rigorous test for discriminant validity in variance-based SEM ( Hair et al., 2017; Henseler et al., 2015). As presented in Table 3, all HTMT values fall below the conservative threshold of 0.90, indicating that the constructs are empirically distinct from one another.

Collectively, these results confirm that the measurement model satisfies the necessary conditions for convergent and discriminant validity, as well as internal consistency reliability. Therefore, the constructs are deemed valid and reliable, and are suitable for subsequent structural model analysis.

The results of hypothesis testing are summarized in Table 4, which presents both direct and indirect effects among constructs.

The analysis of direct relationships in this study reveals that Performance Expectancy (X1), Effort Expectancy (X2), and Facilitating Conditions (X3) exert positive and statistically significant effects on Behavioral Intention to Use (Y1). These findings are supported by path coefficients of 0.720 (p = 0.000), 0.669 (p = 0.000), and 0.228 (p = 0.000), respectively, all with t-statistics above the 1.96 threshold and p-values below 0.05, thus confirming the acceptance of the proposed hypotheses. This suggests that the greater the perceived performance benefits and ease of system use, along with the availability of technical and organizational support, the stronger an individual’s intention to adopt the technology.

Furthermore, Behavioral Intention to Use (Y1) is shown to have a robust and significant direct effect on Use Behavior (Y3), with a path coefficient of 0.857 and a p-value of 0.000. This finding underscores behavioral intention as a principal determinant of actual system usage, in line with the core tenets of the Unified Theory of Acceptance and Use of Technology (UTAUT) as introduced by Venkatesh et al. (2003). In contrast, Trust (Y2) does not exhibit a statistically significant direct effect on Use Behavior, as reflected by its minimal path coefficient of 0.005, a t-statistic of 0.528, and a p-value of 0.597. This suggests that trust in the system, while conceptually relevant, may not independently drive actual usage behavior in the absence of strong performance and usability perceptions.

The analysis of indirect effects further validates the mediating role of Behavioral Intention to Use (Y1) in the relationships between the three exogenous variables (X1, X2, X3) and Use Behavior (Y3). Specifically, Performance Expectancy indirectly affects Use Behavior via Behavioral Intention, with a significant path coefficient of 0.617 (p = 0.000). Similar indirect effects are observed for Effort Expectancy and Facilitating Conditions, with significant coefficients of 0.573 (p = 0.000) and 0.352 (p = 0.000), respectively. These results reinforce the critical function of behavioral intention as a psychological mechanism that translates user perceptions into actual system usage.

However, Trust (Y2) again shows no significant indirect influence on Use Behavior through Behavioral Intention, as evidenced by a low path coefficient (0.005) and a p-value of 0.551. This non-significance implies that trust may serve as a baseline or hygiene factor important for ensuring acceptance, yet insufficient on its own to motivate usage behavior without complementary factors such as performance benefits or system usability.

Overall, the findings provide empirical support for the UTAUT framework, which emphasizes the importance of Performance Expectancy, Effort Expectancy, and Facilitating Conditions in shaping behavioral intention, which subsequently drives technology use behavior. Moreover, the results highlight that the integration of Trust into the UTAUT model should be contextually evaluated, as trust may not be a universally salient factor across all technological adoption scenarios. From a managerial standpoint, these insights suggest that system developers and implementers should prioritize enhancing users’ perceptions of performance, ease of use, and institutional support mechanisms to foster sustainable and effective system adoption.

The analysis reveals that performance expectancy exerts a strong and statistically significant influence on behavioral intention to use BRISPOT. This finding corroborates the original proposition of the UTAUT framework ( Venkatesh et al., 2003), which posits performance expectancy as the most influential determinant of technology adoption intention. In the context of this study, relationship managers (RMs) are more likely to adopt BRISPOT when they perceive that the system effectively supports their performance targets, such as accelerating credit evaluation, reducing paperwork, and enhancing service delivery for SME clients. The strength of this relationship suggests that perceived instrumental value is a critical motivator in organizational settings, aligning with prior empirical evidence in financial services ( Alalwan et al., 2017; Dwivedi et al., 2019). Therefore, improving the alignment between BRISPOT’s capabilities and RMs’ job-related performance expectations is essential to enhance adoption rates.

The results also demonstrate that effort expectancy has a significant positive impact on behavioral intention. This finding is consistent with prior research emphasizing the importance of ease of use in technology acceptance models ( Davis, 1989; Venkatesh et al., 2003). Within the operational dynamics of BRI, ease of system navigation, intuitive interfaces, and minimized complexity are key factors influencing adoption decisions. RMs are more inclined to use BRISPOT when the learning curve is perceived as low and when system interactions do not hinder their workflow. The implication is that BRISPOT’s usability especially for users who may not be highly digitally literate should be continuously improved through user-centered design and targeted training interventions. This aligns with SCT’s emphasis on self-efficacy, where individuals with greater perceived control over system use are more motivated to adopt it ( Bandura, 1986).

The empirical findings confirm that facilitating conditions significantly influence behavioral intention, underscoring the role of organizational and technical support in shaping user intentions even prior to actual system use. This is particularly relevant in BRI’s operational environment, where infrastructure readiness, system availability, and responsive technical assistance are essential enablers. While UTAUT traditionally associates facilitating conditions with actual use behavior, this study demonstrates their anticipatory effect on intention formation, consistent with Rogers’ (1995) theory of perceived support during the adoption decision process. It suggests that efforts to improve BRISPOT’s infrastructure such as stable internet access, system uptime, and helpdesk support can indirectly but substantially elevate user intention.

The study finds strong empirical support for the relationship between behavioral intention and use behavior, in line with the foundational assumptions of UTAUT and Theory of Planned Behavior ( Ajzen, 1991). This indicates that RMs who express high intention are indeed more likely to engage with BRISPOT frequently and meaningfully. The high path coefficient and statistical significance suggest minimal friction between motivation and action, provided other conditions remain favorable. However, this linkage also highlights the importance of monitoring behavioral metrics (e.g., login frequency, transaction completion) to assess whether expressed intentions translate into consistent usage, which is a critical metric for digital transformation success.

Contrary to expectations, trust does not significantly moderate the relationship between behavioral intention and use behavior. This finding deviates from previous studies in e-commerce and financial technology contexts ( Gefen et al., 2003; Pavlou, 2003), where trust often strengthens the translation of intention into action. In the case of BRISPOT, the lack of significant moderation may reflect high baseline trust in internal systems provided by a reputable institution like BRI, thereby reducing variance in the trust construct across users. Alternatively, the results may suggest that performance-driven and structural factors (such as job requirements or system mandates) overpower the effect of psychological confidence in the system. This underscores the context-specific nature of trust and its influence, suggesting that future studies could explore trust’s role more deeply through qualitative inquiry or by distinguishing between trust in system versus trust in data accuracy or organizational intention.

This study investigates the key determinants influencing the adoption of BRISPOT, a digital loan application platform developed by Bank Rakyat Indonesia (BRI), within the context of relationship managers (RMs) serving the SME segment. Drawing upon the Unified Theory of Acceptance and Use of Technology (UTAUT) and enriched by Social Cognitive Theory (SCT), the findings empirically validate that performance expectancy, effort expectancy, and facilitating conditions significantly affect behavioral intention to use the system. Furthermore, behavioral intention is found to have a strong and positive effect on actual use behavior, reinforcing the predictive strength of intention in digital adoption models. However, contrary to theoretical expectations, trust does not moderate the relationship between behavioral intention and use behavior, suggesting that its influence may be context-dependent or mediated by other latent variables in organizational settings.

The study offers theoretical contributions by extending UTAUT with a contextualized moderating variable (trust), while also integrating SCT to acknowledge the cognitive and environmental dynamics of internal system adoption. From a practical standpoint, the findings underscore the importance for banking institutions to invest not only in the functional and technical quality of digital platforms but also in the structural support and user training that shape adoption behavior. A user-centric approach that enhances perceived usefulness and ease of use, while ensuring operational readiness, is imperative for driving meaningful digital transformation in the banking sector.

Despite its contributions, this study has several limitations. First, the cross-sectional nature of the data limits the ability to capture changes in user perceptions or behaviors over time. Longitudinal data could provide richer insights into how behavioral intention and actual usage evolve, particularly in response to system updates or organizational policies. Second, the study focuses exclusively on relationship managers within the SME segment of a single state-owned bank, which may limit the generalizability of the findings across other employee roles, segments, or financial institutions. Third, the study’s reliance on self-reported measures of use behavior may be subject to social desirability bias or overestimation, which could affect the accuracy of reported adoption levels.

In addition, although trust was theorized to moderate the relationship between intention and behavior, the operationalization of trust may not have fully captured its multidimensional nature (e.g., trust in system, data, management, or external regulations), which might explain its non-significant effect. Future studies may consider a more granular approach in measuring trust or examining its mediating rather than moderating role.

Building on these limitations, future research could adopt a longitudinal research design to examine dynamic shifts in adoption behavior as users gain experience and as digital platforms evolve. Expanding the scope to include different banking roles (e.g., credit analysts, branch managers) or customer-facing staff in various regions would enhance the external validity of the findings. Moreover, future studies should explore additional constructs from the technology acceptance and behavioral sciences literature, such as digital literacy, organizational culture, leadership support, or resistance to change, to enrich the explanatory power of the model.

Furthermore, the role of trust in digital system adoption warrants deeper investigation. Qualitative or mixed-methods research could uncover the nuanced perceptions and experiences that shape trust-related decisions in organizational technology use. Finally, comparative studies across different banks or sectors (e.g., insurance, fintech, government services) may provide valuable insights into contextual variations in technology adoption drivers, contributing to the development of more robust and adaptive theoretical models.