In the past few years, multiple industries in the world, including in Indonesia, have been disrupted by digital transformation. Banking and Financial Services, Media and Telecommunication, Retail and Distribution, Transportation, and many other industries are amongst the ones being heavily disrupted. The digital disruption had causes incumbent business closed regardless how big or small their sizes are or pushed the incumbent businesses to continue transform for them to compete back with new disruptive players. In Indonesia, we have seen how Gojek has disrupted transportation and delivery business, Halodoc has disrupted healthcare industry, Traveloka has disrupted travel and hospitality industry. More and more large businesses have been and will continue to be disrupted. 

While banking has always adapted to cultural, regulatory, and technological shifts, the pace of change has accelerated dramatically within the past half-century. When changes happen quickly, the bigger picture is often overlooked. Each incremental innovation in the industry has served to enable greater flexibility and finer control. At the same time, this has decentralized core banking functions and made them more abstract. Take, for example, hoe money has evolved from a commodity basis to one established by fiat, and how institutions have thus been able to manage it better of profit from its flow. Similarly, we may think of what derivatives have done for the management of risk [1]. 

As the preferences and interests of tech-savvy customers keep on changing, we witness the rise of digital transformation in banking. According to reports, by 2006, 80% of all US banks were providing internet banking services – and the trend has shown no signs of slowing down. In 2009, Ally Bank was founded – the world’s very first all-digital bank. A study by Fiserve in 2010 demonstrated that both online and mobile banking were growing at a faster pace than even the internet [2].

Recent reports state that 80% of banking customers globally are regular users of mobile banking technology. But today, the very fabric of the financial industry is transforming as a result of DeFi and blockchain technology.       

Decentralized finance (DeFi) is an emerging model for organizing and enabling cryptocurrency-baseda transactions, exchanges and financial services. DeFi's core premise is that there is no centralized authority to dictate or control operations [3]. The blockchain can be considered a purely distributed data store with additional properties such as being immutable, append-only, ordered, time-stamped, and eventually consistent [4].

Technology is advancing so rapidly that, by 2026, many experts believe numerous western countries will be officially cashless [2]. In Indonesia, many digital banks have been born that offer various kinds of innovations and initiatives that have never existed before in Indonesia or even in the world. It offers faster and multiple service options and provides customers with omnichannel experiences. The omnichannel model of interaction with customers enables banks to simultaneously achieve several key goals of increasing their own business efficiency: increase sales while reducing their cost and improving the quality of customer service. The model can be used not only in banking, but also in other forms of retail business where it is possible to collect detailed statistics and build a factor analysis of conversion through a sales funnel [5]. In other words, digital transformation ensures swift and seamless banking services on any device.

Banks perform many things and their main role is to take in funds (called deposits) from those with money, pool them, and lend them to those who need funds. Banks are intermediaries between depositors (who lend money to the bank) and borrowers (to whom the bank lends money). The amount banks pay for deposits and the income they receive on their loans are both called interest [6]. This kind of bank is named conventional banking systems and generally known as a bank. But there is another type of bank that has very unique on providing their services based on Islamic Law “Sharia”. 

The characteristic of Islamic banking is operated by partnership and mutual benefits principle and it provides an alternative banking system with mutual benefits both for the public and the bank. This system will give priorities to aspects related to fairness in transaction and ethical investment by underlining the values of togetherness and partnership in production, and by avoiding any speculative activity in financial transaction. By providing various products and banking services supported by financial scheme under Islamic Law. Islamic banking will be a credible alternative that can be benefited by all of people without exception, especially the people of Indonesia[7]. In this project, we will research in scope of Sharia Banking environment.

Since its inception as a Sharia Business Unit as part of PT Bank Tepat Tbk in 2010, Bank Tepat Syariah has embraced and reached the productive underprivileged women, a segment the banking sector has not touched. It all started from a piloting project that focused on serving the customers from the productive underprivileged segment in three communities on Banten and Pandeglang area. The project continued to expand in 2011 to serve more customers in Sumatera, Banten, Jakarta, West Java, Central Java, East Java and East Nusa Tenggara (NTT). The company’s mandate is to deliver empowerment activities and financial literacy for underprivileged women in Indonesia. Bank Tepat Syariah provides access to banking products and services in Sharia principles to help the intention and realize the customer’s aspiration for a better life.

On 14 July 2014, Bank Tepat Syariah was officially registered as the 12^th Sharia Commercial Bank in Indonesia by the spin-off of PT Bank Tepat Tbk’s Sharia Business Unit and the conversion of PT Bank Teman Barudanharta. A conventional bank located in Central Java and had been established since 1991 which also served the productive underprivileged community with the latest ownership under the Triputra Group before merged in 2014. On 8 May 2018, Bank Tepat Syariah was officially a publicly listed company by the code BTPS. As the only Sharia Commercial Bank in Indoensia with the primary focus to develop financial inclusion by empowering the productive underprivileged customers, Bank Tepat Syariah has always strived to provide added value and make a difference in the life of every customer in addition to delivering and enhancing product and service delivery to positively impact millions of Indonesian as the actual manifestation of the aspiration to become the Mercy for the world.

Inspired by Grameen Bank which has been working all over Bangladesh since inception to reduce poverty through easy financial access to the rural poor, especially women, Bank Tepat Syariah engages in banking services that empower productive underprivileged communities, especially women. According to the Grameen Bank study, providing loans to women in the underprivileged segment is more effective. The main reason was that a woman was a “better fighter” against poverty than a man. According to Muhammad Yunus, the founder of Grameen Bank, a woman went to more extraordinary lengths to improve her children’s health and nutrition and to further educate her children. Simply put, women are more effective in managing the loan she gets. The field of development has come to a similar conclusion in the past few years, with many aid workers affirming that the best way to fight poverty is to strengthen the positions of women and girls [8]. 

To Engage in this endeavor, Bank Tepat Syariah is led by five directors appointed by the Annual General Meeting (AGM), A mandatory annual assembly of a company's executives, directors, and interested shareholders, on 21 April 2023. It Consists of four directorates:

• Financing Business

• Funding and New Business

• Compliance and Risk

• Finance, Treasury and Operations, as well as Business 

Figure 1: Excess Salary Payments Profile (As of Q2 2022)

Figure 2: Excess Salary Payments Profile (As of 2022)

Development, Internal Audit, Information Technology, and Human Capital. 95% of Bank Tepat Syariah employees are women, of which 47% are high school graduates. Bank Tepat Syariah servers millions of underprivileged women in 23 provinces, from Aceh to East Nusa Tenggara

In Bank Tepat Syariah, the calculation of payroll payment for the current month is processed on the 10^th of the month (if 10^th is a holiday the calculation will be processed in the working day before or after 10^th). Therefore, all information related to termination must be received before the calculation date, so that the salary for the terminated employee is correct. Normally a termination application must be submitted one month before resigning (one-month notice). For absenteeism, a salary withholding notice must be sent before the calculation day so that the final salary is not paid before the termination process is complete. 

The termination rate at Bank Tepat Syariah is consistently high, with over 4400 employees terminated every year (Figure 2). This data indicates a high potential risk of delays in the termination process if done manually. Such delays can significantly burden operational costs and may result in operational losses for the company. Based on historical data (Figure 1), the potential operational losses due to termination reporting delays amounted to 641.65 million Indonesian rupiah in 2015, gradually decreasing to half of that amount in 2016 and 2017. However, this figure remained a significant burden for the company. Therefore, by the end of 2018, the pilot implementation of online termination processes was initiated, resulting in a substantial reduction in operational losses. The termination delay-related losses dropped to only 235.84 million Indonesian rupiah in 2018 and sharply declined to just 148.32 million Indonesian rupiah in 2019.

Figure 3: Terminated Employees Profile (As of Q2 2022)

Figure 4: Transaction Rejection (June –August 2023)

The significant decrease in operational losses caused by termination reporting delays in 2019 was due to the full implementation of online terminations. By utilizing online termination, both resignations and employee absenteeism requests can be processed faster and more efficiently, allowing termination data to be submitted promptly. The potential delays caused by manual submission processes are significantly reduced. Bank Tepat Syariah conducted a large-scale campaign among its employees to encourage the use of the application. Through a training-to-trainer mechanism, extensive training was provided to leaders from each team, enabling them to deliver further training to their respective teams.

In 2019, there were 4434 termination requests (Figure 3), and the excess salary payments were also at a historically low figure for Bank Tepat Syariah, totaling 148.32 million Indonesian rupiah. When compared to 2015, where there were 5673 requests and the potential operational losses due to termination delays amounted to 641.65 million Indonesian rupiah (4.3 times higher than in 2019), this demonstrates the effectiveness of using online terminations when implemented properly.

In 2020, the improvement continued with a decrease in the number of delayed transactions, despite a nominal increase. However, the total number of requests decreased compared to 2019. However, starting from 2021, there was an increase in the number of delayed termination transactions submitted, rising from only 198 requests in 2020 to 362 in 2021 and further increasing to 430 in 2022. This increase was also accompanied by a rise in the potential operational losses, starting from 204.39 million Indonesian rupiah in 2020 to 361.86 million rupiah in 2021, and further escalating to 428.34 million Indonesian rupiah in 2022. The potential operational losses due to these delayed termination submissions are even higher than the losses incurred before the introduction of the online termination application.

Table 1: Construct and the Resources

This could be possible because 2019 was the year when we conducted a massive campaign promoting the use of online terminations. In 2020, employees still remembered the training they received in 2019. However, starting from 2021, there were numerous organizational changes, employee promotions, rotations, and a lack of continuity in the training-to-trainer mechanism from the previous supervisors to the new ones. This resulted in many employees being unaware of the basic procedures for using the online termination application.

This is supported by two basic interviews with two supervisors (from MMS SIMPANG EMPAT ASAHAN and MMS LURAGUNG KUNINGAN), whose subordinates faced difficulties in using the online termination system. The requests submitted by their Community Officers were rejected by Human Capital up to four times, and these supervisors failed to validate the requests properly, despite clear notes from Human Capital indicating the reasons for previous rejections. Both of them and their Community Officers failed to understand the clear instructions provided in the application and did not comprehend the rejection notes written by Human Capital. In line with this, from the rejection data by Human Capital between June and August 2023, it was found that there were 250 termination rejections due to human error (Figure 4). The most common mistakes were discrepancies between the resignation letter dates and the dates inputted by employees into the online termination system.

Employees' difficulties in understanding the application usage could be due to their lack of readiness or the user interface of the application not being user-friendly. Every employee facing challenges in using the application will be guided by the project team to ensure correct usage, preventing similar issues from occurring in the future. Digital transformation is less about technology, but people [9]. A great application is of no use and benefit if it cannot be used properly. What needs to be considered is whether the current user interface is easy to comprehend and understand for the employees. This ensures that it can be used effortlessly, almost on autopilot, without the need for direct guidance from someone who understands it better.

This is related with what it is happening of Community Officer. From the above conditions, management needs to first evaluate the readiness of its Community Officer in accepting digitalization or there are things that need to be adjusted in the application so that it makes it easier for Community Officers to understand and use the application without having to have guidance or direction from their superiors.

Drawing insights from the literature review, the author discovers that the Technology Acceptance Model (TAM) stands out as the predominant theory for assessing acceptance of termination online systems. Consequently, the author opts to incorporate TAM as the central construct in the conceptual framework for this research. Additionally, the author identifies certain external factors that have been demonstrated to influence TAM constructs, ultimately impacting employees' acceptance of termination online systems. These external factors are selected based on their widespread recognition among previous researchers, establishing their reliability through various case studies. The author also integrates external factors that have been proven to significantly affect employees' acceptance of termination online systems, tailoring the selection to align with the target population in this study. All these elements are amalgamated to formulate the conceptual framework. 

Subsequently, the research proceeds to examine the assumed relationships among each element in the conceptual framework through a quantitative methodology using survey. The objective is to pinpoint the factors that exert a substantial influence on employees' acceptance of the termination online system and evaluate Bank Tepat Syariah's present performance on each of these factors. As the construct represents a latent variable that cannot be directly observed or measured, a series of indicators or items must be developed to deduce the value of each construct. The chosen items to gauge all constructs are borrowed from prior research, with modifications made to accommodate the specific conditions of Bank Tepat Syariah, ensuring the validity of the measurements.

To analyze the quantitative data, the author employs the Partial Least Squares-Structural Equation Modeling (PLS-SEM) approach using SmartPLS: Ringle, C. M., Wende, S., and Becker, J.-M. 2015. "SmartPLS 3." Boenningstedt: SmartPLS GmbH, http://www.smartpls.com. There are 41 items to measure the 10 constructs with the detail shown in the Table 1. The full description of each item is available in Appendix A.

The target respondents of this research are the active employee of Community Officer to Senior Business Manager (or other experienced employee as Community Officer). Author distributed the questionnaire from using Whatsapp or Microsoft Teams. There are 12.011 Business Managers and Senior Business Manager per 21 November 2023. So, the calculation to get the number of sample using Slovin.

Description:

N:            Samples

N:            Total population

E:            Error tolerated level (5%)

Based on the calculation above the minimum sample is 387 but the author set sample of this research will be 450 employees. And the data will be analyzed once targeted sample of 450 has been acquired. 

The Survey structure consists of three sections. The First section asks about the participants’ data (name, gender, age, education, job, year of service, and region). The second section will measure each indicator of the model’s construct using a five-point Likert scale where one represents “strong disagree” and five represents “strong agree”. The third section pertains to questions regarding experiences or challenges faced in using the application and is related to suggestions.

PLS-SEM is selected due to its capability to analyze relationships among multiple constructs concurrently within a complex model, involving numerous constructs and indicators. Additionally, it is well-suited for situations with a small sample size requirement and non-normally distributed data [10]. Within PLS-SEM, there exist two primary components: the measurement model and the structural model. The measurement model, also known as the "outer model," assesses the connections between latent constructs (such as variables or factors) and their corresponding indicators (items or measures). The latter refers to the “inner model”, which measures the constructs’ relationship [11]. The analysis result will tell the author which variable significantly influences Bank Tepat Syariah employee acceptance of termination online system and the relation between each factor.

In this study, it is necessary to establish the characteristics of respondents as a depiction of the respondent profile, which serves as the source of primary data.

Table 2: Characteristic Profile

Based on the data in Table 2, it can be observed that out of 450 respondents, the majority are female, totaling 439 respondents (97.60%), while the remaining 11 respondents (2.40%) are male. It can be observed that the majority of respondents are aged 26 – 35 years, totaling 259 respondents (57.60%), followed by respondents aged under or equal 25 years, amounting to 171 respondents (38.00%) and respondents aged 36 - 45 years are 20 (4.40%). It can be observed that the majority of respondents are Undergraduate Employees, totaling 212 respondents (47.10%), followed by High School Graduate Employee, amounting to 189 respondents (42.00%) and the rest are from Diploma and Graduate. It can be observed that the majority of respondents are Community Officer (CO), totaling 318 respondents (70.70%), followed by Business Manager (BM), amounting to 101 respondents (22.40%) and the rest are from Senior Business Manager and Non MMS. Business Manager (BM), Senior Business Manager and Non MMS have experience as a Community Officer. Based on the data in It can be observed that the majority of respondents have less than or equal 5 years of service, totaling 232 respondents (51.60%), followed by 6 to 10 years of service, amounting to 145 respondents (32.20%) and the rest are 73 respondents (16,2%). It can be observed that the majority of respondents from Java, totaling 298 respondents (66.20%), followed by Sumatera, amounting to 112 respondents (24,80%) and the rest are 40 respondents (8,9%) from other islands.

This study employs a variance-based or component-based approach using the Partial Least Squares (PLS) method. The examination of the results of structural equation modeling with the PLS approach is conducted by considering the outcomes of the measurement model (outer model) and the structural model (inner model) of the researched model. The following is the model obtained through the Partial Least Squares (PLS) method:

Testing the outer model can be done by examining convergent validity, discriminant validity, and composite reliability. The measurement model elucidates the connection between each survey item and the constructs it is designed to measure. This phase is conducted to evaluate the validity and reliability of each item in accurately depicting its respective variable. Validity pertains to the extent to which an instrument effectively measures what it is intended to measure, while reliability concerns the consistency of the measure employed in the study [12]. There are two main techniques to assess the measurement model: convergent validity and discriminant validity [13].

Figure 5: Transaction Rejection (June –August 2023)

Table 3: Convergent Validity Results

Table 4: Discriminant Validity

Figure 6: Boostraping Path Diagram Result

Table 6: R-Square Result

Table 7: Hypothesis Testing Result

Convergent validity is employed to identify instrument items that can be used as indicators of the overall latent variable. The test results are measured based on the magnitude of the loading factor (outer loading) of the construct indicators and the value of Average Variance Extracted (AVE). Convergent validity is considered met if the Factor Loading of each indicator is > 0.7, and the AVE of each is > 0.5. The results of the convergent validity test are presented in the following Table: 3

Based on the table above, it is evident that all indicators have factor loading values > 0.7, and the AVE values for each construct variable are > 0.5. This indicates that all indicators in this study satisfy convergent validity and can proceed to the next testing phase.

After testing convergent validity, the next step in outer model testing that needs to be conducted is discriminant validity. The results of the discriminant validity test are presented in the following Table: 4

From the table above, it can be concluded that all dimensions or constructs are distinct from each other. The diagonal indicates the square root of the AVE values for each dimension or construct. Based on the values above, higher correlations are obtained within constructs and dimensions compared to correlations with other constructs and dimensions. This indicates that all dimensions and constructs can be considered valid in forming the research model.

To ascertain the reliability of each research construct, testing is conducted by examining the values of Composite Reliability and Cronbach's Alpha for each construct. To achieve good reliability, the composite reliability value should be greater than 0.7, and the Cronbach's alpha value should be greater than 0.6. The results of composite reliability are presented in the following Table: 5

Based on the table above, it can be explained that the results of the composite reliability testing indicate good outcomes because all latent variables are reliable, having composite reliability values greater than 0.7. This suggests that all indicators serve as reliable measures for their respective constructs. Additionally, from the table, it can be observed that all latent variables have Cronbach's alpha values above 0.6.

The Inner Model is a test on the structural model conducted to examine the relationships between latent constructs. In this study, the inner model testing is performed by displaying the R2 values for endogenous latent constructs. Subsequently, the structural model in the inner model is tested using the predictive relevance (Q2) values. The following is the result of the bootstrapping path diagram in the inner model using SmartPLS 3.0:

The R2 values indicate that the higher the R2 value, the better the predictive model of the proposed research model. The following are the results of the R2 values for endogenous variables (variables that are influenced) Table 6:

Based on the table above, the R-Square value for the Perceived Enjoyment variable is 0.258, meaning that Perceived Enjoyment is influenced by Subjective Norms by 25.8%. Then, for the Perceived Ease of Use variable, an R-Square of 0.705 is obtained, indicating that Perceived Ease of Use is influenced by UIUX, Information, Communication, Computer Self-Efficacy, Subjective Norm, and Perceived Enjoyment by 70.5%. The R-Square value for the Perceived Usefulness variable is 0.885, indicating that the Perceived Usefulness variable is influenced by UIUX, Information, Communication, Computer Self-Efficacy, Subjective Norm, and Perceived Ease of Use by 88.5%. Moving on to the Attitude Toward Using variable, an R-Square value of 0.644 is obtained, indicating that the Attitude Toward Using variable is influenced by Perceived Enjoyment, Perceived Ease of Use, and Perceived Usefulness by 64.4%. Lastly, the Behavioral Intention to Use variable has an R-Square value of 0.572, meaning that the Behavioral Intention to Use variable is influenced by Perceived Usefulness and Attitude Toward Using by 57.2%.

A model is considered to have relevant predictive value if the Q-square value is greater than 0 (>0). The predictive relevance value is obtained using the formula:

Q² = 1 – (1 – R₁²) (1 – R₂²) (1 – R_n²)

Q² = 1 – (1 – 0.258) (1 – 0.705) (1 – 0.885) (1 – 0.644) (1 – 0.572)

Q² = 0.996

The Q-Square calculation in this study yields a value of 0.996. Therefore, it can be concluded that the model in this study has a highly relevant predictive value.

The values of path coefficients, t-statistics, and p-values for the constructs in this study are presented in the following Table 7:

Hypothesis testing in this study employs a 5% level of significance. Therefore, the critical value that must be satisfied is 1.96. The positive or negative influence between exogenous latent constructs and endogenous latent constructs is examined based on the path coefficient values, while the significance of the influence is assessed by the t-statistics or p-value. Based on the table above, conclusions can be drawn regarding the hypotheses, including:

• User Interface & User Experience (UIUX) has a positive and significant effect on perceived usefulness (PU)

• User Interface & User Experience (UIUX) has a positive and significant effect on Perceived Ease of Use (PEOU)

• Information (INF) has a positive and significant effect on perceived usefulness (PU)

• Information (INF) has a positive and significant effect on Perceived Ease of Use (PEOU)

• Communication (COM) has a positive and significant effect on perceived usefulness (PU)

• Communication (COM) has a positive and significant effect on Perceived Ease of Use (PEOU)

• Computer Self-Efficacy (CSE) has a positive and significant effect on perceived usefulness (PU)

• Computer Self-Efficacy (CSE) has a positive and significant effect on Perceived Ease of Use (PEOU)

• Subjective Norm (SN) has a positive and significant effect on perceived usefulness (PU)

• Subjective Norm (SN) has a positive and significant effect on Perceived Ease of Use (PEOU)

• Subjective Norm (SN) has a positive and significant effect on Perceived Enjoyment (PE)

• Perceived Enjoyment (PE) has a positive and significant effect on Perceived Ease of Use (PEOU)

• Perceived Enjoyment (PE) has a positive and significant effect on Attitude Toward Using (ATT)

• Perceived Ease of Use (PEOU) has a positive and significant effect on perceived usefulness (PU)

• Perceived Ease of Use (PEOU) has a positive and significant effect on Attitude Toward Using (ATT)

• Perceived Usefulness (PU) has a positive and significant effect on Behavioral Intention to Use (BI)

• Perceived Usefulness (PU) has a positive and significant effect on Attitude Toward Using (ATT)

• Attitude Toward Using (ATT) has a positive and significant effect on Behavioral Intention to Use (BI)

From the results of the quantitative analysis, the Behavioral Intention to Use variable has an R-Square value of 0.572, indicating that 57.2% of the Behavioral Intention to Use variable is influenced by Perceived Usefulness and Attitude Toward Using. This suggests a relatively good level of employee acceptance of the online termination application, although not exceptionally high. This value also indicates a reasonably good level of readiness for digital transformation at Bank Tepat Syariah.

However, contrary to this, based on the initial data, there has been an increase in delays in the submission of termination requests since the initial implementation of the application. As this study is the first research on the readiness for using online termination at Bank Tepat Syariah, it cannot be determined whether there has been a decrease in the Behavioral Intention to Use variable compared to the previous year.

This study also reveals the existence of other constructs not included in the conceptual framework that influence employees' desire and acceptance of using the online termination application. These unknown constructs affect the Behavioral Intention to Use variable by 42.80%. These unidentified constructs contribute to the decline in the effectiveness and readiness of using online termination, and if researched and studied, they could significantly impact the increase in employee acceptance of the online termination application and, in general, enhance employee readiness for digital transformation.

The other unknown constructs could be related to employee workload, especially at the end of the month when they need to meet specified targets, causing them to focus on performance achievement rather than the neatness of their administrative work. Currently, the emphasis has only been on rewarding employees who achieve success, but there are no penalties for employees who do not carry out the termination process neatly as stipulated. Therefore, in future research, it is hoped that the unknown constructs identified in this study will be included.

A sequence of actionable implementation plans is made for Bank Tepat Syariah within six months to help them improve the termination online system into better termination online system based on the survey. Improvements and adjustments need to be made according to the feedback received in the survey, both in terms of appearance, menu presentation, and the existing workflow, including adding some new features needed by employees. Because UI/UX has a higher priority for immediate improvement, Bank Tepat Syariah should prioritize UI/UX changes as the first step. These adjustments also need to be balanced with improvements in the delivery of information related to online termination and adjustments in the communication mechanism between employees and the central office.

In the first month, Bank Tepat Syariah, should focus on collecting the requirements from all stakeholders and from the employees. In this stage, Bank Tepat Syariah needs to collect a list of requirements as much as possible, which will then be prioritized based on risk levels, importance, and needs. 

In the second month, the focus is on finding solutions to technical system and network challenges. In the preparation of this, the IT team of Bank Tepat Syariah must collaborate with the online termination vendor. The results of the requirements compilation in the first and second months are the development of the agreed-upon Business Requirement Documents (BRD) by all relevant parties.

In the third month, the Functional Specification Document (FSD) is started to be prepared as a plan to build the system according to the list of changes that will be made in online termination. In this preparation, blueprints and mockups for the new application to be developed by the vendor are also created.

From the fourth month to the middle of the fifth month, the vendor focuses on improving online termination in the development environment. The middle of the fifth month until the end of the sixth month involves System Integration Testing by IT and User Acceptance Test by Human Capital (HC).

System Integration Testing (SIT) is a type of testing conducted to ensure that all system components in the application are well integrated and function as expected. SIT involves the interaction between various system components to ensure they are integrated correctly and can exchange information as needed. The function of SIT is to verify whether the entire system functions correctly, whether all components work as expected, and more.

User Acceptance Testing (UAT) is testing aimed at ensuring that the application can meet the components in the business document and is acceptable to users. This testing needs to gather feedback from users to determine if the application is easy to use and meets the business needs of the company or not. UAT involves reviewing each function in the application to ensure it aligns with user requests and complies with existing business documents. UAT is usually the final phase of the software testing process and is conducted after the software is fully developed and internally tested (e.g., SIT, functional testing). The goal of UAT is to ensure that the software is ready for use and can meet the needs of the business and end-users.

In the sixth month, the implementation and data migration of the new online termination system are carried out with limited roll-out starting in seventh month, followed by mass application usage in the same month.

In the preceding section, we delved into the analysis derived from the author's conducted survey, presented the proposed business solution for Bank Tepat Syariah, and outlined the implementation plan for the proposed solution. Chapter V aims to draw a comprehensive conclusion based on the analysis and interpretation outcomes, addressing the research question posed in Chapter I. Additionally, this chapter will provide recommendations, summarizing suggested actions for relevant stakeholders as outlined in the proposed implementations, and offer suggestions for future research to conclude this section.

This subchapter concludes the conclusion of the whole research starting from chapter 1 which mainly discusses the business background and issues until the research result to be analyzed in chapter IV. The writer will discuss how the research questions and objectives in chapter I have been answered and fulfilled. 

According to the findings of this research, it can be concluded that, overall, employees, particularly Community Officers, are capable of adapting to the use of an application, in this case, the online termination system. However, in practice, Community Officers need clearer information and communication mechanisms when facing challenges or seeking explanations for aspects they do not understand. 

Considering the employee profile, where nearly 40% are located outside Java, and even those in Java often come from small towns far from major city centers, attention should be given to the language used, which is mostly tailored to high school graduates. Any information related to the application must be clear and relevant to them. Further study is needed to determine the most effective way to convey information to them. 

Communication from the central office to community officers should be explicit, and the help portal should be clearly communicated to employees. This is because employees feel they struggle to find assistance when encountering difficulties. Despite the availability of the help portal, survey results indicate that the portal has not been adequately and uniformly publicized. 

The current challenges related to information or communication would significantly decrease if the User Interface and User Experience (UIUX) of the application were more user-friendly, facilitating the employees in using and utilizing the application correctly, specifically in the context of the online termination system. With improved UIUX, it is expected that employees, especially community officers, would require fewer guidelines and can readily comprehend the application they are using.

In conclusion, employees' readiness and adaptability to digitalization are influenced by the UIUX of the application they use, clear information during orientation (including user guides in relevant language), and the availability of communication channels that facilitate the expression of encountered challenges. Additionally, individual factors also significantly influence employees' adaptability in using applications.

From the survey, data revealed that the top three most frequently reported related to UIUX challenges are perceived application slowness, network issues, and difficulty understanding the instructions/notifications within the application.

The application slowdown occurs on specific dates when many employees access the application, especially on payday when numerous employees access it to retrieve their pay-slips. It's important to note that the online termination system is one module within an ecosystem of employee self-service on a shared platform. Consequently, if one module experiences high traffic, it draws resources from other modules. Therefore, technical adjustments are necessary to ensure that the termination online module has a separate segment from others. This way, each module can utilize resources specifically allocated to it, without affecting the resources of other modules.

As previously mentioned, community officers come from small towns, and they operate in remote areas where internet coverage is not as robust as in larger cities. Hence, an offline mode needs to be introduced. The offline mode will activate when the network is lost or weakened, allowing employees to continue transactions. Once the network is restored, the application will automatically synchronize.

Clarity in labeling, instructions, and notifications within the application is crucial. Currently, the online termination system uses English, causing some employees to struggle to understand the application's purpose. Therefore, a language feature needs to be implemented to change the language according to the user's preference. Additionally, further research is needed on the naming of labels, instruction templates, and notifications within the online termination system to ensure better understanding and usability.

To enhance the digital transformation readiness of employees, especially community officers at Bank Tepat Syariah, regular training and learning sessions should be conducted. Socialization and regular refreshers related to the applications they use are crucial and should be incorporated into the mandatory curriculum for employees each year. Recognizing and rewarding employees who actively contribute to digital transformation is essential, and considering consequences for employees unwilling to adapt to digitization should also be taken into account.

In the preceding section, the author has put forward an integrated implementation derived from the proposed business solution. The implementation plan delineates specific courses of action, serving as recommendations for the business stakeholders at Bank Tepat Syariah. Reflecting on the research journey, the author offers several suggestions for future researchers. 

Although the UIUX survey results are relatively high at 4.16 (out of 5), qualitative feedback from employees reveals challenges and difficulties in using the application.

Therefore, it is advisable to reconsider conducting qualitative research to gain a deeper understanding of UIUX that best suits the employees. Consider conducting periodic research on termination online to track changes in Behavioral Intention to Use, enabling Bank Tepat Syariah to take prompt corrective actions. As mentioned in the previous chapter, there may be undetected constructs in this study. Therefore, in future research, explore these constructs to enhance the understanding of digital transformation readiness. Consider increasing the number of target respondents in future research to obtain more accurate data. Termination online is one module of Employee Self Service, and similar research should be conducted on other modules.

1. Microsoft Corporation. The Future Banking Ecosystem: Evolution and Innovation in the Digital Era. 2017, info.microsoft.com/rs/157-GQE-382/images/Future-banking-ecosystem.pdf.

2. “Technology in Fintech and the Story of Online Banking.” FinTech, 2022, fintechmagazine.com/banking/fintech-timelines-and-the-story-of-online-banking.

3. Kerner, Sean M. “Decentralized Finance (DeFi).” TechTarget, 2022, techtarget.com/whatis/defi nition/decentralized-finance-DeFi.

4. Daniel, D. Blockchain Basics: A Non-Technical Introduction in 25 Steps. Apress, 2017.

5. Vasiliev, S. A. and E. R. Serov. “Omnichannel Banking Economy.” MDPI, 2019.

6. Gobat, Jeanne. “Banks: At the Heart of the Matter.” International Monetary Fund (IMF), imf.org/en/Publicatio ns/fandd/issues/Series/Back-to-Basics/Banks.

7. Otoritas Jasa Keuangan (OJK). “Sharia Banking.” ojk.go.id/en/kanal/perbankan/Pages/Bank-Syariah.aspx.

8. Kristof, Nicholas and Sheryl WuDunn. “The Women’s Crusade.” The New York Times, 23 Aug. 2009, nytimes.com/2009/08/23/magazine/23Women-t.html.

9. Frankiewicz et al. “Digital Transformation Is About Talent, Not Technology.” Harvard Business Review, 2020, hbr.org/2020/05/digital-transformation-is-about-talent-not-technology.

10. Hair, J.F. et al. “Partial Least Squares Structural Equation Modeling (PLS-SEM).” European Business Review, vol. 26, no. 2, 2014, pp. 106–121.

11. Hair, J. et al. “An Updated and Expanded Assessment of PLS-SEM in Information Systems Research.” Industrial Management and Data Systems, 2017, pp. 442–458.

12. Sekaran, Uma, and Roger Bougie. Research Methods for Business: A Skill-Building Approach. Wiley, 2010.

13. Al-Emran et al. “PLS-SEM in Information Systems Research: A Comprehensive Methodological Reference.” Advances in Intelligent Systems and Computing, vol. 845, 2019, pp. 644–653.