Navigating Customer Satisfaction: Unravelling The Dynamics Of Chatbot Performance, Trust in Technology

Abstract

This study explores the dynamics of customer satisfaction in chatbot interactions, with a focus on how chatbot  performance influences users’ trust in technology. Conducted in Bangalore City, Karnataka, the research targets  customers of both private and public sector banks. A multi-stage random sampling method was used, beginning  with all banks in the city and narrowing to four major institutions: HDFC Bank, ICICI Bank, State Bank of India,  and Bank of Baroda. Data were collected from 350 customers across selected branches of these banks. The study  applies advanced statistical techniques, including Structural Equation Modelling (SEM) and moderation analysis,  to examine the relationships among key variables. The findings highlight the critical role of chatbot performance  in shaping customer satisfaction, with technological trust acting as a mediator and technical literacy serving as a  moderator in influencing overall customer experience in digital banking. 

Keywords: Customer satisfaction, Chatbot interactions, Trust in technology, Chatbot performance 

1.Introduction 

Since the "customer service revolution" about 20 years ago, business research has prioritised  customer happiness and customer-focused organisations. Business consultants and enterprises  have developed techniques to identify customer-pleasing organisations, assess happiness, and  construct quality improvement processes based on feedback. While corporate research has  focused on customer service and happiness, it is not exclusive to the private sector. Publicly  sponsored organisations embracing commercial techniques provide valuable insights and  research. Federal, state, and municipal governments are increasingly measuring their  performance and impact on individuals they serve. Initiatives to "reinvent" government, such  as school reform, privatisation, and managed care, prioritise customer service and satisfaction.  The European Union is reforming social services with a focus on social inclusion and "user  involvement" to enhance quality[1].

Artificial intelligence is not a new notion in India. For decades, research groups and institutions  have been experimenting with various AI technologies, notably in the subject of social change.  AI is becoming more mainstream, with large corporations and start-ups investigating numerous  opportunities as supporting technology becomes more available and inexpensive.technology  significantly impacts industries and everyday tasks. It has several applications and is used  differently across the globe. Artificial intelligence has recently gained popularity among the  public. Artificial intelligence mimics the cognitive abilities of humans. To grow more accurate  and human-like, AI chatbots are taking the place of human responders. The use of chatbots,  sometimes known as virtual assistants, is growing in popularity as a result of developments in  natural language processing, artificial intelligence, machine learning, and neural networks.  Interactive questions let chatbots communicate with people more effectively. To expand the  chatbot industry, several cloud-based chatbot services have surfaced, such as Watson from  IBM, Cleverbot, and ELIZA chatbot. Conversation between people and robots has greatly  improved in recent years, with more responsive conversational agents. This article discusses  the generalisation and description of AI chatbots[2]. 

Chatbots are adaptable and show a lot of potential. Human care providers often struggle to  manage resources and get exhausted. Chatbots are available 24/7 due to scalable technology,  ensuring quick access for consumers. Chatbots may dramatically decrease expenses and  automate corporate procedures. Chatbots, such as Amazon Alexa and Apple Siri, are becoming  more popular for their practical applications. Watson, IBM's question-and-answer programme,  is a great example. Chatbots have a wide range of applications outside information technology.  In 2020, chatbots accounted for over 85% of consumer interactions. Here's an example:  "Businesses can use chatbots for internal communication in addition to using them for external  communication (with customers)" . Large organisations often utilise chatbots for staff  assistance, recruitment as well as training. Meet Franks is a digital assistant that links talent  and corporations discreetly. Live chat interfaces are becoming more popular for providing real time customer care, particularly in e-commerce. Chatbots are often used to replace human chat  service representatives. Although AI-based chatbots have been extensively adopted for cost and time savings, they often fail to meet consumer expectations, potentially leading to decreased  user engagement[3]. 

Chatbots are thought to provide round-the-clock assistance, seven days a week, which  improves customer interactions by cutting down on bank customer wait times. Because they  make common banking tasks simpler for consumers to do, chatbots are seen as beneficial. 

Additionally, using chatbots for various financial procedures is easier and costs nothing. It's  argued that chatbots pose less of a risk to privacy and data security. The chatbots are a  trustworthy source since their responses are up to date and precise[4]. 

Chatbot apps may serve as a useful tool for personal consultation and learning via interactive  tactics and user-friendly interfaces, apart from providing a range of successful interpersonal  interactions. A chat bot's ability to reach a large audience via automated, personalised messages  and a messaging system is its main benefit. Furthermore, chatbots may be used in the workplace  due to the growing use of mobile technology, the lack of time and location constraints, and the  development of interactive education techniques[5]. 

This paper is organised as follows: the third portion presents our analysis's methodologies and  data collecting, while the second section evaluates pertinent literature. The findings are  explained and our results are described in the fourth part. The article is concluded in the last  part. 

2.Literature Review  

[6] This study investigates how cognitive and peripheral inputs affect experience  characteristics, resulting in chatbots, building on the principles of the Model of Elaboration  Likelihood (ELM). The user wants to suggest. In the buy and post-purchase phases, the study  used a strong interpretive sequencing mixed-method research design and incorporated  competence as well as warmth as mediating variables.[7] This paper provides an in-depth  analysis of the manner how artificially intelligent (AI) and NLP, or natural language processing,  are pushing the boundaries of convention, revolutionising modern marketing strategies, and  providing insightful knowledge about consumer behaviour and tactical decision-making. 

[8] This study aims to provide a novel contribution by offering a thorough, well-organized  roadmap that describes the fundamental ideas required for successful combining automation  and process optimisation within the framework of state-of-the-art artificial intelligence[9] This  study explores the intricate interplay between social media and marketing tactics in influencing  the global economic landscape. Understanding the critical roles that social media and  marketing play becomes imperative for companies, governments, and other stakeholders in an  age when globalisation & digitalization are driving revolutionary changes. scholars as well. 

[10] A new age of improved digital offering dependability, optimized supply chain operations,  and instantaneous access to priceless data and analytics has been brought about by the 

incorporation of AI. In its endeavour to close the knowledge gap and enable the effective  integration of AI into business planning, this article aspires to perfection.[11] The report ends  with recommendations for further research on personalised marketing, with an emphasis on  cutting-edge technologies including wearables, big data, blockchain, artificial intelligence, and  the internet of things. investigate new approaches to creating individualised experiences for  both online and offline platforms. 

[12] The study puts the rise of smart tourism in perspective and explains how technology and  data are changing the dynamics of the sector. In summary, it considers potential uses while  highlighting the need of a well-balanced combination of data and technology to fulfil the  potential of smart tourism.[13] This article offers a thorough analysis of scientific research on  AI's effects on digital marketing initiatives. By looking at several scholarly articles, this study  aims to provide light on how artificial intelligence (AI) tools are affecting strategies, processes,  and outcomes in the area of digital marketing. 

[14] This study used a serial mediation mechanism to examine the effect of human portrayal in  chatbot on the intention to follow health advice, using psychological separation and trust  towards the chatbot counsellor. The results showed that participants in the simulated human  condition expressed a stronger willingness to heed chatbot-generated emotional wellness  advice than did those in the mechanical representation condition. [15] Because of digital  technology and media, the situations in which companies interact with their customers are  changing. The authors provide a number of plausible extensions of the present study's findings  as ideas for future academic research. 

3.Research Gap 

The study gap stems from a lack of investigation into how confidence in technology mediates  the link between chatbot performance and customer pleasure, as well as a limited grasp of how  technical literacy moderates this relationship. Current research focuses mostly on direct  impacts, ignoring these critical mediating and moderating variables. Addressing these gaps is  critical for gaining a more complete knowledge of customer-bot interactions and may give  significant insights into the design and deployment of successful chatbot systems. 

4.Aim of the Study 

This study's goal is to look into and comprehend the elements that influence customer  satisfaction in chatbot encounters, with a particular emphasis on chatbot performance and its 

effect on users' faith in technology. The study aims to give insights that may help increase  chatbot performance, overall customer happiness, and a better understanding of technology's  developing role in moulding user trust. 

5.Objective 

OB1: To evaluate and quantify the direct influence of chatbot performance on customer  satisfaction. 

OB2: To examine the mediating effect of trust in technology in the relationship between chatbot  performance and customer satisfaction. 

OB3: To explore the moderating role of technical literacy in shaping the relationship between  chatbot performance and customer satisfaction. 

6.Hypothesis 

H1: Chatbot performance has a significant direct positive effect on customer satisfaction. 

H2: Trust in technology significantly mediates the relationship between chatbot performance  and customer satisfaction. 

H3: Technical literacy significantly moderates the relationship between chatbot performance  and customer satisfaction. 

7.Research Methodology

                         Fig.1 conceptual frame work 

8.Research Design 

This study explores Navigating Customer Satisfaction: Unravelling The Dynamics Of Chatbot  Performance, Trust In Technology, using both surveys and qualitative methods. Surveys gather  numerical data on satisfaction and preferences, while interviews and focus groups delve into  reasons behind perceptions, addressing aspects like trust in AI. This mixed-method design aims 

to offer a comprehensive understanding of factors influencing Navigating Customer  Satisfaction - Unravelling The Dynamics Of Chatbot Performance, Trust In Technology. 

9.Data Collection Methods 

A structured survey will collect quantitative data on Navigating Customer Satisfaction:  Unravelling The Dynamics Of Chatbot Performance, Trust In Technology. Stratified sampling  ensures a diverse sample, using electronic and in-person distribution. Data analysis includes  descriptive and inferential statistics. Validity and reliability are ensured through pre-testing,  Cronbach's alpha, and ethical considerations. This approach aims to yield reliable insights into  Dynamics Of Chatbot Performance, Trust In Technology customer views on Chatbot  Performance. 

• Sampling 

The quantitative study on " Navigating Customer Satisfaction: Unravelling The Dynamics Of  Chatbot Performance, Trust In Technology" will employ a meticulous sampling strategy. The  target population, bank customers Satisfaction Chatbot Performance, Trust In Technology, will  be stratified based on characteristics such as age, income, and banking frequency. Random  sampling within each stratum will ensure diverse perspectives are represented. The sample size  will be determined through statistical calculations balancing reliability and feasibility. The goal  is to generalize findings to the broader population while maintaining efficiency. Efforts to  minimize selection bias and enhance external validity will be ongoing, providing a  comprehensive snapshot of Navigating Customer Satisfaction: Unravelling The Dynamics Of  Chatbot Performance, Trust In Technology. 

• Sample Size  

The research "Navigating Customer Satisfaction: Unravelling the Dynamics of Chatbot  Performance and Trust in Technology" carefully created a sample of 350 participants, expertly  balancing dependability and manageability. This enlarged sample size boosts statistical power,  allowing a complete examination of varied consumer sentiments. This purposeful decision,  which is consistent with structural equation modelling (SEM) principles, emphasises the  research's dedication to producing reliable and relevant outcomes. 

• Sampling Technique 

Our research on "Navigating Customer Satisfaction: Unravelling the Dynamics of Chatbot  Performance and Trust in Technology" used a rigorous sampling approach, including stratified random sampling, random sampling with many stages, and random sampling, to ensure a  comprehensive and representative sample of participants. The population was divided into  groups based on essential factors such as gender, age, income, region, and education, and  people were selected at random. This systematic technique sought to capture variety, improving the quality and reliability of our data, by investigating possible differences in results  across various demographic groups in the areas of Chatbot Performance and Trust in  Technology. 

• Sample Design 

This research includes Bangalore City, the biggest in Karnataka, and aims to analyse consumer  preferences and behaviours in both private and public sector banks. Using a multi-stage random  selection technique, the study first covers all banks in Bangalore. In the next phase, four banks  with large market capitalization are explicitly chosen: HDFC, ICICI, State Bank of India, and  Bank of Baroda. The systematic selection of branches for data collection yielded a total sample  size of 350 clients from 28 State Bank of India branches, 9 Bank of Baroda branches, 11 HDFC  Bank branches, and 5 ICICI Bank branches. This comprehensive method attempts to give  important insights on clients' banking experiences in Bangalore City. 

          Table 1:Total Number Of Bank Branches In Bangalore city 

        Table 2: List Of Selected Private And Public Sector banks In Bangalore City

10.Data Collection 

Our study on " Navigating Customer Satisfaction: Unravelling The Dynamics Of Chatbot  Performance, Trust In Technology " involved 350 participants surveyed through structured  questionnaires. Ethical considerations were prioritized, ensuring informed consent and data  security. A stratified random sampling method, considering demographics, was used for  inclusivity. Participants chose between in-person interviews or online surveys for flexibility.  This approach aimed to gain reliable insights into diverse Navigating Customer Satisfaction:  Unravelling The Dynamics Of Chatbot Performance, Trust In Technology. 

10.1.Data Analysis 

In our work "Navigating Customer Satisfaction: Unravelling Chatbot Performance Dynamics,"  we used Structural Equation Modelling and Moderation analysis. Our extensive data analysis,  which included descriptive and inferential statistics, revealed important insights into the  interaction of key elements driving customer satisfaction. These results inform banks' strategic  choices, upgrading Chatbot services to improve client happiness and loyalty in the ever 

changing the AI environment of the banking business and beyond. 

• Analysis of Structural Equation Modelling (SEM) 

In this study, we use structural equation modelling, or SEM, to analyse complicated interactions  between many variables concurrently. SEM integrates regression and factor analysis, providing  a comprehensive understanding of relationships within a theoretical framework. It helps verify  and adjust research hypotheses, revealing detailed patterns and insights into the dynamics between MSME performance management, modern marketing strategies, and digital marketing  capabilities. 

• Chatbot Performance: Satisfaction levels result from setting and attaining consumer goals.  This entails understanding consumer perceptions, strengths, weaknesses, opportunities, and  threats. Strategies are then developed and implemented to achieve these goals. 

• Customer Satisfaction: Consumer perception drives an entity's growth and success, requiring  effective management and a balance between expansion and core values for long-term  sustainability. 

• Trust in Technology: Consumer trust in technology grows as banks leverage Artificial  Intelligence (AI) to enhance data analysis, develop strategies, and improve communication  with stakeholders. 

11.Results 

Table 3 Demographic variables

• Gender: The variable "Gender" has three categories – Male, Female, and Transgender. Among the 350  respondents, 175 (50.0%) identified as Male, 133 (38.0%) as Female, and 42 (12.0%) as  Transgender. 

• Marital Status: The "Marital Status" variable consists of two categories – Married and Single. Out of the total  respondents, 210 (60.0%) reported being married, while 140 (40.0%) identified as single. 

• Age Group: Five groups are identified by the "Age Group" variable: 18–25 years, 26–35 years, 36–45 years,  46–60 years, and Above 60+ years. Each category shows the respondents' percentage overall;  the highest number is 22.0% for the 18–25 age range. 

• Education Qualification: For the "Education Qualification" variable, respondents are categorized based on their  educational levels – Primary school level, Higher secondary level, Under Graduate level, Post  Graduate level, and University/Ph.D./Tertiary. The highest percentage (24.0%) falls under the  Primary school level. 

• Occupation: The "Occupation" variable includes categories such as Students, Govt. employee, Private  employee, Professional, House Maker, Business/Employer, and Retired. Among the  respondents, the highest percentage (18.0%) identified as Students. 

• Monthly Income: The "Monthly Income" variable is divided into four categories – Below 20000, 20000-50000,  50000-100000, and Above 100000. Each category represents a percentage of the total  respondents, with 26.0% reporting a monthly income in the range of 50000-100000 and Above  100000. 

• SEM (structural equational modelling) Structural Equation Modelling (SEM) is a flexible statistical method that describes complicated  relationships among variables, whether latent or visible. Its ability to analyze intricate causal  pathways, integrate latent components, test several hypotheses at once, account for  measurement error, evaluate model fit, and combine aspects of factor analysis and regression  are just a few of its special features. SEM is an essential tool for research in disciplines like  psychology, sociology, economics, and beyond because it can be used to validate theoretical  models, examine the effects of interventions or policies, and simplify complex datasets. This  allows for more thorough and accurate data analysis and hypothesis testing. 

• Measurement model and validity: Measurement models and validity are indispensable in research as they establish a structured  framework for ensuring the accuracy and meaningfulness of data. Measurement models clarify  the relationships between observed variables and their underlying constructs, enabling  researchers to assess complex concepts. 

Validity, on the other hand, ensures that the  measurement instruments precisely capture the intended constructs, safeguarding against  misleading or incorrect conclusions. Both measurement models and validity are essential  components in research, serving as the foundation for reliable and credible findings, which is  paramount for informed decision-making and advancing knowledge across diverse fields.

Table 4:Regression Weights: (Group number 1 - Default model)

Table 5:KMO and Bartlett's Test 

The KMO test and Bartlett's tests are used to determine whether factor analysis is suitable. The  factor analysis was validated by the very significant (P = 0.000) Bartlett's Examination, which  produced a KMO score of 0.955, indicating excellent sample quality.

Confirmatory Factor Analysis (CFA) verified our instrument by identifying factor loadings  greater than the 0.5 threshold, highlighting its excellent measuring capabilities. Items with  factor loadings less than 0.6 were eliminated. Table 4 provides details on internal consistency,  as measured by Results obtained after CFA, together with average variation extracted (AVE)  and the composite reliability (CR). Discriminant validity. is supported by Table 5, indicating  that the instrument can accurately assess the required components. 

Table 6:Post CFA, Cronbach Alpha, Factor Loadings

Discriminant validity 

Discriminant validity is a conceptual validation quality, not a particular statistical test. It  guarantees the distinctness of variables in a research, which is necessary for proper  measurement. Techniques such as confirmatory component analysis (CFA) and correlation  analysis demonstrate that constructs are indeed distinct. This eliminates duplication, improving  the accuracy of data processing and interpretation in research. 

Table 7: Discriminant validity Test

Table 5 shows good discriminant validity for the research constructs. Diagonal values (AVE  square roots) for Customer Satisfaction, Chatbot Performance, Trust Technology, and  Technical Literacy are more than 0.50, indicating uniqueness. Construct separation is shown  by off-diagonal correlations that are lower than AVE square roots. These data confirm that  variables may accurately assess various features of the study constructs. 

Table 8: Model Fit Summary 

It was determined that the following were appropriate representations of the sample data:  centeredness of fit (χ2 = 963.044), relative fit index (RFI) = 0.960, comparative fit index (CFI)  = 0.913, incremental fit index (IFI) = 0.914, goodness of fit (GFI) = 0.990, and normalised fit  index (NFI) = 0.992. These numbers are much higher than 0.90. Both the Root Mean Square  Error of Adaptation (RMSEA) = 0.064 and the Rational Mean Square Residuals (RMSR) =  0.042 are below the 0.080 criterion. As shown by the model's RMSEA of 0.064, RMR of 0.042,  GFI of 0.990, and CFI of 0.913, the data was well-fitted by it. 

Proposed Hypothesis: 

H1: Chatbot performance has a significant direct positive effect on customer satisfaction.

Table 9: Regression Weights: (Group number 1 - Default model) 

The hypothetical structural equation model shows a substantial positive association (β=0.969,  P<0.05) between Chatbot Performance (independent variable) and Customer Satisfaction  (dependent variable). This implies that Chatbot Performance has a significant influence on  increasing Customer Satisfaction in the analysed environment. With a standardised coefficient of 0.969, the path linking Chatbot Performance and Customer  Satisfaction demonstrates a strong positive relationship. Large correlation coefficient values  (C.R. values) and strong fit indices (p-values > 0.05 in Table 8) confirm the statistical  significance of this association, indicating that the model is robust. 

Table 10: Model Fit Summary 

The model has a strong fit, as shown by a χ2 of 165.698 and several fit indices: NFI = 0.958,  IFI = 0.967, GFI = 0.957, RFI = 0.923, and CFI = 0.967, all above 0.90. Additionally, the RMR  (0.030) and RMSEA (0.078) readings are less than the crucial threshold of 0.080. 

H2: Trust in technology significantly mediates the relationship between chatbot  performance and customer satisfaction.

    Table 11:Regression Weights: (Group number 1 - Default model) 

The research found that Trust Technology mediates the favourable association between Chatbot  Performance and Customer Satisfaction. Chatbot performance has a considerable effect on  Trust Technology (β=0.555, C.R.=16.337) and directly impacts Customer Satisfaction  (Chatbot: β=0.790, C.R.=29.837; Trust Technology: β=0.089, C.R.=3.375). This emphasises  the importance of effective Chatbot Performance and Trust Technology in improving Customer  Satisfaction. 

Table 12:Standardized Indirect Effects (Group number 1 - Default model) 

Table 10 shows the standardised indirect effects for Group 1 in the default model,  demonstrating the impact of one construct on another via intermediates. For example, CUSTOMER Satisfaction's indirect influence on Trust Technology via Chatbot Performance is.050, demonstrating the extent of the effects. Indirect impacts of Trust Technology are  indicated as.000, indicating its low influence in this model. 

Table 13:Model Fit Summary 

The model has a strong fit with a χ2 of 389.878 and fit indices surpassing 0.90, including NFI  = 0.910, IFI = 0.930, GFI = 0.912, RFI = 0.952, and CFI = 0.929. Additionally, RMR (0.070)  and RMSEA (0.074) values are less than the key 0.080 barrier, confirming the model's  sufficiency. 

H3: Technical literacy significantly moderates the relationship between chatbot  performance and customer satisfaction. 

Table 14:Regression Weights: (Group number 1 - Default model)

A Structural Equation Model (SEM) moderating Z score (Technical Literacy of Customers) is  shown in Table 12 and looks at the link between Zscore (Chatbot Performance) and Zscore  (Customer Satisfaction). The findings show that customer satisfaction and chatbot performance  have a strong positive correlation (β=0.737, P<0.05), and technical literacy has a major impact  on customer satisfaction (β=0.167, P<0.05). 

Moderation testing: 

The moderator variable in this study is Zscore (technical literacy of customers), the dependent  variable is Zscore (customer satisfaction), and the independent variable is Zscore (chatbot  performance). Using SPSS, the findings are computed by generating interaction terms from the  standardized score of the variables. 

Table 15: Regression Weights: (Group number 1 - Default model)

Testing Zscore (Technical Literacy of Customers) as a moderator indicates a significant  favourable influence (β=0.040, P<0.05) of the interaction term between Zscore (Chatbot  Performance) and Zscore (Technical Literacy of Customers) on Zscore (Customer  Satisfaction). Despite contradicting the hypothesised nature, the fit indices in Table 14  corroborate the model's overall goodness of fit, emphasising the relevance of Technical  Literacy in the Chatbot Performance-Customer Satisfaction link. 

Table 16: Model Fit Summary 

The model has an excellent fit with a χ2 of 23.339 and strong fit indices above 0.90, including  NFI = 0.971, IFI = 0.980, GFI = 0.975, RFI = 0.923, and CFI = 0.980. Furthermore, RMR  

(0.031) and the RMSEA (0.067) values are lower than the key 0.080 threshold, indicating the  model's adequacy. 

12.Discussion

In examining the hypotheses presented, it is evident that the performance of chatbots holds a  pivotal role in shaping customer satisfaction. The first hypothesis underscores the direct  positive impact of chatbot performance on satisfaction, aligning with the established notion  that efficient and effective interactions contribute to overall positive customer experiences.  Moving beyond this direct relationship, the second hypothesis introduces the nuanced role of  trust in technology as a mediator, suggesting that as trust increases, the positive effects of  chatbot performance on satisfaction are further amplified. This highlights the importance of  cultivating user trust in technology to enhance the effectiveness of chatbot interactions. The  third hypothesis adds the moderating role of technical literacy, demonstrating that users'  technological skill may alter the strength of the link among chatbot performance and pleasure.  Acknowledging these interrelated factors, businesses must adopt a holistic approach,  considering not only the technical capabilities of chatbots but also fostering user trust and  accounting for diverse levels of technical literacy. In doing so, a more comprehensive  understanding of the intricate dynamics involved in the user-chatbot interaction can be attained,  guiding the development and implementation of chatbot systems tailored to meet varied user  needs and ultimately optimize customer satisfaction. 

Conclusion 

The research concluded by looking at three theories on technical literacy, technological trust,  and chatbot performance in relation to customer happiness. First, using a well-fitting model  (p>0.05), the study validated H1, showing a substantial positive association between Chatbot  Performance and Customer Satisfaction (β=0.969, p<0.05). Second, with standardised  coefficients of 0.555 (p<0.000) and 0.790 (p<0.000), respectively, it was shown that Trust  Technology mediates the association between Chatbot Performance and Customer Satisfaction,  supporting the validation of H2. Additionally, there was a favourable direct impact of Trust  Technology on Customer Satisfaction (β=0.089, p<0.005). Finally, H3 suggested technical  literacy as a moderator. The findings showed that Chatbot Performance and Technical Literacy  significantly positively influenced Customer Satisfaction (β=0.040, p<0.05). This surprising  result implies that, in contrast to the relationship's predicted nature, technical literacy has a  moderating effect. Overall, the research emphasises how crucial effective chatbot performance 

is, how crucial technological trust is as a mediator, and how technical literacy functions as a  moderator when it comes to determining consumer pleasure in the digital age. 

References 

1. C. for the S. of S. Policy, “Customer satisfaction: What the research tells us,” p. 42,  2007. 

2. Aishwarya Gupta, “Introduction to AI Chatbots,” Int. J. Eng. Res., vol. V9, no. 07,  2020, doi: 10.17577/ijertv9is070143. 

3. S. Morsi, “Artificial Intelligence in Electronic Commerce: Investigating the  Customers’ Acceptance of Using Chatbots,” J. Syst. Manag. Sci., vol. 13, no. 3, pp.  156–176, 2023, doi: 10.33168/JSMS.2023.0311. 

4. N. R. Sanil, J. K. Johnson, E. Small, and F. Bank, “RESEARCH IN ENGINEERING  MANAGEMENT OPTIMISING BANK ’ S BRAND EXPERIENCES ON SOCIAL  MEDIA : THE ROLE OF CHATBOTS AND INTERACTIVE TOOLS,” pp. 363–371,  2023. 

5. S. Shafiee, L. L. Zhang, and K. M. Rasmussen, “Improving Financial Literacy and  Supporting Financial Decisions: Developing a Personalized Configurator,” J. Knowl.  Econ., no. 0123456789, 2023, doi: 10.1007/s13132-023-01651-9. 

6. Y. K. Dwivedi, J. Balakrishnan, A. M. Baabdullah, and R. Das, “Do chatbots establish  ‘humanness’ in the customer purchase journey? An investigation through explanatory  sequential design,” Psychol. Mark., vol. 40, no. 11, pp. 2244–2271, 2023, doi:  10.1002/mar.21888. 

7. M. Kamal and A. S. Himel, “Redefining Modern Marketing: An Analysis of AI and  NLP’s Influence on Consumer Engagement, Strategy, and Beyond,” Eig. Rev. Sci.  Technol. , vol. 7, no. 1, pp. 203–223, 2023. 

8. A. Aldoseri, K. N. Al-Khalifa, and A. M. Hamouda, “A Roadmap for Integrating  Automation with Process Optimization for AI-powered Digital Transformation,” 2023,  doi: 10.20944/preprints202310.1055.v1. 

9. K. Ambili, “Navigating growth : Unraveling the interplay between marketing strategies  and social media in the economic development of nations,” vol. 10, no. 02, pp. 1144– 1151, 2023.

10. H. Allioui and Y. Mourdi, “Unleashing the Potential of AI: Investigating Cutting-Edge  Technologies That Are Transforming Businesses,” Int. J. Comput. Eng. Data Sci., vol.  3, no. 2, pp. 2737–8543, 2023. 

11. S. Chandra, S. Verma, W. M. Lim, S. Kumar, and N. Donthu, “Personalization in  personalized marketing: Trends and ways forward,” Psychol. Mark., vol. 39, no. 8, pp.  1529–1562, 2022, doi: 10.1002/mar.21670. 

12. H. Hoang, “Navigating the Rise of Smart Tourism: Implications of Technology and  Data for Sustainable Industry Growth,” Brawijaya J. Soc. Sci., vol. 3, no. 01, pp. 1–18,  2023, doi: 10.21776/ub.bjss.2023.003.01.1. 

13. T. M. Tung, “An Analysis of the Scientific Researches on how Artificial Intelligence  affects Digital Marketing Initiatives,” Migr. Lett., vol. 20, pp. 1331–1337, 2023, doi:  10.59670/ml.v20iS8.5266. 

14. G. Park, J. Chung, and S. Lee, “Human vs. machine-like representation in chatbot  mental health counseling: the serial mediation of psychological distance and trust on  compliance intention,” Curr. Psychol., 2023, doi: 10.1007/s12144-023-04653-7. 

15. H. Nam and P. K. Kannan, “Digital Environment in Global Markets: Cross-Cultural  Implications for Evolving Customer Journeys,” J. Int. Mark., vol. 28, no. 1, pp. 28–47,  2020, doi: 10.1177/1069031X19898767. 

16. Y. Bao, W. Gong, and K. Yang, “A Literature Review of Human–AI Synergy in  Decision Making: From the Perspective of Affordance Actualization Theory,”  Systems, vol. 11, no. 9, 2023, doi: 10.3390/systems11090442. 

17. E. M. Lase and F. Nkosi, “Human-Centric AI: Understanding and Enhancing  Collaboration between Humans and Intelligent Systems,” Algorithm Asynchronous,  vol. 1, no. 1, pp. 33–40, 2023, doi: 10.61963/jaa.v1i1.49. 

18. P. GÜLTAŞ, “Strategic Insights: Navigating Digital Transformation and Customer Centricity for Enhanced Business Success,” InTraders Int. Trade Acad. J., vol. 6, no.  2, pp. 223–249, 2023, doi: 10.55065/intraders.1393775. 

19. S. (2019) TONTS, “Chatbots_ will they ever be ready?,” 2019. 

20. A. Hovsepyan and K. Johansson, “Study of AI Service Providers in IT Consulting,  Marketing, and Law,” 2023.