Abstract

The study aims to assess the potential of artificial intelligence (AI) for improving the customer experience in the banking sector by designing and developing a digital human sustaining the customer care. Digital humans are conversational robots that have transformed traditional human-human interactions into new disruptive machine-human interactions that are more reliable and exceptional, but also fragile. The study analyses Algho, a virtual assistant platform developed by QuestIT company, to automate the banking customer care service, and in particular the loan request. The key aspect of this process was creating and consolidating an effective Knowledge Base, the indispensable basis for a successful performance of the digital human. The research evaluates the performance of the virtual assistant, equipped with a digital human interface, through the creation of an Algho tester that can verify the effectiveness of the knowledge base. The results show that AI-based technologies can have a positive impact on business processes. The originality of the study lies in the training of a digital human according to business needs with a perspective of creating an optimised customer experience.

Keywords – digital transformation, digital marketing, artificial intelligence, customer experience, innovation

Paper type – Academic Research Paper

1    Introduction

Digital transformation is “a process that aims to improve an entity by triggering significant changes to its properties through combinations of information, computing, communication, and connectivity technologies” (Vial, 2019), which allows artificial intelligence-based technologies to emerge and be applied to support business activities and processes. Artificial intelligence (AI) supporting the customer experience is revolutionising the way companies interact with consumers (McLean and Osei-Frimpong, 2019). By developing AI technologies strategically, several key touch points can be triggered that could bring significant benefits to companies and the possibility of increasing customer satisfaction. In this context, chatbots emerge that act as a gateway to an immediate and satisfying experience and, by incorporating machine learning functions, learn from their own “mistakes” and user behaviour (Tebenkov and Prokhorov, 2021). The advantages of using chatbots are many, the most important of which is the 24/7 availability, i.e. the immediacy of such tools that can offer services at any time, leading the company to improve customer satisfaction and consequently its brand reputation.

The evolution of chatbots is represented by “Digital Humans”, which go beyond the obsolete concept of chatbots by embodying humanoid forms to relate with the customer in a more empathetic way (QuestIT, 2023). Digital Humans are conversational robots that represent the empathic evolution of virtual assistants. These tools are of enormous importance not only in the economic sphere but also in the social sphere, as they have transformed traditional human-human interactions into new disruptive machine-human interactions that are more reliable, exceptional but also fragile (Kaczorowska-Spychalska, 2019), in which a crucial part is the creation of trust (Przegalinska et al., 2019).  The anthropomorphic nature of the digital human under analysis highlights a number of psychological consequences that can potentially increase the engagement of the user with which it interacts (Alabed et al., 2022; Go et al., 2019), but can also occur a negative impact as a result of the chatbot disclosure, driven by a subjective human perception against machines, despite the objective competence of AI chatbots (Luo et al., 2019).

Despite the growing interest in the applications of Digital Humans for improving the customer experience, the role of digital technologies in improving the customer service experience through the adoption of digital human tools is still in an infancy phase. With the aim to cover this gap, the purpose of this paper is to assess the strategic potential of artificial intelligence for improving the customer experience (Ameen et al., 2021) in the banking sector. The aim is to design and develop a digital human, as a fintech application tool (Zarifis and Cheng, 2022; Hwang and Kim, 2021), as well as a digital marketing leverage (Paschen et al., 2019). Through the realisation of a virtual assistant trained for the purpose of automating the financing process on a banking platform, the advantages, opportunities and limitations of the research will be analysed. The crucial point in this process is the elaboration of a well-structured Knowledge Base, the indispensable basis for an effective performance of the digital human, which will be verified by the creation of a Knowledge Base tester.

Findings show that integrating AI systems can positively impact business processes.

Hence, the remaining sections of this paper are organised as follows. Section 2 presents an overview of the current literature, section 3 presents the research methodology. Then, section 4 is dedicated to the exposition of the findings, section 5 discusses the results in the light of the existing literature and section 6 concludes the paper by highlighting the implications.

2      Literature background

Artificial intelligence found its roots in 1950, with the publication of the article “Computing machinery and intelligence” written by Alan Turing on this avant-garde topic. It developed concretely from the 1940s onwards, a period in which the term “cybernetics”, i.e. the discipline that deals with the study of processes concerning “communication and control in the animal and in the machine” (Wiener, 1948), became widespread.

In order to meet consumers’ needs, financial intermediaries can deploy digital technologies that optimise operational processes, especially AI technologies that can speed up credit recovery activities, reduce credit management costs, optimise more complex processes, maximise performance, and enhance touchpoints for customer care services.

2.1 AI-enabled tools: Digital Humans

The main AI technology supporting customer care services is chatbots, ‘intelligent agents’ (Turing, 1950; Poole and Mackworth, 2010) based on systems aimed at communicating with users using natural language processing (Adam et al., 2020); they are designed to imitate human speech in approximating written text or vocal speech as best as possible to interact with people via a digital interface (Ling et al., 2021; Thomaz et al., 2020). They become the technological reflection of humans leading to “the dehumanisation of what is human and the humanisation of technology in all its manifestations” (Kaczorowska-Spychalska, 2019).

As a result, chatbots are gradually replacing human service agents on websites, social media platforms, and messaging services. According to industry experts, the market for chatbots and related technologies is projected to surpass $1.34 billion by 2024 (Wiggers, 2018). While some experts argue that chatbots can improve customer service while also reducing costs (De 2018), others fear that they may have a negative impact on firms and customer service (Kaneshige and Hong 2018). Hence, determining the optimal design and implementation of chatbots for customer service purposes remains an open question, as businesses weigh the benefits and drawbacks of these tools in delivering efficient and effective customer service (Crolic et al., 2021).

The shift from chatbot to digital human represents a move towards creating AI systems that can simulate human-like interactions, emotions, and behaviours to a greater extent than current chatbot technology. A digital human goes beyond the scripted responses of a chatbot and can respond more naturally to complex queries and even understand nonverbal communication such as facial expressions, tone of voice, and body language. The development of digital humans involves using advanced algorithms and machine learning techniques, including deep learning and natural language processing, to create more sophisticated and personalised interactions with users, as well as being equipped with a digital human interface (QuestIT, 2023).

2.2 Digital Humans in the banking sector

The digitalisation of businesses, in which digital technologies are used to change business models and create new revenue opportunities, is transforming the way that service providers and customers interact (Oviatt and Cohen, 2015; Hennig-Thurau et al., 2010; Pousttchi and Dehnert, 2018; Schmidt et al., 2017). As new technologies emerge, it can be assumed that the digitalisation-enabled transformation in services will be even further amplified (Gomber et al., 2018). The financial service sector has already experienced the impact of digitised services such as mobile wallets, payment apps, and automated wealth advisors that have entered the market as replacements for established banking services (Basole and Patel, 2018). To fully realise the potential of digitalisation-enabled service transformation, it is important for those leading the transformation to see its potential from a customer perspective (Kandampully et al., 2021; Lähteenmäki et al., 2022).

The finance industry has been at the forefront of utilising AI technology, including chatbots, and is seeing an increase in the number of financial institutions adopting chatbots to enhance customer service (Adobe, 2019; Sarbabidya and Saha, 2020; Suhel et al., 2020). This has led to progress in research on chatbot technology in the finance industry, with many researchers studying algorithms and systems that reflect the unique characteristics of the industry. Financial firms face challenges in providing customers with the right information due to the complexity of the financial system, resulting in time and labour lost on simple inquiries (Chaitrali et al., 2017).  Chatbots have been proposed as a solution to answer frequently asked questions using natural language processing (NLP), providing customers with quick and efficient service (Elcholiqi and Musdholifah, 2020; Rajbabu et al., 2019; Yu et al., 2020).  Several banks have embarked on a digital transformation process (Zhang, 2021), and chatbots are widely recognized as a vital and necessary element of this transformation, as well as a sustainable strategy for the development of the banking industry (Forbes, 2021). Chatbots are frequently employed by banks in their marketing efforts, sales activities, and customer relationship management practices (Eren, 2021), enabling them to deliver quick, affordable, and customised services to their customers. Nguyen et al. (2021) suggest that to keep customers using chatbot services in banking, providers must ensure their chatbots are trustworthy, useful, and meet customer needs. They can do this by focusing on three quality aspects of chatbot services: information quality, service quality, and system quality. Information quality is critical for users’ trust and satisfaction, and providers must offer relevant and up-to-date information. Service quality is essential to provide accurate information and prompt responses to users’ queries. Personalised services, addressing customers by name, can alleviate uncertainty and build trust. System quality is the most important factor in building users’ trust in chatbots, and banks must provide a stable, attractive, and adaptable interface with 24/7 support. Additionally, they must consider systematic risk and users’ privacy when developing chatbots, particularly in emerging markets with weak legal protection. Providing high-quality chatbot services can result in many benefits for banks, such as a good reputation and positive image.

Digital Humans are the right tool for banking customer care as they facilitate navigation, provide live insights, are multimodal, predictive, inclusive, and optimise work, automating first-level processes, facilitating the work of operators and reducing costs (QuestIT, 2023).

The study therefore aims to enrich the existing literature with more knowledge on the shift from chatbots to digital humans, by describing the key features of these innovative AI-enabled digital tools. The focus of the work is on the automation of banking customer care, thanks to the Algho platform, to examine the potential impact of digital humans on the customer experience.

3      Research methodology

The study adopts the case study (Yin, 2003) of Algho, a technology produced and developed by QuestIT and a registered trademark of The Digital Box. Algho is a virtual assistant, based on latest-generation 3D technology with human features, with the aim of creating an empathic relationship with the user and thus establishing a new and effective User Experience. The research question driving this study is: How does the implementation of Digital Human for banking impact the customer experience?

To address this question, a digital human was created and developed using the Algho Platform in order to automate the financing process on a banking website. Fundamental in this phase is certainly the process of creating and consolidating a “Knowledge Base” capable of understanding and solving the user’s specific problems and earning the user’s trust. To evaluate the effectiveness of the virtual assistant, an Algho Knowledge Base tester was created to optimise the performance.

4          Findings

The first step in proceeding with this study was to realise the digital human according to business needs. We proceed to choose the name, the voice, the language, the gender. We then proceeded to insert a welcome and welcome back message. The digital human was created and trained completely in Italian, as the bank for which it was developed is Italian.

Figure 1. Creation of the Digital Human

 Next, we set the colours of the chat, the company logo and settings related to the speed of the assistant’s response. The latter setting allows you to set a slower response type, if you are trying to simulate human interaction, or a faster one.

There is the possibility of setting up the ‘Live chat’ function, which allows a human operator to enter the chat and replace the assistant in the event that the user, for example, asks to interface with a real person, or in the event that after a pre-set total of questions that the digital human is unable to answer, the automatic call to the human operator starts. By means of the ‘Parameters’ setting section, one can set the degree of flexibility, confidence and correspondence one wishes to give the virtual assistant. Specifically, the correspondence refers to the degree of importance the digital human attaches to the semantic and syntactic component in order to provide the most suitable answer to the question received from the user. The “Reinforcement” section allows the user to set various types of learning that the assistant can perform automatically. This function allows the user to rate the digital human’s answer positively or negatively, thus making him/her remember whether a particular answer was useful or not in order to help him/her with future questions. The risk, however, is that the answer is voted negatively even if it is correct, for whatever reason or problem not arising from the chatbot.

After setting these technical settings, we entered the phase of creating the knowledge base of the digital human. The platform provides pre-set basic questions for small talks, and to create custom questions, one must formulate the question and enter an appropriate answer that satisfies different variations of the question. Conversational forms can also be added, along with a link function to guide users to relevant pages. Multiple answers can be created for a single question and targeted according to the user’s focus. To train the virtual assistant, indirect or direct information can be sought to convert the user into a customer. Trust and a suitable tone of voice must be established to adapt to a broader range of customers, especially in the banking sector. A solid knowledge base must be created to understand and solve the user’s specific problems, building trust with the user.

The Knowledge Base Test was the key point to improve the virtual assistant’s ability to answer questions correctly. In fact, thanks to this test, which was carried out by entering all the questions in the virtual assistant’s Knowledge Base with variants that a user could ask in an excel file uploaded directly to the Algho platform, the latter provided the result as a percentage of this performance. Thus, the Knowledge Base was optimised until a 100% performance result was obtained.

5          Discussion

In our study, we examined the historical evolution of chatbots and how they have now turned into Digital Humans, conversational robots that represent the empathic evolution of virtual assistants. Specifically, we analysed the case study of the Algho platform, which gives the possibility of training a Digital Human to assist users by reshaping the User Experience.

Then, we designed and trained a Digital Human for banking customer service, working on the possibility of optimising banking processes such as a loan request. This process represents a great advantage for the market, giving the possibility of optimising costs and employing human resources in more complex and difficult to predict activities. Several studies indicate that the use of such AI tools offers substantial advantages, especially with regard to interactions with customers, who form more personal bonds with robots than with an AI without a physical embodiment.

A wide range of potential functions and advantages offered by chatbots are the subject of reflections that will certainly be deepened in the coming years, in the face of new interactions with customers that represent an interesting win-win. The customer in fact obtains a constantly evolving ‘product’ that satisfies his needs, and companies are motivated to optimise it to increase the customer lifetime value (Kaczorowska-Spychalska, 2019).

The provision of accurate, timely, and pertinent information by chatbots can enable users to make prompt and accurate financial decisions. Users are likely to experience increased satisfaction when they perceive chatbots as trustworthy (Gao et al., 2015). Notably, the quality of service is the primary determinant of user satisfaction (Nguyen et al., 2015).

According to Nguyen et al. (2015) customers tend to continue using chatbot services only if they perceive them as reliable and useful and if their needs are met. Therefore, banking service providers should pay close attention to three key quality aspects of chatbot services – information quality, service quality, and system quality – and ensure that users’ expectations are met. Understanding customers’ expectations is a vital first step for banks to provide timely solutions that satisfy them. Hence, to promote customers’ intentions to continue using chatbot services, their expectations must be met or even exceeded. Furthermore, given the significant role of perceived usefulness in the relationship between satisfaction and continuance intention, banks must ensure that their chatbot services are error-free, as service failures can prevent customers from getting what they are looking for, causing dissatisfaction. Banks should also anticipate users’ most common questions or requests and program chatbots to perform their tasks efficiently. In addition, interactions between users and chatbots should be efficient and user-friendly (Nguyen et al., 2015).

Endowed with personality and based on self-improvement mechanisms, Digital Humans could become our faithful reflection in the future (Kaczorowska-Spychalska, 2019).

6          Conclusion and implications

Having created and trained the virtual assistant, the enormous potential of such a tool emerged, as not only does it have a simple and intuitive interface, it also allows the digital human to be trained and tested in an iterative manner. This makes it possible to constantly improve its performance in order to offer the user an innovative and immersive customer experience.

The main managerial implications that emerge from our study are first the possibility of automating certain processes currently entrusted to human resources so that certain business activities and processes can be optimised. Second, a growing awareness has emerged with respect to the benefits that these tools can bring, but a low ease of use, so the focus on which there is a need to work is on increasing knowledge of these technologies. In addition, future research, building on the implications of our study, could deepen this topic by analysing the psychological factors underlying human-machine interactions to improve and optimise the customer experience, reducing the current resistance with respect to these technologies and thus increasing the propensity to adopt them.

The ideal evolution of the project we have outlined is represented by the future possibility of placing one’s trust in AI tools that, in a completely automated manner, can on the one hand provide all those operational banking services such as applying for a loan, opening an account, or requesting a debit or credit card, and on the other hand also provide all those advisory services that, based on the customer’s need and financial situation, can help him or her orientate themselves in the investment choices that are best suited to them.

Underlying possible future developments is an awareness on the part of individuals of how such technologies can improve their customer experience. Furthermore, in an optimal process improvement and optimisation in addition to automating the totality of banking services, a possible future development that would be of incredible importance is the possibility of automating customised advisory services according to customer requests. This process would be the natural evolution of the so-called robo-advisors, i.e. those tools that are aimed at providing investment solutions in a totally, or almost totally, automated manner; in order to improve the customer experience supported by artificial intelligence tools.

Future research could also analyse and test the reaction and behaviour of users when using such a tool, in order to better understand the role of artificial intelligence on the customer experience.

References

Adam, M., Wessel, M., & Benlian, A. (2020). AI‐based chatbots in customer service and their effects on user compliance. Electronic Markets, 1–19.

Adobe. (2019). Digital Trends: Financial Services in Focus. Retrieved from https://www.ad obe.com/content/dam/acom/uk/modal-offers/2019/DT-Report-2019/Econsultancy-2019-Digital-Trends-Financial-Services.pdf.

Alabed, A., Javornik, A., & Gregory-Smith, D. (2022). AI anthropomorphism and its effect on users’ self-congruence and self–AI integration: A theoretical framework and research agenda. Technological Forecasting and Social Change, 182, 121786.

Ameen, N., Tarhini, A., Reppel, A., & Anand, A. (2021). Customer experiences in the age of

artificial intelligence. Computers in human behavior, 114, 106548.

Basole, R. C., & Patel, S. S. (2018). Transformation through unbundling: Visualizing the global FinTech ecosystem. Service Science, 10(4), 1–18.

Chaitrali, S. K., Amruta, U. B., Savita, R. P., & Satish, S. K. (2017). Bank chat bot – an intelligent assistant system using NLP and machine learning. International Research Journal of Engineering and Technology, 4(5), 2374–2377.

Crolic, C., Thomaz, F., Hadi, R., & Stephen, A.T. (2021). Blame the Bot: Anthropomorphism and Anger in Customer–Chatbot Interactions. Journal of Marketing, 86, 132 – 148.

De, Akansha (2018), “A Look at the Future of Chatbots in Customer Service,” ReadWrite (December 4),
https://readwrite.com/2018/12/04/a-look-at-the-future-of-chatbots-in-customer-service/.

Elcholiqi, A., & Musdholifah, A. (2020). Chatbot in bahasa Indonesia using NLP to provide banking information. Indonesian Journal of Computing and Cybernetics Systems, 14(1), 91–102.

Eren, B.A. Determinants of customer satisfaction in chatbot use: Evidence from a banking application in Turkey. Int. J. Bank Mark. 2021, 39, 294–311. [CrossRef]

Forbes. Every Bank Needs A Chatbot (Or Two) For Its Digital Transformation. 2021.
https://www.forbes.com/sites/ronshevlin/2021/03/15/every-bank-needs-a-chatbot-or-two-for-its-digital-transformation/?sh=798b1c9275d7

Gao, L.; Waechter, K.A.; Bai, X. Understanding consumers’ continuance intention towards mobile purchase: A theoretical framework and empirical study–A case of China. Comput. Hum. Behav. 2015, 53, 249–262. [CrossRef]

Gomber, P., Kauffman, R. J., Parker, C., & Weber, B. W. (2018). On the FinTech revolution: Interpreting the forces of innovation, disruption and transformation in financial services. Journal of Management Information Systems, 35(1), 220–265.

Hwang S, Kim J., (2021) Toward a Chatbot for Financial Sustainability. Sustainability, 13(6):3173.

Kaczorowska-Spychalska,D., (2019) How chatbots influence marketing. Management,23(1) 251-270.

Kandampully, J., Bilgihan, A., Bujisic, M., Kaplan, A., Jarvis, C. B., & Shukla, Y. S. (2021). Service transformation: How can it be achieved? Journal of Business Research, 136, 219–228.

Kaneshige, Tom and Daniel Hong (2018), “Predictions 2019: This is the Year to Invest in Humans, as Backlash Against Chatbots and AI Begins,” Forrester (November 8), https://go.forrester.com/ blogs/predictions-2019-chatbots-and-ai-backlash/.

Hennig-Thurau, T., Malthouse, E. C., Friege, C., Gensler, S., Lobschat, L., Rangaswamy, A., & Skiera, B. (2010). The impact of new media on customer relationships. Journal of Service Research, 13(3), 311–330.

Lähteenmäki , I , Nätti , S & Saraniemi , S 2022 , ‘ Digitalization-enabled evolution of customer value creation : An executive view in financial services ‘ , Journal of Business Research , vol. 146 , pp. 504-517 . https://doi.org/10.1016/j.jbusres.2022.04.002

Ling, E. C., Tussyadiah, I., Tuomi, A., Stienmetz, J., & Ioannou, A. (2021). Factors influencing users’ adoption and use of conversational agents: A systematic review. Psychology & Marketing, 2021.

Luo X., Tong S., Fang Z., Qu Z., (2019) Frontiers: Machines vs. Humans: The Impact of Artificial Intelligence Chatbot Disclosure on Customer Purchases. Marketing Science 38(6):937-947.

McLean, G., & Osei-Frimpong, K. (2019). Hey Alexa… examine the variables influencing the use of artificial intelligent in-home voice assistants. Computers in Human Behavior, 99, 28–37, 2019.

Nguyen DM, Chiu Y-TH, Le HD. Determinants of Continuance Intention towards Banks’ Chatbot Services in Vietnam: A Necessity for Sustainable Development. Sustainability. 2021; 13(14):7625.

Oviatt, S., & Cohen, P. R. (2015). The paradigm shift to multimodality in contemporary computer interfaces. Synthesis Lectures on Human-Centered Informatics, 8(3), 1–243.

Paschen, J., Kietzmann, J. and Kietzmann, T.C. (2019), Artificial intelligence (AI) and its

implications for market knowledge in B2B marketing, Journal of Business & Industrial

Marketing, Vol. 34 No. 7, pp. 1410-1419.

Poole, D.L. and Mackworth, A.K. (2010), Artificial Intelligence: Foundations of Computational Agents, Cambridge University Press, Cambridge.

Pousttchi, K., & Dehnert, M. (2018). Exploring the digitalization impact on consumer decision-making in retail banking. Electron Markets, 28, 265–286.

Przegalinska, A.K., Ciechanowski, L., Stróz, A., Gloor, P.A., & Mazurek, G. (2019). In bot we trust: A new methodology of chatbot performance measures. Business Horizons.

QuestIT, (2023). https://www.alghoncloud.com/artificial-human/

Rajbabu, M., Prabhuraj, P., & Jeyabalan, S. (2019). An intelligent behavior shown by chatbot system for banking in vernacular language. International Research Journal of Engineering and Technology, 6(3), 1210–1212.

Sarbabidya, S., & Saha, T. (2020). Role of chatbot in customer service: A study from the perspectives of the banking industry of Bangladesh. International Review of Business Research Papers, 16(1), 231–248.

Schmidt, J., P. Drews and I. Schirmer (2017), “Digitalization of the banking industry: a multiple stakeholder analysis on strategic alignment,” Twenty-third Americas Conference on Information Systems, Boston, 2017.

Suhel, S. F., Shukla, V. K., Vyas, S., & Mishra, V. P. (2020). Conversation to automation in banking through chatbot using artificial machine intelligence language. In Paper presented at the 2020 8th international conference on reliability, infocom technologies and optimization. Noida: IEEE.

Tebenkov, E., and Prokhorov, I., (2021) Machine learning algorithms for teaching AI chat bots, Procedia Computer Science, Volume 190, 735-744, ISSN 1877-0509, https://doi.org/10.1016/j.procs.2021.06.086.

Thomaz, F., Salge, C., Karahanna, E., & Hulland, J. (2020). Learning from the Dark Web: Leveraging conversational agents in the era of hyper‐privacy to enhance marketing. Journal of the Academy of Marketing Science, 48(1), 43–63.

Turing, A. (1950). Computing machinery and intelligence. Mind, 59: 433–460. Reprinted in Haugeland [1997]. 5, 40

Vial, G. (2019). Understanding digital transformation: A review and a research agenda. The journal of strategic information systems : incorporating International Information Systems, 28(2), .

Wiener, Norbert, 1948, Cybernetics, or Communication and Control in the Animal and in the Machine, Cambridge, The MIT Press.

Wiggers, Kyle (2018), “Google Acquires AI Customer Service Startup Onward,” VentureBeat (October 2), https://venturebeat.com/
2018/10/02/google-acquires-onward-an-ai-customer-service-startup/.

Yin, R.K. (2003). Case Study Research: Design and Methods. Sage. Thousand Oaks, California.

Zarifis, A., Cheng, X., (2022) A model of trust in Fintech and trust in Insurtech: How Artificial Intelligence and the context influence it. Journal of Behavioral and Experimental Finance, 36 [100739]

Yu, S., Chen, Y., & Zaidi, H. (2020). A financial service chatbot based on deep bidirectional transformers. arXiv:2003, Article 04987. Zhang, K. What Does Smart Banking Look Like in 2021? (2021). Available online: https://www.sld.com/blog/brand-strategy/smart-banking-2021/