Nyamweya is a Kenyan scholar who has done many years of research on a diversity of topics
Overview and Research Questions
The research aims to examine the impact of artificial intelligence tools on customer experiences in China’s financial industry. Modern commerce requires businesses to employ strategies that enable them to manoeuvre their landscape. Conditions in the market that inevitably influences business performances such as stiff competition and fluctuating product/service demands continuously compel businesses to assess and alter production and management policies (Parise et al., 2016). Alam (2020) argues that customer satisfaction and loyalty is one of the most effective ways of improving and guaranteeing top firm performance. Moreover, the scholar suggests that technology use in business enables organisations to ensure service quality, which in turn improves customer experience. In support of the claim, Sujata et al. (2019) posit that artificial intelligence (AI) promises to revolutionise global markets in a bid to improve customer experiences. The findings of the study reveal that AI can adequately bridge the gap between customers and the business guaranteeing the former enormous possibilities of increased comprehension of products and services rendered. Particularly, AI promotes value creation and proffers prospects for substantial market innovation (Wu et al., 2020).
According to Westerheide (2020), China aims to become the initial artificial intelligence superpower and has already laid the groundwork for the accomplishment of this objective throughout its industries. McKinsey and Company (2020) opines that AI enables banks and other companies providing financial services to meet the ever-rising customer expectations. According to the source, AI employs smart servicing and provides intelligent propositions. The Chinese government continuously invests in AI through research and development (RandD) and strives to implement it in various sectors to promote rapid development. The nation has imposed a 2025 target by which it must connect and upgrade all its sectors including the financial services industry to balance supply and demand while refining service provision and goods production. With an ever-increasing AI adoption rate Chinese companies seek to transform service delivery within their sectors tremendously (Xueqing, 2019). In this regard, the study will contribute to the existing knowledge by investigating how the adoption of AI tools in the Chinese financial service industry will influence customer experience. The following research questions will guide the course of the study:
- What are the impacts of AI personalisation on customer experience in financial services?
- What are the impacts of AI quality of service on customer experience in financial services?
- What are the impacts of AI hassle-free service on customer experience in financial services?
2. Literature Searches
The review of literature plays a pivotal role in the development of a research project. For this reason, it potentially becomes a vital starting point for establishing a strong foundation for an investigation. Essentially, the rationale is to proffer a critical contextual understanding of the phenomenon under scrutiny from the perspective of other scholars (Winchester and Salji, 2016). Otherwise stated, the literature review not only entails scouring relevant avenues but also critical inspection of existing knowledge to develop a basic understanding of a problem. Snyder (2019) refers to the literature review as the building block of scholarly examinations and notes that it entails developing one's study around existing knowledge. Contextually, the process enables the investigator to focus the paper in anticipation and preparation for answering and addressing key issues in the research question. Normally, the preliminary step involves searching, locating, and collecting texts and publications that contain relevant information that provide insights on the problem under investigation.
Presently, the internet and associated technological advancement have simplified the search process considerably making it easier for researchers to access pieces of literature for scrutiny; a notable improvement in the investigative process in relation to past decades. The internet has made a difference, enabling scholars to obtain information quite easily. In their discussion, Winchester and Salji (2016) pointed out that the difficulty in managing the bulk material that ensues from the search process is one of the main underlying issues that affect the literature review process. To address the challenge the extant review process will endeavour to use a systematic approach as Snyder (2019) suggests in his analysis of the literature review methodology. The systematic process according to Snyder (2019) involves integrating standpoints and findings derived from empirical results collected from various publications. The study will thus, conduct a systematic literature review in the section to lay the ground for the study to respond to the research questions adequately. Derived from the step-by-step review concept, the systematic literature review is a methodology utilised in most quantitative studies. Similarly, systematic literature review occurs in meta-analyses which fundamentally involve the synthesis of information and results from a collection of quantitative researches. The suitability of the use of the methodology hinges on the principal nature of quantitative studies to demand a critical focus of the examination as determined by the research questions formulated to guide the investigation (Winchester and Salji, 2016). In this case, systematic literature review will attempt to focus the analysis on the financial service sector to respond to each of the research questions.
Interestingly, systematic literature reviews have quite varied characteristics, which make them suitable for use in the other two research methods; the qualitative and mixed research methods. Normally, this occurs through the alteration of specific facets of the principal technique to suit the needs of the research, which usually has direct links to the research problem, objectives, questions, or hypothesis in some instances. Since the present study utilises the quantitative research method, the researcher will summon and implement a full-fledged systematic literature review complete with a rigorous and critical assessment of the subtle features of studies published on the subject under investigation. Particularly, the review will accentuate the analysis of the study aims, key methodological procedures, the findings, and the ultimate implications on the anticipated research outcomes of the present study. Despite the laborious and time-consuming nature of the systematic literature reviews, its use could potentially prove beneficial to the investigation in the long term by providing a relevant framework to build the study upon (Winchester and Salji, 2016).
Artificial intelligence, a new phenomenon in the digital age is an under-researched area in academia. Significant yet, a gap exists in the literature that establishes the influence of artificial intelligence on financial services. In this case, the systematic literature will enable the review to present an exhaustive criticism of the challenges and opportunities that arise (Winchester and Salji, 2016) thanks to the implementation of artificial intelligence in the financial service sector. Essentially, this will help the investigator gather crucial information that enhances the individual's understanding of the research topic. Besides, it will enable the researcher to acquire key insights assessed, observed, and reported by past scholars on artificial intelligence and the methodology used within the studies. Significantly, this will help the researcher refine the present study to ensure the study outcomes meet the stipulated research aims.
Moving on, the literature review process will have three notable stages including, literature search, literature selection, and literature write-up (actual review). The review process will begin with the search procedure where the researcher engages specific practices aimed at locating only relevant pieces of literature that provides insights on artificial intelligence. Firstly, the researcher will need to identify the relevant resource databases which contain the pertinent pieces of literature required for the review. The researcher will conduct searches on known academic databases such as Google Scholar, Jastor, Academic search premier, Emerald, ProQuest, and Ebscohost. Important yet, the scholar will use notable academia-based databases for the search process including China Academic Journals, CADAL (China Academic Digital Associative Library), and Hong Kong Journals. The university search engine is another suitable alternative. Essentially, the use of multiple search avenues/databases is to enhance the outcomes and quality of the search process. The researcher will design specific search terminology and combinations that will enable them to collect relevant resources for use in the review. In this case, keywords such as 'artificial intelligence', customer experience, financial services. Better still, the researcher will endeavour to utilise notable synonyms such as 'AI', 'robotics', 'client perception', 'job satisfaction' and 'banking services' to replace the keywords for expansive information and to improve search outcomes for literature that directly address issues raised in the research question. Additionally, the process will entail the use of the relevant keywords along with Boolean operators such as AND, OR, and NOT to locate relevant studies to use in the review process as in the following examples: (a) Artificial Intelligence AND customer experience, (b) job satisfaction OR customer experience.
Another key element is the implementation of the inclusion and exclusion parameters, which enables the researcher to refine the literature search process. First, since this is quantitative research the researcher will give top priority to past studies conducted using a similar research method. For this reason, the research will strive to limit and reduce the literature bulk by restricting resource suitability to methodology type. Secondly, the review will limit the search to time frames (2016-2020) to ensure information credibility and validity. After accumulating enough resources, the researcher will proceed to review the literature by assessing the findings and comparing them to establish common links and disparities.
3. The Overall Research Strategy
The quantitative research method will be used in the extant study. Fundamentally, its choice for the research is informed by various factors. One overriding factor is the need to harmonise subsequent elements of the study with the previously selected systematic literature review system for the literature review. Apuke (2017) clarifies that the quantitative research methodology chiefly quantifies and analyses variable quantities within a study to obtain definite outcomes. Important still, it sustains research processes that purpose to answer the who, how, what, how much, how many, when, and where questions. Normally, these kinds of studies utilise given statistical methods to collect and examine numerical data. Secondly, the choice of the quantitative research method for this study hinges on the focus of the research question and research aims which seek to test the influence one phenomenon has on another. The characteristic element of the quantitative method allows the researcher to appraise key variables to find the association between them (Guetterman and Fetters, 2018). In this case, the study will strive to ascertain the connection between artificial intelligence and customer experience, the main variables in the study. The rationale for this attribute is the capacity of the method to sustain the objective appraisal of the research problem since it supports the use of primary data collected directly from credible sources and analysed using statistical techniques to enhance result accuracy.
On the other hand, the qualitative research method would not sustain the study aims satisfactorily since its principal tenets are not appropriate for answering the research question. Singh et al. (2020) provide that qualitative studies are primarily explorative, and the lack of objectivity is their main limitation. The main disparity as outlined by Guetterman and Fetters (2018) revolves around their principal aims. The qualitative research strategy explores people's feelings and thoughts. On its part, the quantitative research strategy seeks to measure people's (specific population) thoughts and feelings (Queirós et al., 2017). Another notable difference is that while quantitative methods use larger samples the qualitative methods mostly utilise smaller samples. The major benefits of adopting the quantitative research method include increased accuracy, objectivity, speed and ease of use, effectiveness, and last but least it enables the generalisation of study outcomes since it allows the use of large samples (Queirós et al., 2017).
The qualitative research method primarily seeks to develop an understanding of complex phenomena or to elucidate their occurrence in specific contexts (Queirós et al., 2017). On the contrary, the present study looks to focus the investigation on establishing links between key factors such as AI personalisation, AI quality of service, and AI hassle free service. While the qualitative method will employ an explorative approach to assess the occurrence of these elements without a particular focus, the quantitative method endeavours to determine the impact of each factor on customer experience independently. Essentially, this makes it possible for the study to offer informed insights and issue practical recommendations that help players in the financial services industry enhance the customer experience. The analysis process of data collected using the quantitative research method will entail the use of numerical representation such as pie charts, tables, and graphs. The identified strategy will help in evaluating the relationship between artificial intelligence tools and the customer experience in the Chinese financial service industry. It will assess whether the relationship is poor or strong.
In addition, the study will employ the research philosophy of positivism. The choice of the quantitative research method agrees with the positivism research philosophy, which is the basis for objective reasoning in the world of research (Alharahsheh and Pius, 2020). The positivism research philosophy creates a favourable context for the development and implementation of various related procedures that will be discussed in the subsequent sections such as research design, data collection, and analysis. The principles of the positivism philosophy are founded on the perspective that factual information obtained via direct observation is reliable (Ryan, 2018). Moreover, positivism is primarily dependent on measurable observation's which ultimately culminate in the statistical examination. The main difference between the positivism and interpretivism research philosophy is that the former employs a deductive research approach while the latter adopts an inductive research approach.
4. Research Design Features
Based on the design of the research questions, the study essentially aims to examine the effect or association between phenomena within the research. Overall, four key variables have been identified for examination including customer experience, AI personalisation, AI quality of service, and AI hassle-free service. To collect, record, and analyse data that can show the relationship between different data set, the study must employ quantitative methods (Bryman, 2016). For this reason, the researcher will need to select research designs that allow the examination of various variables (data set) to reveal their association. The condition informs the suitability of the correlational research design which conveniently allows the study to assess and establish the relationship between the dependent and the independent variables in a study (Bryman, 2016). For instance, in this case, the design will ultimately reveal how AI personalisation, AI quality of service, and AI hassle-free service (independent variables) relate (influence) to customer experience (dependent variable). This way, the study can respond to the research questions which seek to reveal the association between the mentioned variables.
The extant research will employ the survey research strategy to collect data for analysis. For one the choice of the design hinges on the use of the quantitative research method, which allows the use of surveys to help generate statistical data. As established above, the quantitative method supports the objective examination of research problems based on the research philosophy of positivism. In addition, the study endeavours to investigate phenomena in the field of business and marketing hence, the use of surveys as a method for collecting data. Qualtrics (2020) defines surveys as the research methodology that entails gathering data from a group of human participants while aiming to generalise the outcomes of the study on the broader population. The suitability of this choice depends on various characteristics of surveys that make them appropriate for use in quantitative studies (Fulton, 2018). Firstly, surveys have low-cost budgets, which make them considerably convenient for most scholars. Secondly, surveys are extensive in nature, hence, can sustain and facilitate the acquisition of a broad range of information relevant to the large population that the study endeavours to examine. It allows the study to describe the features of a sample thereby guaranteeing more accurate data, which in turn enhances the quality of the study results (Qualtrics, 2020). Thirdly, surveys score highly on flexibility since they present a variety of options for administration to the researcher. The extant study will opt for online surveys owing to their ease of use and convenience in comparison to other modes such as telephone, paper, and face-to-face interview surveys. Furthermore, Sokolova and Titova (2018) argue that online surveys are cheaper to administer than phone and paper surveys for example. Fourthly, surveys are dependable as they address some ethical requirements that may affect the study process such as ensuring anonymity.
Moving on, the study will design and administer questionnaires to a designated sample of participants. Researchers conveniently use the instrument in their studies to collect quantitative data. The main merit of the tool concerns its practicability and suitability for collecting considerable information from large populations within a short time (Fulton, 2018). Secondly, since it generates statistical data, the tool helps avoid human interference because the outcomes are dependent on software analysis. Thirdly, it sustains an objective assessment of the problem, which is a primary property of the positivism philosophy (Fulton, 2018). The instrument will be designed to gather the information that directly responds to the research question and will constitute three different sections. The first section will contain warm-up questions related to the research topic that fundamentally acquaint the respondent with the larger survey thematic questions. The second section will contain questions that seek to assess the demographic features and composition such as gender, age, experience distribution across the sample. On its part, the third part will evaluate both the independent and dependent variables including AI personalisation, hassle-free services, quality of services and customer experience. The evaluation will be through the Likert scale to examine respondents' degree of agreement or disagreement with the statements (where 1 – Strongly agree, 2 – agree, 3 – neutral, 4 – disagree, and 5 – strongly disagree). As determined earlier, the researcher will administer the questionnaire online in compliance with the current COVID-19 social distancing regulations as is the circumstance with paper surveys. The questionnaire will be administered to customers of financial services companies in China.
In research, sampling pertains to the processes that enable researchers to recruit participants for a study (Amato et al., 2017). Based on the already selected procedures in the above sections, the study will employ non-probability sampling techniques, which is a primarily non-random sampling criterion. The study will employ the convenience sampling technique where the participants will partake in the study based on their availability. The main advantage of the use of the technique is due to its extremely easy, speedy, cost-effective, and convenient nature (Amato et al., 2017). The demerit of the approach is that it limits the chances of participation in the survey for individual customers (Amato et al., 2017). Since the study uses the survey strategy for data collection, the researcher can recruit a large sample. In the previous section, this has been identified as one of the main advantages of using surveys to gather information in a study (Qualtrics, 2020). Thus, to obtain large amounts of data that can be generalised on the larger population, the researcher will recruit 200 participants. The target population will be the customers of 5 banking companies in China who will answer questions designed to obtain information regarding their experiences with artificial intelligence. The study will be seeking to recruit participants within the age groups of between 18 and 45.
The researcher will contact social media admins and request access to the financial institutions' social media (WeChat) platform to interact with the customers. Keen to adhere to the stipulated ethical research frameworks the researcher will briefly contact the customers one at a time (based on customer's convenience) and inform them summarily concerning their intent and enquire if they are willing to engage them further. Upon agreeing, the customer will then be forwarded copies of the informed consent into their inboxes for them to review carefully and provide their consent. The investigator will post the survey on the wenjuanxing.com website, a known survey platform in China. Eventually, the researcher will avail a link to consenting participants which will direct them to the uxspot.com website where they can respond to the questions.
Study limitations refer to the specific research procedures or conditions that ultimately have an adverse influence on the study outcomes. According to Bryman (2016) study limitations manifest in various key elements of the study including methods, data collection, sampling, data use, generalisability, and practice application. To begin with, the researcher will acknowledge their lack of sufficient knowledge in conducting quantitative studies. As outlined throughout the paper, quantitative studies incline mostly towards statistical techniques that demand considerable prior practice and experience. Some of the notably challenging elements of the quantitative method include designing the questionnaire, software analysis of the data, and the interpretation of the results. If done wrongly, may impair the researchers' ability to offer rational recommendations and sustain practical application of the findings. Moreover, the mentioned limitations may have a direct effect on the validity and reliability of the data collection instrument and the analysis process ultimate. Following the researcher's inexperience, the questionnaire design may fail to address/cover all the variables in the study adequately thereby, hampering effective problem assessment. Eventually, the study may not accurately determine the type of relationship that exists between the dependent variables and the independent variables. This may diminish the validity of the quantitative study. Similarly, the reliability of the study instruments will be examined by assessing their capacity to provide similar results consistently. For instance, the use of a large sample in a quantitative methodology may increase the reliability of the study. Other notable limitations include participants having limited knowledge regarding artificial intelligence rendering them incapable of providing appropriate responses for the survey questions. Further, some of the recruited respondents might fail to answer all questions while others may offer informed responses. Additionally, some may find it difficult to use the online survey website thereby making it impossible to fill the online survey altogether.
7. Data Management, Ethics, and Safety
Data management and ethical considerations are integral components of the research process that must be outlined sufficiently in research methodology. During the actual research process, the researcher must ensure to adhere to stipulated ethical requirements and more so, those that pertain to human participant engagement, data handling, and its safety. To begin with, the researcher will ensure to obtain permission from the relevant authorities; the university research department, and the banking companies' management. Similarly, the researcher will need to obtain the participants' informed consent before partaking in the survey. The respondents must be given ample time to read and understand the contents of the form before providing their approval. The study will strive to promote confidentiality in the research process to enable the participants to engage freely without any anxieties or fears of incriminations. The researcher will caution participants against providing/revealing their actual location, names, identification numbers, area of work, and other personal details. In addition, the scholar will remember to inform the participants that participation is voluntary and that they are free to withdraw at any point without prior notice. As well, the researcher must clearly state that the survey and information provided will be used in the present investigation strictly. Important still, the individual will ensure that no one else will have access to the data gathered from the survey. Lastly, all the information recorded from the survey will be stored in a password-enabled pc. The researcher will ensure not to share the password with any other persons. The data management plan will be appended in the proposal appendix to provide further datils concerning the research project.
Alam, S. (2020) ‘Artificial Intelligent Service Quality to Increase Customer Satisfaction and Customer Loyalty (Survey of PT. Telkomsel Customers)’, In First ASEAN Business, Environment, and Technology Symposium (ABEATS 2019) (pp. 100-104). Atlantis Press.
Alharahsheh, H., and Pius, A. (2020) ‘A Review of Key Paradigms: Positivism VS Interpretivism’, Global Academic Journal of Humanities and Social Sciences, 2(3), pp. 39-43.
Amato, P., Brisebois, E., Draghi, M., Duchaine, C., Fröhlich‐Nowoisky, J., Huffman, J., ... and Thibaudon, M. (2017) ‘Sampling Techniques’, Microbiology of aerosols, pp. 23-48.
Apuke, O. D. (2017) ‘Quantitative Research Methods: A Synopsis Approach’, Kuwait Chapter of Arabian Journal of Business and Management Review, 33(5471), pp. 1-8.
Bryman, A. (2016) Social research methods. Oxford university press.
Fulton, B. R. (2018) ‘Organizations and Survey Research: Implementing Response Enhancing Strategies and Conducting Nonresponse Analyses’, Sociological Methods & Research, 47(2), pp. 240-276.
Guetterman, T. C., and Fetters, M. D. (2018) ‘Two Methodological Approaches To The Integration Of Mixed Methods And Case Study Designs: A Systematic Review’, American Behavioral Scientist, 62(7), pp. 900-918.
McKinsey and Company (2020) Reimagining customer engagement for the AI bank of the future. Available at: https://www.mckinsey.com/industries/financial-services/our-insights/reimagining-customer-engagement-for-the-ai-bank-of-the-future# [24th November 2020].
Parise, S., Guinan, P. J., and Kafka, R. (2016) ‘Solving the Crisis of Immediacy: How Digital Technology Can Transform the Customer Experience’, Business Horizons, 59(4), pp. 411-420.
Qualtrics (2020) How to create a great survey. Available at: https://www.qualtrics.com/experience-management/research/survey-basics/ [Accessed 11 December 2020]
Queirós, A., Faria, D., and Almeida, F. (2017)’ Strengths and Limitations of Qualitative And Quantitative Research Methods’, European Journal of Education Studies, 3(9), pp. 369-387.
Ryan, G. (2018). Introduction to positivism, interpretivism and critical theory. Nurse researcher, 25(4), 41-49.
Singh, L., Bode, L., Davis-Kean, P., Berger-Wolf, T., Budak, C., Chi, G., ... and Kreuter, F. (2020). Study Designs for Quantitative Social Science Research Using Social Media. The Future of Quantitative Research in Social Science, pp. 1-27.
Snyder, H. (2019) ‘Literature Review as A Research Methodology: An Overview and Guidelines’, Journal of Business Research, 104, pp. 333-339.
Sokolova, N. G., and Titova, O. V. (2018) ‘Online Surveys in Marketing Management: Opportunities, Advantages and Disadvantages’, Bulletin of Kalashnikov ISTU, 21(2), pp. 90-95.
Sujata, J., Aniket, D., and Mahasingh, M. (2019) ‘Artificial Intelligence Tools for Enhancing Customer Experience’, International Journal of Recent Technology and Engineering (IJRTE), 8(2), pp. 2277-3878
USC Libraries (2020) Research guides: Organizing your social sciences research paper: Limitations of the study. Available at https://libguides.usc.edu/writingguide/limitations#:~:text=Definition,the%20findings%20from%20your%20research [Accessed 11 December 2020]
Westerheide, F. (2020) China – The First Artificial Intelligence Superpower. Available at: https://www.forbes.com/sites/cognitiveworld/2020/01/14/china-artificial-intelligence-superpower/?sh=22db52b2f053 [Accessed 24th November 2020].
Winchester, C. L., and Salji, M. (2016) ‘Writing a literature review’, Journal of Clinical Urology, 9(5), pp. 308-312.
Wu, F., Lu, C., Zhu, M., Chen, H., Zhu, J., Yu, K., ... and Cao, X. (2020) ‘Towards a new generation of artificial intelligence in China’, Nature Machine Intelligence, 2(6), pp. 312-316.
Xueqing, J. (2019) Artificial intelligence adoption gathering momentum in China. Available at: https://global.chinadaily.com.cn/a/201903/06/WS5c7f29b2a3106c65c34ed0b5.html