The Impact of Artificial Intelligence (AI) On Customer Relationship Management

International Journal of Management and Accounting, 5(5), 74-88,
Publisher homepage: www.universepg.com, ISSN: 2707-4641 (Online) & 2707-4633 (Print)
https://doi.org/10.34104/ijma.023.0074090
The Impact of Artificial Intelligence (AI) on Customer Relationship

Management: A Qualitative Study
Jeremy Fei Wang, Ph.D*
Management Information Sytems, Director of the China Program, 74 King Street Saint Augustine, FL 32084, United States.
*Correspondence: [email protected] (Jeremy Fei Wang, Ph.D, Associate Professor of Management Information Sytems,
Director of the China Program, 74 King Street Saint Augustine, FL 32084, United States).
ABSTRACT
Ever since the commercialization of the Internet in the '90s, technology has been evolving faster than ever with
the advent of cloud computing, social media, ubiquitous mobile devices, Internet of things (IoT), blockchain,
and more. A staggering number of three billion internet users, five billion mobile users, and six billion devices
are now connected through this massive global network of networks, facilitating customer information
exchange and interaction never before seen in history. Driven by recent technological advances in computing
power, big data, high-speed internet connection and easier access to models built with advanced algorithms,
Artificial Intelligence (AI) is the next wave of innovation, which has already come into widespread awareness
in the consumer world with the emergence of virtual assistants and chatbots (e.g., Amazon's Alexa, Apple's
Siri, Google's Assistant), image recognition (e.g., Facebook Photos, Google ImageNet), personalized recom-
mendations (e.g., Netflix, Amazon) and autonomous driving (e.g., Tesla, Google Waymo). This qualitative
research study intends to learn about the impact of AI on customer relationship management (CRM),
specifically in the area of customer service of problem resolution. Most prior research focuses on the AI
technologies leveraged in CRM systems, such as machine learning, natural language processing, voice
recognition, chatbots, data analytics, and cloud infrastructure. Few extant studies have used a qualitative
research methodology to gather data from industry experts to truly understand the impact of AI technologies on
customer relationship management, especially in the area of customer service and problem resolution. This
study aims to fill this research gap. This research contributes to the literature on AI in the context of CRM and
is of value to both academics and practitioners as it provides a detailed analysis and documentation of the
impact of AI on the customer service domain.
Keywords: Artificial intelligence (AI), Internet of things (IoT), and Customer relationship management (CRM).
INTRODUCTION: proper protocols on the relevant informants who have

The objective of this qualitative research study is to extensive real-world industry experiences with AI-
learn about the impact of AI on customer relationship driven CRMs. Since the adoption of AI in the CRM
management (CRM), specifically in the area of custo- systems is a relatively new phenomenon, the qualita-
mer service of problem resolution. The focus of this tive research methodology will enable us to have an in-
qualitative study is to conduct depth interviews using depth understanding of its behavior and impact and
UniversePG I www.universepg.com 74
Wang JF / International Journal of Management and Accounting, 5(5), 74-88, 2023
also better interact with the interviewees through an series of hard-selling or simple customer service to soft
emergent design. In addition, the qualitative research skills such as the relationship building and emotional
provides an opportunity for us to examine existing connectivity (Singh et al., 2019). Therefore, this study
theoretical and conceptual foundations in the AI-driven also aims to learn the impact of AI on customer service
CRM context. The central questions of the study are job loss and change of job roles.
designed to analyze the general, industry, and firm- 1) What is the firm-specific or the industry-level
specific impact of AI on customer service and problem impact of AI on service job loss and change of
resolution job roles of customer service personnel?
2) How would AI change the job roles and required
1) What are the most remarkable features of AI
skills of customer service professionals?
used in the area of customer service and problem
resolution? Conceptual Background
2) How did these AI features come about and what General Literature Review of AI in the CRM Space
factors drove the implementation of these AI Recent technological advances in Artificial Intelligence
features? (AI) technologies, especially in the fields of machine
3) What were the key factors that made the AI fea- learning, deep learning, neural networks and big data
tures impactful in the area of customer service (Moreno and Redondo, 2016; Zhang et al., 2018),
and problem resolution? What were the key ubiquitous mobile computing (He et al., 2019) have
barriers that had to be overcome? fueled the growth of the next-generation digital plat-
4) What is the nature and pace of the change in forms (Khalid et al., 2019; Rai et al., 2019; Zhang et
customer service and problem resolution that al., 2019), which have progressively achieved human
these AI-driven CRMs have produced? Will this (sometimes super-human) level of performance in the
pace of change slow down, remain the same, or various areas including autonomous driving, medical
accelerate over the next decade? diagnosis (e.g., cancer screening), the robots/ drones,
5) What role has the specific respondent's firm chatbots, virtual assistants, the language translation,
played in the area and how has it assisted other the governance monitoring (e.g., copycats, content
firms? What key accomplishments and challen- violation), complex game playing and recommendation
ges have been encountered? systems. AI features embedded in customer relation-
6) What are the specific steps or programs the firm ship management (CRM) platforms also create new
has taken (or helped other firms) to support possibilities for customer experience, with rich insights
customer service and problem resolution using into customer needs (Kumar Deb et al., 2018). These
AI technologies? innovations have been driven by a manifold increase in
processing power, lower-cost hardware, and the
Also, the main threats posed by AI are job losses as a
exploding creation and availability of customer data
result (Siau & Wang, 2018; Siau & Yang, 2017) and
(Gantz et al., 2017).
the skills of these professionals may also shift from a
Fig. 1: The Four-Part Method of Inquiry for Ethnography (Chakravarti & Crabbe, 2019).
The qualitative depth interview employed by this study problems and the provide recommendations. These
is based on McCracken, (1988) four-phase implement- technologies have produced effective applications for
tation framework, which calls for a careful review of customer service & problem resolution, including Next
“analytical categories and relationships” and “cultural Best Action (NBA) systems based on the predictive
categories and relationships”. This framework (Fig. 1) analytics, chatbots, virtual agents and agent assist sys-
has been adapted by (Chakravarti & Crabbe, 2019) to tems. The impacts of these AI technologies on the
fit a variety of ethnographic methods. customer service & problem resolution are multifold,
including operational excellence, customer intimacy,
Analytic Categories and Concepts
service productivity & quality, competitive advantage,
The analytic categories establish the domain of the
service excellence, customer satisfaction, & ultimately,
investigation & organize the knowledge in the domain.
return on investment (ROI). These analytic categories
In the specific domain of AI-driven CRM systems, a
are identified in the formal inter-view protocol as the
plethora of technologies, including machine learning
elaboration probes and are used during the inquiry.
(ML), deep learning (DL) or neural network, natural
The relationships of these analytic categories are the
language processing (NLP), voice speech/ recognition,
depicted in the following figure and are the further
image recognition, and other tools are used to discover
explored in the depth interview.
insights, identify patterns, answer inquiries, the solve
Fig. 2: Relationships of Analytic Categories.
Table 1 explains the analytic categories and concepts key technical terms appeared frequently in the answers
of AI technologies most relevant to the customer from various informants during the interviews.
service and problem resolution context. Some of these
Table 1: Analytic Categories and Concepts.

AI Analytic Category Definition CRM Context
Artificial Intelligence AI is the concept of having AI can be further classified into Weak and Strong AI. Weak
(AI) machines “think like humans” & AI, also known as narrow artificial intelligence, focuses on
perform tasks. such as reasoning, specific narrow tasks such as customer service and problem
planning, learning, and resolution. The strong AI, or general AI, where machines
understanding language acquire cognitive and self-aware capabilities similar to
(Salesforce.com, 2017). human intelligence, will take years to appear in CRM
systems. The term AI is widely used in the CRM space. AI-
driven CRM refers to CRM systems with various levels of
AI capabilities for handling specific tasks.
Machine Learning (ML) ML is the core driver of AI and All AI-driven CRMs are powered by machine learning,
involves computers learning from which analyzes and learns from massive amounts of
data with minimal programming. customer data.
ML programs the desired outcome
and feeds data to train the machine
to achieve the outcome on its own
(Salesforce.com, 2017).
Deep Learning (DL) and DL is a subset of machine learning The reason for the recent surge in the use of AI in CRM is
Neural Network and uses complex algorithms that mainly due to relatively recent advances in deep learning,
mimic the brain’s neural network to which creates outputs in much more sophisticated ways
learn a domain with little or no using multiple layers of interconnected neurons (called
human supervision (Salesforce.com, Neural Networks).

2017).
Natural Language NLP uses machine learning NLP is widely used in customer service applications such as
Processing (NLP) techniques to find patterns within chatbots and virtual assistants. It is also capable of doing
large data sets in order to recognize sentiment analysis to understand how customers feel about
natural language (Salesforce.com, products or services.
2017).
Voice Recognition (VR) VR, alternatively referred to as Voice recognition has gained prominence and use with the
speech recognition (SR), is the advent of intelligent personal assistants, such as Apple's
ability of a machine or program to Siri, Amazon's Alexa, and Microsoft's Cortana. It has been
receive and interpret dictation or to increasingly used in customer service applications to
understand and carry out spoken address customer needs.
commands (Rudrapal et al., 2012).
Image Recognition (IR) IR is a term for computer IR, or computer vision, is a subfield of machine learning,
technologies that can recognize which has already been extensively used in autonomous
certain people, animals, objects, or driving, security, healthcare, social media, and visual
other targeted subjects through the search. It has a lot of potential in the customer service area.
use of algorithms and machine Salesforce has launched Einstein Vision with a set of
learning concepts (Techopedia, application program interfaces (APIs) designed to help
2019). companies build customer relationship management apps
with image recognition features.
Big Data Big Data is the raw fuel of AI, The recent explosion of high volume, high velocity, and
which includes large amounts of high variety of customer data provide the ideal inputs for
structured or unstructured CRM machine learning models, which make these systems
information that provide the inputs smarter and smarter.
for surfacing patterns and making
predictions (Salesforce.com, 2017).
Cultural Categories and Concepts 2) CRM vendors produce the AI-driven applications
The cultural categories call for a detailed inventory of utilizing AI leaders’ frameworks or the research
the key features of the researcher’s experience with the work, e.g., Salesforce, NetSuite, and SAP.
focal topic (Chakravarti & Crabbe, 2019, p. 75). This 3) CRM implementers are the usually IT consulting
also allows recognition and admittance of matches and firms partnered with multiple CRM vendors, e.g.,
mismatches with other cultural categories which may Accenture, Deloitte & Touche, and Perficient.
merge from the respondent’s subjective experience 4) CRM Facilitators are usually technology evange-
(McCracken, 1988). As a former practitioner in the lists from AI leaders, CRM vendors, & imple-
management and IT consulting industry, my personal menters, who coach and train customers and
experience tells us that it is crucially important to advocate the adoption of AI-driven CRMs, e.g.,
assess the cultural categories from the industry and Microsoft IT evangelists, IBM Watson consul-
functional area perspectives. In the domain of AI- tants, Salesforce Einstein Analytics & Discovery
driven CRMs, the following are the cultural categories consultants, and Perficient AI consultants.
identified based on the major stakeholders in the CRM 5) CRM Followers adopt the AI frameworks or
ecosystem (the functional cultures) for this specific applications from the leaders and vendors once
study. The depth interview questions and probes are the tools are well established. Both CRM imple-
designed to analyze the relationships among these menters & customers can fall under this category.
entities. 6) CRM customers/clients are companies of various
1) AI leaders conduct fundamental AI research and sizes that use CRM systems to serve their own
development, e.g., Google, Microsoft, Amazon, users.
and IBM.
Qualitative Study Design remains flexible both before and throughout the actual
Available Methods research (Marshall & Rossman, 2014). It encompasses
Compared with other research methods in which res- a variety of accepted methods and structures, including
earchers’ hypotheses and the procedures are predeter- four major types which are the most commonly used.
mined, the research design in qualitative research They are summarized in Table 2.
Table 2: Qualitative Study Design Methods, Adapted from Astalin, (2013, pp. 120-122).
Method Description
Phenomenology This refers to a study of phenomena (e.g., events, situations, experiences, or concepts), which
describes something that exists as an integral part of the world in which we are living.
Ethnography This refers to a methodology for descriptive studies of cultures and peoples, which requires fieldwork
by the investigator using data collection techniques including:
Participant-observation
Interactive depth interviews
Interactive group interviews
Life histories/autobiographies
Other (unobtrusive) observational methods (e.g., garbology)
Other elicitation methods (nonverbal)
Grounded This refers to a type of qualitative research methodology that allows theory/theories to emerge from
Theory the data that is collected. It employs a systematic yet flexible process to collect data, code the data,
make connections, and see what theory/theories are generated or are built from the data.
Case Study This refers to a research method, which is used to describe an entity that forms a single unit such as a
person, an organization, or an institution.
Chosen Method and topics that need to be covered during the

This qualitative study aims to the conduct personal conversation, usually in a particular order.
interviews with four industry experts to gain insights 3) The interviewer follows the protocol but is able
into the impact of AI technologies on customer service to follow topical trajectories in the conversation
and the problem resolution. Therefore, the proposed that may stray from the guide when he or she
research method for this study is an interactive depth feels this is appropriate.
interview which is a commonly used ethnographic data
collection technique. Moreover, following Brinkmann, Interview protocol
(2016) interview guidelines, this study has chosen the Preliminary Draft
semi-structured interview method among the spectrum The preliminary interview protocol was the developed
of different types of the interviews (i.e., structured, based on the central research questions of this study, a
unstructured, & semi-structured). The semi-structured careful literature review of the use of AI in the space
interview is defined as “an interview with the purpose of customer service and problem resolution, and an
of descriptions of life world of the interviewee in order assessment of the professional backgrounds of the res-
to interpret the meaning of described phenomena” pondents so that the scope of topics are appropriate for
(Brinkmann & Kvale, 2015, p. 6). Compared with the the interactive depth interview. In addition, elaboration
structured and unstructured interviews, the semi- probes are developed to the facilitate data collection
structured interview has the following characteristics according to the proposed analytic and the cultural
(Cohen & Crabtree, 2006), which fit the requirements categories in the section 3. The preliminary inter-view
of this study well. protocol was initially tested with a senior product
manager with the Coupa Software and responsible for
1) The interviewer and respondents engage in a delivering a roadmap for the search and shopping
formal interview. module within the larger Coupa platform suite. Based
2) The interviewer designs and uses a formal on the initial interview observations and the expert
interview protocol, which is a list of questions feedback, we made the following adjustments to the
preliminary interview protocol. this interview protocol. First, the open-ended questions
are difficult to analyze due to the varied responses
1) For Q1, the time frame was specified to be “the
from respondents. Second, the flexibility of using the
last 10 years” of the AI implementation in CRMs
elaboration probes in the interview may reduce reli-
to avoid any confusion about earlier AI features
ability, as respondents may not receive the same probe
used in legacy systems; the scope of implement-
questions and it would be hard to compare answers.
tation was specified to be the general application
Third, in order to get enough data from respondents
space for various businesses to avoid potential
during the one-hour-long interview, the questions are
confusionwithfirm-specific implementation.
designed to include subparts, which makes it cumber-
2) For Q2, the question was made more concise by
some for respondents to remember to answer all parts
removing “to the best of your knowledge", which
of a specific question.
was a given condition.
3) For Q5, the question was modified to distinguish Informant Selection Process/Criteria
between the firm-specific versus industry-level In order to obtain reliable and useful data for this
impacts on jobs and change of job roles. qualitative study, the informants of this study are the
4) The time checks were also adjusted accordingly selected based on the following criteria.
based on the initial test observations.
1) Who has relevant and useful information? Infor-
Final Interview Protocol mants should have extensive experience and
The final interview protocol (Appendix A) is com- knowledge in the AI-driven CRM space.
prised of six questions in a particular order to facilitate 2) Who is interested and available? Informants
the flow of the interview and also standardize the data should be interested in the research topic and be
to be collected from the respondents. Elaboration available for a one-hour-long interview.
probes are developed in case topical trajectories in the 3) Who is willing to provide reliable information?
conversation stray from the inter-view guide. Standard Informants should have a good professional
introduction and closing statements are also used to reputation and be willing to share their honest
ensure the formality and consistency of the interview. answers to the interview questions.
Following Brinkmann’s, (2016) guidelines, the final
Based on these three key criteria, we selected four
interview questions are designed with a purpose and
industry professionals for the final interview from my
refined to invite respondents to give descriptions of
own professional network, which represents a good
their experiences of and insights (lifeworld) into the
mix of professional levels (2 executive levels and 2
AI-driven CRM space and seek to obtain data for
middle management levels), cultural domains (3 from
interpretation of meaning. The key strength of this
the U.S. and 1 from China), functional sectors (3 from
formal interview protocol is that it has adhered to the
consulting firms, 1 from an e-commerce company). All
guidelines and best practices of the semi-structured
four respondents have agreed to share their names in
interview. This allows the interviewer to be prepared
the study.
and appear competent during the interview while
giving the informants the freedom to express their Informant Description
views on their own terms (Cohen & Crabtree, 2006). Respondent 1
With a standardized set of questions, it can provide Respondent 1 is the Chief Strategist of the Customer
reliable, consistent, and comparable qualitative data Experience Platforms at Perficient, Inc. He has more
from respondents across different functional domains than 35 years of the strategic technology advising
and various professional levels. In the meantime, the experience. He and his team build great customer,
interactive elaboration probes offer flexibility in partner, and employee experiences. His primary focus
discovering analytic and cultural categories and an is on the actual technology used in a customer ex-
opportunity for identifying new ways of seeing and perience environment and takes into account his
understanding the topic at hand. Besides the strengths experience with digital platforms such as web content
mentioned above, there are also several weaknesses of management, portals, search systems, the marketing
platforms, and the marketing systems. Looking to the action software across their regional call centers.
future, he and his team are expanding their focus to Before joining Accenture, He was a senior consultant
understand how emerging technologies, such as the at Capgemini, where he focused on the digital trans-
artificial intelligence (AI), relate to those customer formations using technologies such as NetSuite and
experience platforms. It’s quite a broad range of the Salesforce. He later shifted to talent and organizational
technology, but the common thread is they all connect effectiveness work with a heavy emphasis on mergers
with the customer. His primary role is to help his and integrations, talent learning and development, and
clients understand the technology that they need to change management. He has a B.S. in the Information
have in place, why it’s important, and how it integrates Systems from Marietta College (Adapted from respon-
and affects all parts of their business. He has a B.S. in dent-supplied information, 2019).
Computer Science from Purdue University and an
Respondent 4
M.B.A. from the Cleveland State University (Adapted
Respondent 4 is a data analyst from the Chief Cus-
from respondent’s Perficient profile and respondent-
tomer Office (CCO) department at the Alibaba’s head-
supplied information, 2019).
quarters in Hangzhou, China. Alibaba is a Chinese
Respondent 2 multinational conglomerate holding company special-
Respondent 2 is the practice leader and Chief Strategist izing in e-commerce, retail, Internet, and technology.
for artificial intelligence (AI) at Perficient, Inc. She His current role is focused on improving the customer
brings a background in the analytics and unstructured experience for more than 20 million users of the Ali
information management to design and deliver trans- Express, an online retail service owned by Alibaba.
formative AI solutions. Her engineering background Currently, 70% of employees working at the CCO are
aids in her analysis of complex business problems and computer engineers. He and his team embed artificial
her ability to the develop innovative cognitive appli- intelligence into customer services to the serve both
cations. She and her team help clients uncover hidden buyers and sellers. This system has helped to reduce
insights, the create differentiated experiences, identify more than 100 thousand customer agents for the plat-
trends and anomalies, enhance existing applications, form. He has a B.S. in the Information Systems from
and scale their knowledge in AI. She created an award- Marietta College and an M.S. in Management Inform-
winning center of excellence for AI and participated in ation Systems from Texas Tech University (Adapted
a panel with IBM CEO Ginni Rometty to share her from respondent-supplied information, 2019).
knowledge on the AI solutions and strategy. She also
frequently demonstrates thought leadership through Nature of Material Collected
speaking engagements at various industry and techno- The materials collected from respondents are mostly
logy conferences, such as the AI Summit. She has two non-numerical, the descriptive data and therefore, not
engineering degrees, including a B.S. in Engineering appropriate for quantitative analysis using advanced
Physics from the Taylor University and an M.S. in statistical tools. Most of the descriptive data are fairly
Mechanical Engineering from the Columbia University current, with only a small number of the references to
(Adapted from respondent’s Perficient profile, 2019). historical contexts. None of the materials collected are
deemed classified, private use only, or confidential.
Respondent 3 Most of the materials collected from the respondents
Respondent 3 is a management consultant within the are the experience-based while some of them are
Talent & Organization practice at Accenture, servicing knowledge-based derived from other industry sources
clients within the Utilities & Resources industry. His (e.g., Gartner, CIO Daily). Some materials are busi-
current role at Accenture is Analytics and Business ness-oriented in the nature, especially during the
Readiness Lead, responsible for the monitoring and discussions on the demand-side of the AI features.
reporting on the progress/state of the overall transfor- Some materials, on the other hand, are technical in
mation using various data analytics and the metrics nature with many acronyms used. These technical
tracking tools. His current project is at a leading US acronyms are manually noted in the full interview
utility company implementing a new customer inter- transcripts (Appendix B).
RESULTS:
After reading and rereading data collected from the
General Interpretative Approach
respondents, we kept track of the hunches, the inter-
Qualitative data analysis entails certain distinct active-
pretations, and ideas by using the Microsoft OneNote,
ties. The first and most important one is the ongoing
which allowed us to write down ideas anytime and
discovery, which is about identifying themes and the
anywhere. We looked for broad themes and the major
developing concepts and propositions (Taylor et al.,
topics that emerged from the interview transcripts and
2015). Following Brinkmann, (2016) guidelines on the
developed a coding scheme (Fig. 4) to analyze the
semi-structured interviews, this study has adopted a
data. One key observation that has emerged from the
general interpretative approach and conducted a broach
interviews is the dynamic relationship between the
thematic analysis on the interview transcripts. Taylor
demand-side (CRM system customers/clients) and the
et al. (2015) recommend identifying themes by the
supply-side (technology providers) of the implement-
thoroughly exploring the data & propose the following
ation of AI features. Based on the coding template
steps to maintain focus.
suggested by (Taylor et al., 2015, p. 184), I modeled
my coding scheme to the highlight the cultural and
analytic categories that frequently appeared in the
respondents’ accounts of their experiences (Fig. 4).
The coding scheme is applied using a red font to some
sections in the transcripts (Appendix B) to demonstrate
the process.
Fig. 3: Developing Themes and Concepts (Taylor et

al., 2015).
Fig. 4: Analysis: Coding Scheme for the AI-Driven CRM Space.
Emergent Themes AI features in CRM systems. Another emergent theme

Cultural Categories from the supply side is that CRM implementers (e.g.,
The cultural categories emerged when all four (4) Accenture, Deloitte and Touche, Perficient), usually
respondents were talking about the key stakeholders in partner with multiple CRM vendors and advise and
the AI-driven CRM space, especially the AI leaders, facilitate the adoption of these AI-driven the CRM
CRM vendors, CRM implementers, & CRM customers/ systems based on the assessment of customers' needs.
clients. One emergent theme is that CRM customers/ The availability of advanced AI technologies and the
clients want AI features in the CRM system in order to promotion and adoption campaigns by CRM vendors
achieve their strategic goals, such as the reducing and implementers are key supply-side factors that
operational costs and improving customer satisfaction. drive the implementation of the AI features in CRM
These are key demand-side factors that drive the systems. All four respondents predict that due to the
implementation of AI features in CRM systems. From dynamic growth of this ecosystem, the pace of change
the supply-side perspective, one emergent theme is that in customer service produced by AI-driven CRMs will
CRM vendors (e.g., Salesforce, NetSuite, SAP) are the accelerate over the next few years. However, she
leveraging the tools and platforms from AI leaders predicts that the acceleration will not continue all the
(e.g., Google, Microsoft, Amazon, IBM) that conduct way through a decade, though, because 80% of the
fundamental research & development in AI techno- adoption will probably come in the next three to five
logies. These tools are increasingly accessible through years due to the current high demand for AI features in
PaaS (platform as a service). The CRM vendors either CRM systems. Following Taylor et al. (2015) guide-
embed the AI capabilities in their own applications or lines, I constructed the following analytic diagram
integrate with the AI leaders’ platforms (e.g., IBM (Fig. 5) to illustrate the dynamics of the key cultural
Watson). According to respondent 1, the supply side is categories.
playing a predominant role in pushing the adoption of
CRM
Vendors
CRM
Customers
AI CRM
Leaders Implemen-
ters
Fig. 5: Analytical Diagram: Key Cultural Categories in AI-Driven CRM Space.
Emergent Themes solve problems, and provide recommendations. These

Analytic Categories analytic categories were used as elaboration probes in
The analytic categories establish the domain of the the formal interview protocols. The respondents have
investigation and the organize the knowledge in the converged on the several AI technologies which are
domain. In the specific domain of AI-driven CRM frequently mentioned in their responses, including the
systems, a plethora of AI technologies are identified artificial intelligence (AI), machine learning (ML),
and defined in Fig. 2, which are used in CRM systems natural language processing (NLP), voice recognition
to discover insights, identify patterns, answer inquiries, (VR), image recognition (IR), and big data. Several
common emergent themes include 2) Understanding each customer individually and

providing personalized user experience to the
1) Data has fueled the growth of the AI adoptions
achieve customer intimacy;
in the customer service space in recent years.
3) Offering fast & accurate answers and efficiently
Machine learning (ML) is the engine that powers
resolving problems;
these AI features.
4) Achieving higher customer satisfaction and the
2) Voice recognition (VR) and natural language
obtaining competitive advantages over compete-
processing (NLP) are among the top AI features
tors;
adopted in the CRM systems. Virtual agents,
5) Ultimately realizing a satisfactory return on the
chatbots, and agent assist systems are the predo-
investment (ROI) and maximizing profits;
minant use cases in this space. The major
6) Reducing the number of the low-level customer
challenge understands user intent & sentiment.
service personnel and transitioning them to more
3) Predictive data analytics are on the rise and Next
value-added roles, e.g., virtual agent managers
Best Action (NBA) systems are demanded by
and trainers. The impact is more about the role
CRM customers to utilize the customer data
changes, rather than simple job loss.
better.
4) Image recognition (IR) or visual recognition has Integrated Interpretation
seen new use cases in the customer service The descriptive data collected from respondents
space. support the relationships of the cultural and analytic
categories proposed in earlier sections. That is, a
The respondents have also converged on several
plethora of AI technologies are becoming increasingly
themes regarding the impact of AI on customer service
accessible through AI leaders, and are leveraged by
and problem resolution, which include
CRM vendors and implementers to embed or integrate
1) Dramatically reducing the number of service into CRM systems as useful AI features, which in turn
personnel answering simple & routine customer impact customer service and problem resolution.
inquiries;
Fig. 6: Integrated Cultural & Analytic Categories in AI-Driven CRM Space.
DISCUSSION: key findings based on the qualitative analysis of the

Summary of Findings data collected through semi-structured interviews with
This qualitative research study intends to learn about four respondents who have the extensive industry
the impact of the AI on customer relationship man- experiences in this space.
agement (CRM), specifically in the area of customer 1) Overall, AI adoption by businesses for the various
service of problem resolution. The following are the purposes has continued to the gain momentum in
recent years. With a focus on the customer ser- size of this study is very small and the informant
vice and problem resolution domain, the pace of selection process is not random, thereby limiting the
adoption will accelerate in the next few years. generalizability of the findings. Second, due to the
2) The most impactful AI features used in CRM project time constraints, the four informants were
systems include 1) virtual agents, chatbots and selected from our own professional network. Further
agent assist systems, which rely on voice/ speech research could explore strategies to minimize biases
recognition and natural language processing linked to informant selection, ensuring a more
(NLP); 2) Next Best Action (NBA) systems, objective and comprehensive analysis. Two informants
which leverage predictive data analytics; and 3) are from the same company and three informants are
personalization technology, which is powered by from the IT consulting industry, thereby increasing the
advanced machine learning (ML) and the big chance of biased responses. A longer vetting process
customer data. The key factors that made these would help in this aspect. Moreover, for conducting an
AI features impactful in the area of customer in-depth interview, the researcher’s personal inter-
service and problem resolution include 1) cost viewing skills (e.g., proper use of elaboration probes)
savings from reducing the number of low-level could affect the quality of data collected. The test
service personnel; 2) faster and more accurate interview helped us immensely before conducting the
answers to customer inquiries; 3) personalized actual interviews. Lastly, the flexibility of using
customer experience; and 4) overall improved elaboration probes in the interview may reduce
quality of customer service & customer satis- reliability, as the respondents did not receive the same
faction. probe questions, which made it difficult to compare
3) The factors that drive the implementation of answers. Future research could increase the sample
these AI features are mainly two folds – 1) from size and explore strategies to minimize biases linked to
the demand-side perspective, an increasing the informant selection, ensuring a more objective &
number of companies want AI features in their comprehensive analysis. In addition, further research
CRM systems to the obtain both tangible and could delve deeper into the transition of job roles
intangible benefits and achieve their strategic resulting from AI implementation. Analyze how
goals; 2) from the supply-side perspective, AI employees are adapting to new roles such as virtual
leaders, CRM vendors, implementers, especially, agent managers and trainers, and assess the factors
are making the AI technologies and AI-driven influencing a smooth transition.
CRMs more accessible to companies of various
sizes. The key stake holders of AI-driven CRMs CONCLUSION:
have formed an ecosystem, which is pushing the In conclusion, this qualitative study sheds light on the
accelerated adoption of AI features in this space. transformative impact of AI on CRM, particularly in
4) AI has been replacing the jobs of low-level the domain of customer service & problem resolution.
customer service personnel and the some of the The findings underscore the accelerating pace of AI
employees have transitioned to the more value- adoption in businesses, with AI features like virtual
added roles, e.g., virtual agent managers and agents, Next Best Action systems, and personalized
trainers. The impact is more about changing job technology leading the charge. These technologies,
roles, rather than simple job loss. primarily driven by voice & natural language
5) All respondents stated that their firms have processing, predictive data analytics, & machine
formal programs to embrace new technologies learning, are significantly enhancing customer service
(or help other firms), including AI. There is also by enabling cost savings, faster and more accurate
a formal process to evaluate the success of such responses to inquiries, personalized experiences, and
programs. heightened customer satisfaction. Both demand-side
factors, where companies seek AI integration to attain
Limitations & Future Research strategic objectives, & supply-side dynamics, wherein
This study has several limitations. First, the sample AI leaders and CRM vendors facilitate accessibility
and integration of AI technologies, contribute to this AI on customer relationship man-agement. IDC:

rapid AI adoption. Job roles in customer service are Analyze the future.
evolving, with AI replacing low-level service person- 7) He, J., Liu, H., & Li, X. (2019). Mobile app
nel and necessitating a shift towards higher-value recommendation: An involvement-enhanced ap-
roles. While this study's small sample size and the proach. MIS Quarterly, 43(3), 827-849.
potential biases due to informant selection underscore https://doi.org/10.25300/MISQ/2019/15049
its limitations, it serves as a foundational exploration 8) Hossain MM, Faruq MO, and Sharmin K.
of AI’s impact in the CRM landscape. Future research (2022). Measuring the impact of corporate social
should address these limitations, aiming for a broader responsibility on customer satisfaction: insight
and more objective analysis to the further our under- from Cumilla city areas. Can. J. Bus. Inf. Stud.,
standing of this evolving paradigm. 4(1), 12-23.
https://doi.org/10.34104/cjbis.022.012023
ACKNOWLEDGMENT: 9) Khalid, O., Khan, S. U., and Zomaya, A. Y.
I express my heartfelt gratitude to the reviewers and (2019). Big Data Recommender Systems: Vol. 2
editors for their insightful feedback and guidance, Application Paradigms. The Institute of Engin-
which significantly contributed to the refinement of eering and Technology.
this manuscript. Additionally, I extend our appreci- 10) Kumar Deb, S., Jain, R., and Deb, V. (2018).
ation to all individuals and institutions that supported Artificial intelligence creating automated in-
and facilitated this research. Your contributions were sights for customer relationship management.
invaluable in bringing this work to fruition. 11) Marshall, C., & Rossman, G. B. (2014). Desig-
CONFLICTS OF INTEREST: ning Qualitative Research. Sage publications.
I hereby declare that I have no conflicts of interest, 12) McCracken, G. (1988). The Long Interview.
financial or otherwise, that could have influenced or Sage publications.
biased the work presented in this manuscript. 13) Mezgebe A. (2020). Effect of electronic mar-
keting on customer satisfaction: the case of four
REFERENCES: star hotels in Addis Ababa, Int. J. Manag.
1) Astalin, P. K. (2013). Qualitative research de- Account., 2(4), 74-95.
signs: A conceptual framework. Inter j. of social https://doi.org/10.34104/ijma.020.074095
science & interdisciplinary research, 2(1), 118- 14) Moreno, A., and Redondo, T. (2016). Text ana-
124. lytics: The convergence of big data and artificial
2) Brinkmann S. (2016). The Interview: The SAGE intelligence. Inter. J. of Interactive Multimedia
Handbook of Qualitative Research (1st ed.). and Artificial Intelligence, 3(Special Issue on
3) Brinkmann, S., & Kvale, S. (2015). Inter-views: Big Data and AI), 57-64.
Learning the Craft of Qualitative Research Inter- https://doi.org/10.9781/ijimai.2016.369
viewing. Sage publications. 15) Rai, A., Constantinides, P., & Sarker, S. (2019).
4) Chakravarti, D., and Crabbe, R. (2019). Quali- Editor's comments: next-generation digital plat-
tative Research for Consumer Psychology. In forms: toward human - AI hybrids. MIS Quar-
Handbook of research methods in consumer psy- terly, 43(1), iii–x.
chology, (pp. 61-92). Routledge/Taylor and https://dl.acm.org/doi/10.5555/3370135.3370136
Francis Group. 16) Rudrapal, D., Das, S., & Debbarma, N. (2012).
https://doi.org/10.4324/9781351137713-4 Voice recognition and authentication as a pro-
5) Cohen, D., & Crabtree, B. (2006). Qualitative ficient biometric tool and its application in
research guidelines project. Robert Wood John- online exam for PH people. Inter J. of Computer
son Foundation. Retrieved July 2019, from - Applications, 39(12), 6-12.
http://www.qualres.org/HomeSemi-3629.html https://doi.org/10.5120/4870-7297
6) Gantz, J. F., Vesset, D., & Wardley, M. (2017). 17) Salesforce.com. (2017). AI for CRM: A Field
A trillion-Dollar boost: The economic impact of Guide to Everything You Need to Know. Re-
trieved August 2019, from - take notes during the interview, it will be very helpful
https://a.sfdcstatic.com/content/dam/www/ocms/as if I have your permission to video record this conver-
sets/pdf/misc/ai-for-crm.pdf sation so that I accurately capture your opinions and
18) Siau, K., & Wang, W. (2018). Building trust in input. To begin the interview, I would like to know a
artificial intelligence, machine learning, and ro- little bit more about you. Please take 3 minutes to talk
botics. Cutter business technology journal, 31 about your background and current role in the organiz-
(2), 47-53. ation, including how long you have worked here & the
19) Siau, K. L., and Yang, Y. (2017). Impact of nature of your current professional responsibilities.
artificial intelligence, robotics, and machine lear-
ning on sales and marketing. Midwest Asso- Time check: 3 minutes
ciation for Information Systems Conference INTERVIEW QUESTIONS
(MWAIS 2017), Springfield, Illinois. Q1. Think about the implementation of artificial
http://aisel.aisnet.org/mwais2017/48
intelligence (AI) features in CRM systems by
20) Singh, J., Flaherty, and K., Onyemah, V. (2019).
businesses in the last ten years. What comes to
Sales profession and professionals in the age of
mind as among the most remarkable features of AI
digitization & artificial intelligence technologies:
used in the area of the customer service and
concepts, priorities, & questions. J. of Personal
problem resolution? What makes these AI features
Selling & Sales Management, 39(1), 2-22.
stand out in your mind?
https://doi.org/10.1080/08853134.2018.1557525
21) Taylor, S. J., Bogdan, R., & DeVault, M. (2015). Elaboration Probes
Introduction to Qualitative Research Methods: A 1) At least three AI features implemented in the
Guidebook and Resource. John Wiley & Sons. customer service and problem resolution (e.g.,
22) Zhang, S., Sun, A., and Tay, Y. (2018). Deep chatbots)
Learning Based Recommender system: A Sur- 2) Focus on the impact on the customer service and
vey and New Perspectives. ACM Computing problem resolution
Surveys, 1, 1-35. 3) Nature of impact
https://doi.org/10.1145/3285029 4) Note terms that the respondent uses to describe
23) Zhang, S., Yao, L., and Tay, Y. (2019). Deep the scale and locus of impact.
learning based recommender system: A survey 5) If needed, help respondents using an unrelated the
and new perspectives. ACM Computing Surveys, AI example (e.g., the AI-driven fraud detection
52(1), 1-38, Article 5. systems).
https://doi.org/10.1145/3285029
Time check: 8 minutes
APPENDICES
Q2. How did these AI features come about and
Appendix A. Interview Protocol
what factors drove the implementation of these AI
Thank you for agreeing to participate in this interview.
features?
This research aims to the explore and understand the
impact of artificial intelligence (AI) on the customer Elaboration Probes
relationship management (CRM), specifically in the 1) The three AI features identified in Q1.
area of customer service and problem resolution. I am 2) Origins: Needs for operational excellence and
currently interviewing a small number of experts like customer intimacy
you, so that I can draw on your experience to help us New development in AI technologies & big data
understand the nature of the AI impact in this space - 3) Implementation Drivers:
how it happens and what determines its success and
impact on customer service and problem resolution. I AI technology development (e.g., machine learning,
am very interested in your views and opinions on this deep learning, the natural language processing, neural
topic as they reflect your experiences. Although I will networks)
Industry actions and trends 3) Explore the basis on which the predictions are
Firm actions on the supply-side (e.g., the R&D, oper- made regarding both the nature and pace of the
ations, excellence) change
Firm actions on the demand-side (e.g., customer satis- 4) Explore factors related to change in pace
faction, service excellence) 5) Explore the potential relationship between the
Other supply and demand-side factors
Time check: 8 minutes Time check: 10 minutes
Q3. What, in your opinion, were the key factors Q5. Given the nature, pace and impact of AI
that made the AI features impactful in the area of technologies used in customer service and problem
customer service and problem resolution? What do resolution, describe the role that your firm has
you think were the key barriers that had to be played (or help other firms) in the area. What have
overcome? been the key accomplishments and challenges?
Have you witnessed any firm-specific or industry-
Elaboration Probes level impact on service job loss and change of job
1) AI Technology, BIG data, personalization, data roles of customer service personnel?
privacy, regulation, development costs
Elaboration Probes
2) Market level Benefits/costs (e.g., customer
insights, incremental productivity, work without 1) Note terminology used: (e.g., Leader, Follower,
breaks, Challenger, Disrupter, Implementer, Facilitator)
1. content curation & recommendation; appli- 2) Note implicit criteria Demand side, Supply side
cation dev and maintenance costs) 3) Specific key accomplishments: Demand side, the
2. Compatibility Supply side
3. Competitive advantages 4) Specific key challenges: Demand side, Supply
4. Agility side
5. Other 5) Specific challenges (product, process, technology)
3) Market access Product versions 6) Specific impact on job loss and roles if any
1. Service infrastructure Time check: 10 minutes
2. Customer communication (traditional and new
media) Q6. Thinking about the time between the year 2000
3. Service and distribution logistics and now, describe any specific steps or programs
4. Other that your firm has taken (or help other firms) to
4) Role of firm size and resources support customer service and problem resolution
5) Contrast: AI implementations in CRM that failed using AI technologies. Is there a formal process in
place? How (were) are these programs evaluated?
Time check: 8 minutes
Which ones were effective (ineffective)? What made
Q4. How would you describe the nature and pace the difference?
of change in customer service and problem reso- Elaboration Probes
lution that these AI-driven CRMs have produced?
1) Note whether the program is formal/informal
Will this pace of change slow down, remain the
2) Focus Employees (esp. service employees)
same, or accelerate over the next decade?
1. Customers/channel partners
Elaboration Probes 2. Vendors/other business partners
1) Note terminology used: Disruptive/radical/incre- 3. External experts (consultants)
mental/application/other 4. Other
2) Note Categorization criteria: Scale, speed, com- 3) Criteria was evaluation done? Formal/informal
petitive impact, convergence etc. 1. ROI/Profits (short and long term)
2. Productivity (short and long term) truly appreciate your participation and willingness to
3. Tangible demand-side impact (e.g., customer share your views with us. We would like to reaffirm
relationships, esp. customer satisfaction) our original assurance of confidentiality. Would you
4. Tangible supply-side impact (new service have any problems with us thanking you by name for
product, process, technology) your help in our finished report (without attributing
5. Intangible work environment impacts any opinions)?
(morale, empowerment, interaction)
4) Note effectiveness differentiators Record response: [ ] Yes - feel free to thank me by
name [ ] No - do not identify me. Finally, we will be
Time check: 13 minutes very pleased to share a copy of our final report with
you. The report should be completed by the end of this
Before we close for today, would you like to add
summer. Once again, thank you so much for your help.
any other comments related to AI-Driven CRM in
the area of customer service and problem reso- Total: 60 minutes
lution that we may not have covered in our con-
The full interview transcripts are available at:
versation so far?
https://drive.google.com/file/d/1fTF1IlFEt-v89o9v3ohIx
Elaboration Probes None HIjUhFdN_v0/view?usp=sharing
Thank you again for your time and your insights. We
Citation: Wang, Ph.D JF. (2023). The impact of artificial intelligence (AI) on customer relationship manage-
ment: a qualitative study, Int. J. Manag. Account. 5(5), 74-88.
https://doi.org/10.34104/ijma.023.0074090

The Impact of Artificial Intelligence (AI) On Customer Relationship Management

Uploaded by

Copyright:

Available Formats

The Impact of Artificial Intelligence (AI) On Customer Relationship Management

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Impact of Artificial Intelligence (AI) On Customer Relationship Management

Uploaded by

Copyright:

Available Formats

International Journal of Management and Accounting, 5(5), 74-88,

Publisher homepage: www.universepg.com, ISSN: 2707-4641 (Online) & 2707-4633 (Print)

The Impact of Artificial Intelligence (AI) on Customer Relationship

INTRODUCTION: proper protocols on the relevant informants who have

Fig. 2: Relationships of Analytic Categories.

Table 1: Analytic Categories and Concepts.

human supervision (Salesforce.com, Neural Networks).

Chosen Method and topics that need to be covered during the

Fig. 3: Developing Themes and Concepts (Taylor et

Fig. 4: Analysis: Coding Scheme for the AI-Driven CRM Space.

Emergent Themes AI features in CRM systems. Another emergent theme

Fig. 5: Analytical Diagram: Key Cultural Categories in AI-Driven CRM Space.

Emergent Themes solve problems, and provide recommendations. These

common emergent themes include 2) Understanding each customer individually and

Fig. 6: Integrated Cultural & Analytic Categories in AI-Driven CRM Space.

DISCUSSION: key findings based on the qualitative analysis of the

and integration of AI technologies, contribute to this AI on customer relationship man-agement. IDC:

Time check: 8 minutes Time check: 10 minutes

You might also like