Journal of Theoretical and Applied Information Technology

30th September 2023. Vol.101. No 18

© 2023 Little Lion Scientific

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

AI-DRIVEN PERSONALIZATION IN CUSTOMER

RELATIONSHIP MANAGEMENT: CHALLENGES AND
OPPORTUNITIES
Dr. N. VENKATESWARAN1
1
Professor and Dean, Panimalar Engineering College, Department of Master of Business Administration,
INDIA
1
E-mail: profvenkimba2019@gmail.com

ABSTRACT

Artificial Intelligence (AI) has emerged as a transformative technology with the potential to revolutionize
various industries, including customer relationship management (CRM). This research study aims to
explore the role of AI in enhancing CRM practices and improving customer experiences. This research
paper discusses AI and CRM can analyze vast amounts of customer data, automate routine processes, and
deliver personalized experiences on a large scale. By harnessing the power of AI, organizations can gain
valuable insights into customer preferences, behavior patterns, and purchase history, allowing them to tailor
their offerings and communications to individual customers. This paper explores the potential of AI in
enhancing CRM strategies and improving customer experiences, ultimately leading to increased customer
satisfaction and loyalty. The findings of this research will contribute to the growing body of knowledge on
the application of AI in CRM and provide practical insights for organizations seeking to leverage AI
technologies.
Keywords: Customer Experience; Personalization; Customer Preferences; AI-Driven CRM; Accuracy

1. INTRODUCTION deliver personalized experiences at scale [5]. By

leveraging AI, organizations can gain valuable
Customer Relationship Management (CRM) insights into customer preferences, behavior
plays a pivotal role in organizations' success by patterns, and purchase history, enabling them to
fostering strong and lasting relationships with tailor their offerings and communications to
customers. However, with the rapid advancements individual customers.
in technology, the traditional approaches to CRM
are evolving to meet the demands of an increasingly While AI has gained widespread use in
digital and interconnected world. One technology managerial contexts, there has been a limited focus
that has gained significant attention in recent years on AI within management scholarship over the past
for its potential to transform CRM is Artificial two decades [4]. The existing body of AI literature
Intelligence (AI). AI offers new opportunities to has predominantly developed within computer
enhance customer interactions, improve service science and operations research disciplines, with an
delivery, and drive personalized experiences. This emphasis on exploring machine-handled operational
chapter provides an introduction to the role of AI in tasks. Conversely, organizational and management
enhancing CRM practices, highlighting its benefits, research has primarily examined managerial tasks
challenges, and implications for organizations. [2] that are traditionally performed by humans [4].

AI encompasses a range of technologies, One prominent application of AI in CRM is

including natural language processing, machine the use of chatbots and virtual assistants. These AI-
learning, and predictive analytics, that enable powered tools can handle customer inquiries,
computers to simulate human-like intelligence and provide real-time support, and offer personalized
perform tasks autonomously. In the context of recommendations. Chatbots are available 24/7,
CRM, AI can be utilized to analyze large volumes improving customer service by ensuring swift
of customer data, automate routine processes, and responses and reducing customer wait times. Virtual

7392
 Journal of Theoretical and Applied Information Technology
30th September 2023. Vol.101. No 18
© 2023 Little Lion Scientific

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

assistants, equipped with natural language experiences and gain a competitive edge in the
processing capabilities, can understand and respond digital era. [11]
to customer queries, providing an interactive and
conversational experience. As a result, Currently, there is a significant gap in the
organizations can enhance customer satisfaction, literature when it comes to a comprehensive review
engagement, and loyalty through seamless and specifically dedicated to mapping the integration of
efficient interactions. [5] AI in the CRM domain [12] Addressing this gap,
the objective of this paper is to provide an up-to-
Furthermore, AI-driven predictive analytics date overview of the state-of-the-art in AI adoption
empowers organizations to anticipate customer within CRM and identify emerging themes and
needs and behaviors. By analyzing historical data, potential avenues for future research. To achieve
AI algorithms can identify trends, patterns, and this goal, the authors adopt a systematic and
anomalies, enabling organizations to predict transparent approach, utilizing bibliometric
customer preferences, identify upselling or cross- techniques to map the research landscape in an
selling opportunities, and proactively address unbiased manner [14]. Specifically, the study
potential issues. This enables organizations to employs bibliographic coupling to analyze
deliver more targeted marketing campaigns, references, establishing intellectual connections
personalized recommendations, and proactive among articles, and utilizes co-occurrence analysis
customer support, ultimately leading to improved of keywords to gain a comprehensive understanding
customer satisfaction and retention. [10] of the prominent terms, thus enabling the creation
of structural visualizations of the research domain.
However, as AI becomes increasingly
integrated into CRM practices, organizations must To explore the application of deep learning
also consider the ethical implications. Privacy techniques in the context of customer relationship
concerns arise with the collection and utilization of management (CRM) with a particular emphasis on
vast amounts of customer data, necessitating robust predictive analytics. In the paper, El Amrani et al.
data protection and governance measures. discuss the methodology and approach used to
Moreover, the potential for algorithmic bias and implement a predictive CRM system based on deep
discriminatory outcomes must be addressed to learning. They provide insights into the process of
ensure fairness and inclusivity in AI-driven CRM developing and training deep learning models using
systems. large datasets to predict customer outcomes, such as
purchase behavior, churn probability, or response to
This chapter will draw upon the works of marketing campaigns [3].
leading researchers and practitioners in the field to
provide a comprehensive overview of the role of AI Prior reviews within the field have primarily
in enhancing CRM practices. Noteworthy concentrated on specific dimensions, such as
contributions include the research who investigated examining the challenges and applications of Big
the impact of AI chatbots on customer satisfaction Data and AI in customer journey modelling or
and loyalty, and the study by Verhoef et al. (2020), exploring the potential impacts of Big Data and AI
which examined the use of predictive analytics in on key success factors in CRM [13] and consumers’
personalized marketing campaigns. Additionally, decision-making processes [7].
the ethical considerations of AI in CRM will be
explored to gain insights into managing privacy 2. RESEARCH METHODOLOGY
concerns and mitigating algorithmic bias.[6], [9] This chapter presents the research
methodology employed to investigate the role of
By exploring these studies and others, this Artificial Intelligence (AI) in enhancing Customer
chapter aims to shed light on the potential benefits, Relationship Management (CRM). The chapter
challenges, and best practices associated with the outlines the research design, data collection
integration of AI into CRM. Through a methods, data analysis techniques, and ethical
comprehensive understanding of the role of AI in considerations.
enhancing CRM practices, organizations can
The research design for this study is primarily
strategically adopt and effectively utilize AI
exploratory and aims to gain insights into the
technologies to deliver exceptional customer
application and impact of AI in CRM. It involves a
mixed-methods approach, combining quantitative

7393
 Journal of Theoretical and Applied Information Technology
30th September 2023. Vol.101. No 18
© 2023 Little Lion Scientific

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

analysis of CRM data and qualitative methods such H1a: Service experience using AI and CRM is
as interviews and surveys. The use of mixed positively and significantly related to customer
methods allows for a comprehensive understanding satisfaction.
of the phenomenon under investigation. [1]
H1b: Service experience using AI and CRM is
positively and significantly related to
Customer data from CRM systems collected, personalization.
including purchase history, customer interactions,
and engagement metrics. These data will be used H2a: Service experience only with CRM is
for quantitative analysis to evaluate the impact of positively and significantly related to customer
AI on CRM performance indicators. Online surveys satisfaction.
administered to CRM users and customers to gather H2b: Service experience only with CRM is
their perspectives on AI-driven CRM practices. The positively and significantly related to accuracy.
surveys will capture demographic information,
perceptions of AI in CRM, and satisfaction with AI- H3: Customer satisfaction using AI and CRM is
powered interactions. positively and significantly related to accuracy.
H4a: Customer satisfaction has a significant
Descriptive statistics used to summarize and mediating effect between service experience of
analyze CRM data, such as customer demographics, using AI and CRM and accuracy.
purchase frequency, and customer engagement
metrics. Inferential statistical techniques, such as
regression analysis, employed to examine the 3. PROCEDURE
relationships between AI adoption in CRM and The study utilizes an online survey due to its
various performance indicators, such as customer numerous advantages, including flexibility, cost-
satisfaction, loyalty, and revenue. efficiency, prompt results, convenient
administration, and easy storage. To minimize
Thematic analysis conducted to identify and missing data, the survey design incorporated
analyse recurring themes and patterns in the measures to prevent respondents from skipping
interview and survey responses. This analysis will questions.
provide insights into the perceived benefits,
challenges, and implications of AI in CRM. Before initiating data collection, a pilot test of
the questionnaire was conducted. The data was
Upon identifying the articles that specifically gathered in February 2023, ensuring that
focus on the investigated theme, we conducted two respondents were informed about the voluntary
bibliometric analyses, namely bibliographic nature of their participation and their right to
coupling and keywords' co-occurrence, to trace the withdraw at any point. It was explicitly
current state-of-the-art regarding the integration of communicated that there were no right or wrong
AI in CRM contexts. Prior to applying the answers, allowing respondents to provide truthful
bibliometric analysis, we ensured the completeness and accurate responses. Within a two-week period,
of the information within the dataset by manually a total of over 500 responses were received.
adding any missing cited reference data [3]. To Following the removal of invalid responses (i.e.,
initiate the bibliographic coupling analysis, we those with identical answers for all questions), 380
clustered the sampled articles based on shared responses remained for further analysis. The sample
references. This clustering was achieved by size was considered sufficient, taking into account
quantifying the frequency with which two articles the calculation of confidence level, population size,
cited the same references. Both bibliographic and margin of error.
coupling and co-citation analysis are essential
techniques in science mapping. A co-citation The sampling plan outlines the target
network is formed when two articles (nodes) are population, sample size, and sampling techniques to
jointly cited by another document, whereas a be used in the research study investigating the role
bibliographic coupling network is formed when of Artificial Intelligence (AI) in enhancing
both articles (nodes) make references to a third Customer Relationship Management (CRM). The
document, establishing a link. target population for this study includes CRM
The following research hypothesis are postulated professionals, industry experts, CRM users, and
in this study customers who have interacted with AI-powered
CRM systems. The population may vary based on

7394
 Journal of Theoretical and Applied Information Technology
30th September 2023. Vol.101. No 18
© 2023 Little Lion Scientific

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

the specific objectives and scope of the research. a good fit for the model: χ2 = 7.72, df = 2, χ2/df =
The sample size will depend on the research 3.39, p < .001; CFI = 0.96; TLI = 0.92; RMSEA =
objectives, available resources, and the desired level 0.08. The same procedures were applied to the other
of statistical power. It should be large enough to three constructs.
provide meaningful insights while considering
feasibility and practicality. The model fit indices for the second-order
structure of personalization with AI were found to
The measurement items for the study variables be: χ2 = 121.08, χ2/df = 3.32, p < .001; CFI = 0.96;
were derived from previously validated studies, TLI = 0.95; RMSEA = 0.07. Similarly, for customer
demonstrating strong reliability and validity. In this engagement, the indices were: χ2 = 887.04, df =
study, a 7-point Likert scale ranging from 1 265, χ2/df = 3.45, p < .001; CFI = 0.94; TLI = 0.97;
(Strongly disagree) to 7 (Strongly agree) was RMSEA = 0.08. Lastly, for accuracy, the model fit
employed for all the measurement items. This scale indices were: χ2 = 262.67, df = 98, χ2/df = 2.63, p <
was chosen to provide flexibility in responses and to .001; CFI = 0.97; TLI = 0.96; RMSEA = 0.08.
discourage respondents from selecting a neutral
stance. Table 1 Model fit indices for first and second-
order factors
The Likert scale used in this study allowed
participants to express their agreement or AI and CRM χ2 df χ2/df p CFI TLI RMSEA

disagreement on a continuum, ranging from 1 Second

762.75 233 3.39 *** .97 .94 08
Customer order
(Strongly disagree) to 7 (Strongly agree). This range Satisfaction
One
was selected to ensure that respondents did not lean [CS]
order
7.72 2 3.39 *** .96 .92 .08

towards a neutral position, encouraging more Second

121.08 40 3.32 *** .96 .95 .07
diverse and nuanced responses from the target Personalization order

audience. [PER} One

5.92 2 2.96 *** .98 .97 .07
order

Second
Within the respondent pool, approximately half Customer order
887.04 265 3.45 *** .94 .97 .08

identified as female (50.8%). The largest age group engagement

[CE] One
6.78 2 3.39 *** .99 .97 .07
represented was between 18 and 25 years old, order

accounting for over 28 percent of respondents Second

262.67 98 2.63 *** .97 .96 .08
order
(28.2%). When considering educational Accuracy
[ACC] One
background, around 37 percent of respondents order
3.42 2 1.71 *** .99 .99 .04

possessed university degrees (37.1%).

The model fit indices, as presented in Table 1,
demonstrated a good fit for both the first-order and
4. MEASUREMENT MODEL
second-order factors, with all path coefficients
between the indicators and their respective factors
In this study, the majority of variables including
found to be significant at the 0.05 level. Given the
AI and CRM with customer satisfaction,
satisfactory fit of both factor structures, subsequent
personalization, customer engagement, and
analyses incorporated both the first-order and
accuracy were assessed as second-order factor
second-order factors to assess the proposed
constructs. The validity of these constructs was
relationships. The second-order factors were
evaluated following the procedures outlined in [8].
employed to examine the overall effects, utilizing
Specifically, to assess customer satisfaction and its
an aggregated scale comprising the average scores
four dimensions (product/service quality, customer
of the subdimensions across all second-order
service, trust and security, convenience), a
factors, following the approach of [5], [8]. On the
confirmatory factor analysis (CFA) was employed
other hand, the first-order factors were assessed to
using a hierarchical model.
understand the effects of the sub-dimensions within
each factor on the outcome variables.
The second-order factor model demonstrated
acceptable fit indices: χ2 = 762.75, df = 233, χ2/df =
The assessment of the CFA for the proposed
3.39, p < .001; CFI = 0.97; TLI = 0.94; RMSEA =
model prior to hypothesis testing indicated that the
0.08. To further examine the second-order factor
model exhibited satisfactory fit indices [9]. The
structure, a one-factor confirmatory factor analysis
results showed that the model demonstrated
(CFA) was performed on the average scores of the
acceptable fit, as evidenced by the following
four first-order constructs, [5]. The results indicated

7395
 Journal of Theoretical and Applied Information Technology
30th September 2023. Vol.101. No 18
© 2023 Little Lion Scientific

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

indices: χ2 = 932.52, df = 254, χ2/df = 3.67, p < satisfaction and accuracy exceeded 37% and 71%
.001; CFI = 0.95; TLI = 0.95; RMSEA = 0.07. respectively, demonstrating substantial explanatory
Additionally, all items displayed significant capacity across all models.
loadings on their respective constructs.
Hypotheses H1a and H1b propose a positive
The composite reliabilities of all factors association between service experience with AI and
exceeded the predefined threshold of 0.70, CRM and both customer satisfaction and
demonstrating satisfactory internal consistency. personalization. The findings presented in Table 3
Furthermore, the average variance extracted (AVE) provide evidence supporting H1a and H1b, revealing
for each factor surpassed the recommended that service experience with AI and CRM had a
threshold of 0.50, indicating adequate convergence significant impact on customer satisfaction (β =
as suggested by Fornell and Larcker (1981). The 0.24, p < .05) and personalization (β = 0.41, p <
analysis of standardized residuals and modification .001). Hence, H1 is confirmed.
indices revealed no statistically significant changes
required to be made to the model.[11]. Table 2 H2a and H2b propose that service experience
presents the validities of all variables. only with CRM positively affects customer
satisfaction and accuracy. Supporting Hypothesis 2,
To verify discriminant validity, a heterotrait- the findings presented in Table 3 indicate a
monotrait (HTMT) ratio test was conducted in significant impact of CRM service experience on
AMOS. The HTMT results revealed that all both customer satisfaction (β = 0.74, p < .001) and
correlation ratios were below the threshold of 0.80, accuracy (β = 0.29, p < .001). Hypothesis 3 was
indicating acceptable discriminant validity in line supported as the analysis revealed a significant
with the findings of Nguyen et al. (2019). influence of customer satisfaction using AI and
Additionally, the square root of the average CRM have positive relationship on accuracy (β =
variance extracted (AVE) for each construct 0.39, p < .001), indicating that customer satisfaction
exceeded the correlation between constructs, plays a crucial role in fostering accuracy.
providing further evidence of discriminant
validity.[10]. Table 3 Hypothesis test results

Table 2 Discriminant validities for the study Path β Sig

variables Service experience with AI and CRM → Customer 0.24 *
satisfaction
Service experience with AI and CRM → 0.41 **
AVE CR CS PER CE ACC Marker Personalization
CS .83 .95 .91 Service experience with CRM → Customer 0.74 **
PER .91 .94 .52* .95 satisfaction
CE .85 .95 .44* .52* .92 Service experience with CRM → Accuracy 0.29 **
ACC .84 .93 .61* .38* .43* .91 Customer satisfaction using AI and CRM → 0.39 **
Marker .82 .92 .05 .08 .06 .07 .90 Accuracy
R2
Customer satisfaction .39
Note: The diagonal elements represent the square
Accuracy .76
root values of the Average Variance Extracted
(AVE), while the off-diagonal elements indicate the
correlations among constructs. It is worth noting Model fit: χ2 = 223.49, df = 65, χ2/df = 3.65,
that the correlation values between variables, both p < .001; CFI = .99; TLI = .98; RMSEA = .08
with and without the marker variable, remain
unchanged. Additionally, significant correlations The study conducted tests to examine mediation
are denoted by *p < .05." hypotheses in H4a, specifically exploring the role of
customer satisfaction as a mediator between AI and
CS = Customer satisfaction; PER = Personalization; CRM, and accuracy. The findings presented in
CE = Customer engagement; ACC = Accuracy; Table 4 demonstrate that customer satisfaction has
significant effects on the relationships between
5. HYPOTHESIS TESTING RESULTS service experience with AI and CRM, and accuracy.
These results indicate that service experience with
The hypothesis fit indices were deemed AI and CRM directly influences both customer
acceptable: χ2 = 223.49, df = 65, χ2/df = 3.65, p < satisfaction and accuracy. Consequently, customer
.001; CFI = 0.99; TLI = 0.98; RMSEA = 0.08. satisfaction is identified as a partial mediator in the
Moreover, the R-squared values for customer

7396
 Journal of Theoretical and Applied Information Technology
30th September 2023. Vol.101. No 18
© 2023 Little Lion Scientific

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

relationship between service experience with AI and operated interactions. Comparatively, customers
CRM, along with personalization and accuracy. tend to prefer engaging with employees as they
generally have a better experience resulting from
Table 4: Results for mediation testing the personalized and human touch offered by
Mediator Between Estimate Lower Upper P- Type of
employee responses.
value mediation
CS SE PER .08 .02 .13 .018 Partial
CS CRM ACC .29 .25 .37 .001 Partial Table 5 Comparison means of all dimensions of
customer service experience with AI and CRM
The 95% bootstrapping confidence intervals
(CI) for the indirect effects on accuracy and Customer Dimensions Means AI and CRM Means
Dimensions
personalization of service experience with AI and Responsiveness 5.27 Responsiveness 4.92
CRM (CI = 0.24, 0.40) and service experience with Assurance 5.32 Assurance 4.87
CRM (CI = 0.31, 0.46) exclude the value of 0. As a Proactivity 5.21 Proactivity 4.89
Security and Privacy 5.30 Security and 4.72
result, this test provides confirmation of the Privacy
mediation effects of customer satisfaction, thereby
providing support for H4a. Upon conducting a post-hoc analysis, it was
found that customers rate the dimension of
6. DISCUSSION customer responsiveness which is higher (5.27) than
AI and CRM responsiveness dimension (4.92)
This study investigates the impact of both indicating that customers generally have a better
employee service experience and along with AI and experience when interacting with AI and CRM
CRM powered service on customer satisfaction and systems due to the standardized nature of the
accuracy, recognizing their significance in responses.
enhancing operational efficiency and customer
experience within service organizations in the Assurance refers to the ability of the AI and
current service dominant logic era. The study CRM systems to provide correct and reliable
findings suggest that customer satisfaction, information about customers. It involves accurate
personalization, and accuracy are influenced by the data capture, processing, and analysis, which helps
overall experience with AI and CRM. However, in personalizing customer interactions and
when examining the specific dimensions of service improving overall satisfaction. Based on the ratings
experience, these relationships become more of both AI and CRM assurance, customers rate the
intricate and complex. later much higher (4.87 vs 5.32)

6.1 AI and CRM experience with customer Proactivity is an important quality dimension
satisfaction, personalization and accuracy related to AI and CRM (Customer Relationship
Management) systems. AI-powered CRM systems
The study findings indicate that the overall can use predictive analytics to analyze customer
experience with AI and CRM significantly impacts data and behavior patterns to anticipate future needs
customer satisfaction, personalization, and or issues. AI and CRM systems can leverage
accuracy. However, when examining the individual contextual information to deliver personalized and
dimensions of AI and CRM service, none of them proactive engagement with customers. This
demonstrate a significant relationship with customer dimensions why employees' proactivity on AI and
satisfaction. On the other hand, the dimensions of CRM had a significant effect on customer service
competence, friendliness, responsiveness, and experience and was rated higher by customers (5.21
professionalism exhibited by employees exert a vs 4.89).
significant influence on the outcome variable.
AI and CRM systems deal with a vast amount
The dimension of responsiveness, in the context of of sensitive customer data, including personal
employees, refers to their ability to promptly deliver information, transaction details, and communication
services, willingness to assist, and availability to records. It is crucial to ensure the security of this
address customer requests. In contrast, in the case of data to protect it from unauthorized access,
AI and CRM services, responsiveness relates to the breaches, or malicious activities. AI and CRM
timely response of AI tools. Although AI-powered systems should implement secure authentication
tools can provide quick responses, these responses mechanisms to ensure that only authorized users
are often standardized due to the nature of machine- can access sensitive customer data. By addressing

7397
 Journal of Theoretical and Applied Information Technology
30th September 2023. Vol.101. No 18
© 2023 Little Lion Scientific

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

security and privacy as important dimensions, effective service and improved operational
organizations can build trust with their customers efficiency, it is important to exercise caution and
and protect sensitive data in AI and CRM systems not excessively rely on these tools at the expense of
with mean value of (5.30 vs 4.72) minimizing employee involvement in customer
service. This study emphasizes the significance of
7. IMPLICATIONS responsiveness, assurance, personalization, and
accuracy in AI tools, as they are closely linked to
7.1 Theoretical Implications customer satisfaction. Therefore, it is crucial to
focus on reinforcing these strengths to enhance
In the past decade, there has been a significant customer-related outcomes.
increase in the popularity of customer service
experience research as a key marketing strategy for In addition, the study highlights that not all
achieving customer retention and loyalty. It is customer service dimensions are directly impacted
crucial to identify effective antecedents or drivers of by AI and CRM. Therefore, it is crucial for
customer satisfaction to optimize organizational management to accurately identify the specific
performance. This study makes a valuable factors that influence the dimensions relevant to
contribution to customer satisfaction and accuracy organizational outcomes. The research findings
research by examining two important touchpoints in specifically indicate a significant relationship
the consumer purchase journey: the experience and between affective engagement and personalization,
interaction with AI and CRM. While most customer as well as accuracy. In light of this, marketers
satisfaction research primarily focuses on should prioritize strategies that effectively address
aggressive marketing through competitive customers' emotions, such as fostering passion and
promotions or defensive marketing through loyalty excitement, to enhance affective engagement.
programs, this study provides a fresh perspective by
exploring customers' service experience with AI and 8. LIMITATIONS OF FUTURE RESEARCH
CRM. It sheds light on how machines or robots can
potentially contribute to the customer service While the study made every effort to ensure
experience and ultimately impact organizational rigor, it is important to acknowledge certain
performance. limitations. One such limitation is the selection of
only four dimensions to measure the customer
While previous research on AI and CRM has service experience with AI and CRM. The study
predominantly focused on the technical aspects of recognizes that AI and CRM encompass multiple
AI tools, this study goes beyond that by examining dimensions, and by narrowing the focus to only four
their application in the realm of customer dimensions, there is a possibility of inhibiting a
satisfaction and accuracy. It highlights that AI- comprehensive understanding of the impact of AI-
powered CRM services are not solely meant to powered tools on the customer service experience.
enhance operational efficiency but can also be The extendable work from this study may develop
leveraged as a marketing approach to engage comprehensive framework that encompasses a
customers and foster customer loyalty. broader range of dimensions to measure the
Traditionally, customer service has been closely customer service experience with AI and CRM.
linked to customers' behavioral intentions. This would provide a more holistic understanding
However, this study takes a distinct approach by of the impact of AI-powered tools on customer
investigating the customer service experience satisfaction and organizational outcomes. Explore
within the context of AI and CRM. By how the combination of human expertise and AI
incorporating these variables, it aims to understand capabilities can lead to improved customer
their individual and unique contributions to experiences and organizational performance. We
explaining customer satisfaction and accuracy. can also examine how AI tools influence customer
lifetime value, repeat purchases, and advocacy.
7.2 Practical Implications Investigate strategies to foster long-term customer
relationships using AI-powered CRM systems.
The findings from this research can be utilized
by management and marketers to develop effective 9. CONCLUSION
and sustainable strategies, essential for surviving
and remaining competitive. While incorporating AI This study offers valuable contributions to
tools has become common practice for cost- both theory and practice. From an academic

7398
 Journal of Theoretical and Applied Information Technology
30th September 2023. Vol.101. No 18
© 2023 Little Lion Scientific

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

perspective, it provides a comprehensive analysis of learning. International Journal of Computer
the research landscape surrounding the integration Applications, 179(31), 26-33.
of AI in CRM. By employing rigorous literature [4] Fornell C., Larcker D.F. Structural equation
review methodologies and objective bibliometric models with unobservable variables and
techniques, this study maps the diverse subfields measurement error: algebra and statistics. J.
that have emerged within this fragmented literature. Market. Res. 1981;18(3):382–388.
As a result, scholars are empowered with a holistic doi: 10.1177/002224378101800313.
understanding of AI in CRM, enabling them to [5] Jayachandran S., Sharma S., Kaufman P.,
position their own contributions effectively within Raman P. The role of relational information
this dynamic field. processes and technology use in customer
relationship management. J.
In response to the call made by Raisch and Market. 2005;69(4):177–192.
Krakowski (2020) to develop comprehensive [6] Li, Y., & Karahanna, E. (2019). What drives
perspectives on the management AI debate, this users' continuous usage intention of AI-
study identifies and describes three subfields that powered chatbots? Information &
shape and characterize the literature within the Management, 56(8), 103168.
CRM domain. From a managerial perspective, this [7] Klaus, P. and Zaichkowsky, J. (2020), “AI voice
study provides valuable insights for organizations bots: a services marketing research agenda”,
and managers seeking to leverage AI within their Journal of Services Marketing, Vol. 34 No. 3,
CRM systems. It offers practical guidance on pp. 389-398.
harnessing the opportunities presented by AI and [8] Kumar V., Pansari A. Competitive advantage
outlines the key directions in which AI-CRM through engagement. J. Market.
integration is evolving. By incorporating the Res. 2016;53(4):497–514.
recommendations and managerial practices outlined [9] Mittal, S., & Khan, M. L. (2021). Ethical
in this study, organizations can effectively drive the concerns in artificial intelligence-based
integration of AI into their CRM strategies, customer relationship management. Journal of
ensuring a powerful and productive implementation. Business Research, 132, 347-358.
The findings of this study suggest a notable shift in [10] Niknejad, N., Ismail, W., Bahari, M.,
CRM from a data-driven strategy to an AI-driven Hendradi, R. and Salleh, A.Z. (2021),
[2] “Mapping the research trends on blockchain
strategy . Furthermore, recent scholarly research
has taken a broader strategic perspective in technology in food and agriculture industry: a
approaching the topic, aiming to leverage the power bibliometric analysis”, Environmental
of AI to enhance CRM as a whole, rather than Technology & Innovation, Vol. 21, p. 101272.
solely focusing on investigating specific [11] Raisch, S. and Krakowski, S. (2020),
technological applications to improve operational “Artificial intelligence and management: the
efficiency or customer experience within individual automation-augmentation paradox”, Academy
CRM activities. of Management Review, Vol. 46 No. 1, pp. 1-
48
[12] Verhoef, P. C., Kannan, P. K., & Inman, J. J.
REFERENCES: (2020). From multi-channel retailing to omni-
channel retailing: Introduction to the special
[1] Callon, M., Courtial, J.P., Turner, W.A. and issue on multi-channel retailing. Journal of
Bauin, S. (1983), “From translations to Retailing, 96(1), 1-5.
problematic networks: an introduction to co- [13] Zerbino, P., Aloini, D., Dulmin, R. and
word analysis”, Social Science Information, Mininno, V. (2018), “Big data-enabled
Vol. 22 No. 2, pp. 191-235 customer relationship management: a holistic
[2] Colson, E. (2019), “What AI-Driven decision approach”, Information Processing &
making looks like”, Harvard Business Review, Management, Vol. 54 No. 5, pp. 818-846.
available at: https://hbr.org/ 2019/07/what-ai- [14] Zupic, I. and Cater, T. (2015), “Bibliometric
driven-decision-making-looks-like (accessed methods in management and organization”,
20 December 2021) Organizational Research Methods, Vol. 18
[3] El Amrani, R., Baina, A., Rabea, A., & No. 3, pp. 429-472, doi: 10.1177/
Ennouary, Y. (2021). Artificial intelligence 1094428114562629.
and CRM: Predictive approach based on deep

7399