Impact of Customer Relationship Management Dimensions on Customer Retention through Customer Satisfaction: A Customer Perspective of Balkan Transition Economies

One of the main challenges of service providers is the retaining of the acquired customers. Customer Relationship Management (CRM) has proved to be an effective tool in reaching this objective. Many researchers tested the effect of CRM on customer retention (CR), but most of the studies were focused on developed economies and were analyzed from a business perspective. The literature review highlights the lack of studies that examine the impact of CRM on CR from a customer perspective and that are focused on transition economies, especially in Balkan developing countries. Considering this, our paper aims to develop a model that proves the impact of CRM on CR, focused on three Balkan countries (Kosovo, Albania, and North Macedonia). Data was collected using an electronic questionnaire from a sample that consists of 764 residents of the three above-mentioned Balkan countries, all customers of the services sector. The research model shows the impact of three CRM dimensions (key customer focus - KCF, technology-based CRM - TCRM, and CRM knowledge management – KM) on customer retention through customer satisfaction (CS). The model fit and research hypotheses were tested using Confirmatory Factor Analysis and Structural Equation Model. Findings show a positive impact of KCF, TCRM, and KM on CS, which as a result positively affects CR as well. Theoretical and practical implications are discussed.


Introduction
One of the main objectives of a services marketing strategy is to create long-term, beneficial relationships with customers, translated into higher loyalty and retention rates.This is because of the retention costs which are lower than customer attraction costs.According to Many studies investigated the impact of CRM dimensions in the services sector.Herman et al. (2020) state that CRM has a positive impact on long-term customer satisfaction and retention.This is supported by other studies as well (Lo et al., 2010;Al-Qeed, 2017;Al-Gasawneh et al. 2021).Additionally, the study of Khan et al. (2022) in the tourism industry found that customer satisfaction mediates the relationship between CRM and customer retention (CR).But, despite the amount of research conducted in this field, from the literature review, it can be noticed that the vast majority of studies are related to developed countries.It can be noticed a lack of literature regarding developing and transition economies, including Balkan countries.Kapoulas & Ratković (2015) examined e-CRM in tourism in Serbia, but no categorization of dimensions was investigated.The same approach can be found in the study of Stokić et al. (2019) who investigated CRM usage in public libraries in Serbia, Montenegro, and Bosnia and Herzegovina.Other researchers focused on Kosovo and North Macedonia also treated CRM as a general concept without distinguishing between its dimensions and their separate impact (Gashi & Gashi, 2021;Nure, 2018;Rexhepi et al., 2019).Another issue that characterizes those studies is they mostly measure the CRM effect from a business perspective, leaving aside the customer perspective.
Based on the identified gap in the literature, our paper aims to fill this gap by providing a model that investigates the effect of each CRM dimension separately, and their effect on customer satisfaction and retention, focused on three transition economies of the Balkan region, from customer perspective.Therefore, we consider that our model can be used as a pioneering model for transition economies and it can be used as practical guidance for service providers of similar countries.

Research hypotheses 2.2.1. Key customer focus and customer satisfaction
Key customer focus (KCF) relates to a business approach focused on customers.This means that businesses with a customer-centric focus deliver personalized offerings, added value, and cocreation opportunities to their key customers (Sin et al, 2005).As a result, this approach helps businesses achieve a long-lasting competitive advantage (Asikhia, 2010).Many studies found that this business philosophy positively impacts customer satisfaction -CS (Irfan et al., 2013).The study of Cai (2009) conducted with Chinese companies revealed a positive relationship between KCF and CS, which led to better financial performance.The same findings are shown in the model of Aburayya et al. (2020), focused on healthcare centers in the United Arab Emirates, which shows that the customer orientation of employees positively impacts customer satisfaction.Another study with customers in the services sector also found that the customer-oriented approach of employees improved customer satisfaction (Ngo et al, 2020).Based on the above-mentioned studies and findings, we raised the first research hypotheses as follows: H1: Key customer focus has a positive impact on customer satisfaction.

Knowledge management and customer satisfaction
According to Alshourah et al. (2018), knowledge management (KM) as a dimension of CRM, refers to the transformation process of customer data to knowledge, that helps businesses improve their services based on what they learn from and about their customers.Migdadi (2021) treats KM as a CRM dimension that consists of knowledge creation, acquisition, sharing, and application from customers, about customers, and for customers.Businesses that operate based on KM use this knowledge to have higher customer satisfaction and better performance (Abbas & Kumari, 2021).The study of Chaithanapat et al. (2022) found that leadership with a customer knowledge management approach had a positive impact on customer satisfaction, studies as a part of business operational performance.In their research with manufacturing businesses, Anil & Satish (2017) used KM as a practice of total quality management.In their study, they found that KM had a positive effect on CS level.Another study reveals that CRM based on KM enhances marketing performance including CS.The same results are also found in the study of AlQershi et al. (2020).Based on the above discussion, we propose the second research hypothesis for testing: H2: Knowledge management has a positive impact on customer satisfaction.

Technology-based CRM and customer satisfaction
Technology-based CRM helps businesses to collect customer data so that they can profile them into categories, and understand and better satisfy their needs (Sofi et al., 2020).This is enabled by some tools like customer support process automation, customer information systems, and informative or call centers (Ghodeswar, 2001).Hashemzadeh et al. (2011) show that through technology-based CRM businesses collect and process customer data to better meet their needs, and this increases the satisfaction and retention of customers.Kumar et al. (2021) studied electronic CRM usage in the banking industry and found that technologyenabled businesses have updated and real-time customer data.This led to higher service quality and better connection and interaction with their customers, which resulted in improved levels of customer satisfaction.Rashwan et al. (2020) revealed that due to the technology-based CRM used by businesses, customers have easier access to services, at any time, from anywhere, and at a lower cost, and they can customize their services based on their needs and preferences.According to the authors, this increased customer satisfaction with service providers and impacted higher loyalty rates.These findings are also in line with the studies of Khan et al. (2022) and Mohamed et al. (2022).The studies we reviewed led to the third research hypothesis as follows: H3: Technology-based CRM has a positive impact on customer satisfaction.

Customer satisfaction and customer retention
According to Han et al. (2017), customer satisfaction (CS) stands for the level at which the product/service performance has met customer expectations.Authors and practitioners relate CS with customer retention (Al-Ansi et al., 2019;Kim et al., 2020).Lee et al. (2020) investigated customer satisfaction in the restaurant sector and found that it positively impacted customer retention (CR).Amin (2016) studied the behavior of banking sector customers.His research shows that if customers are satisfied with the services provided to them, they tend to be more loyal and continue to be customers of the specific bank.The findings of the Yu et al. (2021) study, conducted in the hotel industry, also show that satisfying customer needs is the key to retaining customers in the long term.Similarly, Cheng et al. (2018) investigated CR in the hotel industry and found that if service managers can implement compensation strategies for customers with complaints, the customers' satisfaction level will get higher and this will make them want to revisit the hotel.Based on this research, we developed the fourth research hypothesis: H4: Customer satisfaction has a positive impact on customer retention.

Instrument for data collection and measurement scale construction
Data was collected using an electronic questionnaire, which consisted of 23 Likert scale questions with five degrees of evaluation: 1=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree.Items used to measure the research variables were derived from previous studies and they were adapted for our research purpose.It should be mentioned that the original items for CRM dimensions were created to measure the constructs in the context of businesses (managers), but we adopted them to be used in terms of customers.Due to this, we excluded the "CRM organization" dimension, which could not be measured using data from customers.So, our research includes three dimensions of CRM.All sources of item construction are shown in Table 1.

Item Source Key customers focus
Through ongoing dialogue, the organization works with us individually to customize their offerings Sin et al., 2005 The organization provides customized services and products to us, as key customers The organization makes an effort to find out what our needs are When the organization finds that we would like to modify a product/service, they make coordinated efforts to do so

Knowledge management
The organization's employees are willing to help us in a responsive manner Sin et al., 2005 The organization fully understands our needs The organization provides channels to enable ongoing, two-way communication with us We can have prompt service from employees of the organization

Technology-based CRM
The organization has the right technical personnel to provide technical support to us Sin et al., 2005 The organization has the right software to serve us Individual customer information is available at every point of contact The organization maintains a comprehensive database of our activity with the organization as customers Has never disappointed me so far Customer retention

Revisit intention
Al-Tit, 2015 Chance of continuing with current service provider for the next year Spreading the word-of-mouth I am a loyal customer of "X" service provider Hennig-Thurau, 2004 My next service order will take place at "X" In the future, I will order my services at "X" Source: Author

Sample description
764 customers, residents of Balkan countries, were part of the sample.62.57% of respondents are 20 -35 years old.The vast majority is female (67.02%).The largest part comes from Kosovo (37.96%), and others are from Albania and North Macedonia.Detailed descriptive data of the sample can be found in Table 2.

Data analyses
To analyze the research data, we used IBM SPSS Statistics and IBM AMOS 26.The first step was to test the measurement scale for reliability and validity.Next, we examined all the necessary parameters to test the model's goodness of fit and the research hypotheses.The results of these analyses were achieved using the Confirmatory Factor Analysis (CFA) and Structural Equation Model (SME).

Construct's reliability and validity
The metrics we used for reliability and validity were Cronbach's Alpha -α, average variance extracted -AVE, and the discriminant validity test.As shown in Table 3, the values of α for each construct are higher than 0.7, which is the threshold value for reaching valid consistency (Kline, 1994).From the results of this parameter, we can indicate that the scale has a good internal consistency.Furthermore, we checked the convergent validity of the construct.According to Becker et al. (2003), the convergent validity is reached if the AVE values of each construct are higher than 0.5.Based on the values of AVE (Table 3), we can state that the convergent validity of our scale has been reached.At last, we tested the discriminant validity too, which indicates if questions of different constructs are different from the questions of other constructs.To be reached, the construct's square root of AVE (values in the diagonal of Table 4), should be greater than the values below the diagonal, which shows the correlation of the construct with other constructs.As the results in Table 4 indicate, the discriminant validity was also reached for all the constructs.

SEM and conceptual research model estimation
The proposed research model was tested using the Structural Equation Model (SEM) analyses.Figure 2 shows the path coefficients of the relationships between constructs, based on the research hypotheses.Results on the figure show that technology-based CRM has the largest impact, explaining 47% of the variance in customer satisfaction.Further, we examined the main model fit indicators, like CMIN/DF (Chi-square per degrees of freedom), CFI (Comparative Fit Index), RFI (Relative Fit Index), TLI (Tucker-Lewis coefficient), IFI (Incremental Fit Index) and RMSEA (Root Mean Square Error of Approximation).The results are shown in Table 5.According to Marsh & Hocevar (1985) and MacCallum et al. (1996), a good model fit shows a CMIN/DF value lower than 5.0, and RFI, IFI, TLI, and CFI values greater than 0.9.As indicated in Table 5, all these thresholds are reached, so it can be inferred that our research model has a good model fit.

SEM and hypotheses testing
The last step of our analysis is about research hypotheses.To test the hypotheses, we used the Structural Equation Model (SEM), using indicators like regression estimates (β coefficients), p values, and critical ratios -t value.All hypotheses with β coefficient greater than 0.1 in 0.05 significance level (p-value) were accepted, based on Huber et al. (2007).As mentioned before, the first three hypotheses assume that the identified variables (key customer focus, knowledge management, and technology-based CRM) have a positive impact on customer satisfaction.As shown in Table 6, all three hypotheses are supported.The first coefficient β = 0.135 in level of significance p = 0.000, shows that key customer focus has a significant positive impact on customer satisfaction, supporting the first hypothesis.Results also show a positive impact of knowledge management on customer satisfaction, supporting the second hypothesis as well (β = 0.262, p = 0.000).Next, findings show that technology-based CRM has also a positive impact on customer satisfaction (β = 0.369, p = 0.000), so the third hypothesis is also supported.Further, through the fourth hypothesis, we proposed that customer satisfaction has a positive impact on customer retention.This was supported, based on the results that show a significant positive relationship between the two variables (β = 0.676, p = 0.000).

Conclusion
This research aimed to test the impact of Customer Relationship Management (CRM) dimensions on customer retention through customer satisfaction, from a customer perspective.These relationships were tested in the services sector of transition economies of three Balkan countries: Kosovo, Albania, and North Macedonia.For this purpose, we used three CRM dimensions, namely: key customer focus, technology-based CRM, and CRM knowledge management.Empirical results show that all the hypothesized relationships between the model variables are supported.
First, our results show a positive significant impact of key customer focus on customer satisfaction.This means that if service providers focus on identifying and filling customer needs and preferences, offer customized and personalized services for them, and enable them to cocreate their offers, this will create more satisfied customers.This finding complies with other studies in developed economies like Aburayya et al. (2020), Cai (2009), Irfan et al. (2013), andNgo et al. (2020).
Additionally, we also found a positive significant impact of knowledge management on customer satisfaction.This emphasizes the importance of customer data collection, processing, and application so that businesses can provide better services and products that improve customer satisfaction rates.Results are in line with other previous studies (Abbas & Kumari, 2021;AlQershi et al., 2020;Anil & Satish, 2017;Chaithanapat et al., 2022).
Furthermore, findings show a positive significant impact of technology-based CRM on customer satisfaction.This shows that businesses that use the latest technology, digitize, and automate customer service processes tend to have higher customer satisfaction rates.This is because it facilitates the service usage by customers and makes it more cost-effective.Similar findings are also shown in the studies of Hashemzadeh et al. (2011), Khan et al. (2022), Kumar et al. (2021), Mohamed et al. (2022) and Rashwan et al. (2020).
At last, our research findings show that customer satisfaction positively impacts customer retention.This proves that satisfied customers are more likely to be loyal and come back, so through CS service providers can develop long-term relationships with their customers.This is also found in previous studies (Al-Ansi et al., 2019;Amin, 2016;Cheng et al., 2018;Kim et al., 2020;Lee et al., 2020;Yu et al., 2021).
Based on our findings, it can be concluded that research results support the proposed research model and significant implications can be derived.In the following part of this section, theoretical and practical contributions are discussed.

Theoretical and practical contribution
Based on the review of existing literature we can imply that our paper contributes to the theory of relationship marketing and customer relationship management of transition economies, specifically Balkan countries.Furthermore, it also sheds light on how CRM dimensions are perceived by customers.It fills the identified gap in the literature by offering a model that shows the importance and impact of CRM dimensions on the creation of longterm customer-business relationships.It can be considered as an initial work in this field, which can be further enriched and expanded.
Apart from the theoretical contribution, this paper has practical implications too.Our model serves as evidence for CRM managers of service providers that, if used properly, CRM can foster stronger and longer customer-business relationships.It proves that if businesses use technology to create knowledge about their customers and focus on their needs, they will have lifelong customers that will increase their overall business performance.

Limitations and scope for future research
Like any other study, our paper has its limitations too.They will be used as a scope for future researchers who like to expand our current study.First, the research sample is very limited, so it can be expanded to other Balkan countries or transition economies of other regions.This would lead to more credible results.Second, we used three CRM dimensions, excluding the CRM organization, but in the future, other studies can investigate the effect of all four CRM dimensions.Third, we focused only on the services sector, while future studies can expand to the tangible products sector, and distinguish between the effect of CRM dimensions on two industry sectors.

2. 1 .
Proposed research modelAfter consulting the literature regarding dimensions and outcomes of Customer Relationship Management (CRM), a research model is proposed.Figure1illustrates the model and the hypothesized relationships between research variables.As shown in the figure, customer retention (CR) is the dependent variable, while key customer focus (KCF), technology-based CRM (TCRM), and CRM knowledge management (KM) are independent variables impacting CR through customer satisfaction (CS).

Figure
Figure 1 Proposed research model Customer satisfactionMy choice to use this company was a wise one Hennig-Thurau et al. 2002 I am always delighted with this firm's service Overall, I am satisfied with this organization Always fulfills my expectationsHennig-Thurau, 2004

Table 2
Sample descriptive data

Table 3
Scale's reliability and validity

Table 5
Model fit indicators

Table 6
Hypotheses testing