Source-linked AI summary
Customer Relationship Intelligence: Integrating CRM and MDM for Enhanced Customer Engagement
Tejasvi c. Addagada
TL;DR
The study examines whether CRM, MDM, and CKM jointly form a CRI framework for improving customer engagement. Using a cross-sectional survey of 100 participants and regression, correlation, and mediation analyses, it finds that CRM and CKM are significant positive predictors, whereas MDM’s direct and mediated effects are not statistically significant. The findings support treating the study as exploratory and motivate larger, longitudinal, sector-specific research.
Problem
The study addresses the need for an integrated CRI framework combining CRM, MDM, and CKM to explain customer engagement.
Method
A cross-sectional survey of 100 participants across four sectors was analysed using Spearman correlations, ordinal logistic regression, and parallel mediation analysis.
Results
CRM (β = 0.717, p = 0.002) and CKM (β = 0.581, p = 0.009) significantly predicted customer engagement, while MDM’s direct effect was positive but non-significant (β = 0.346, p = 0.071).
Takeaways & Limitations
CRM and CKM emerge as the principal direct drivers of customer engagement, while MDM is positioned as a foundational data-quality component within CRI.
Takeaways & Limitations
The sample was severely underpowered for small effects, so findings are preliminary and require replication with larger samples.
Abstract
from arXiv · showhide
This study examines how Customer Relationship Management (CRM), Master Data Management (MDM), and Customer Knowledge Management (CKM) jointly constitute a Customer Relationship Intelligence (CRI) framework for enhanced Customer Engagement (CE). A cross-sectional survey of 100 participants across retail, healthcare, IT, and telecommunications sectors was analysed using Spearman rho correlation and ordinal logistic regression (IBM SPSS). Bivariate correlations were weak and non-significant (r<0.19, p>0.06). Regression identified CRM (beta=0.717, p=0.002) and CKM (beta=0.581, p=0.009) as significant positive predictors of CE; MDM showed a positive but non-significant direct effect (beta=0.346, p=0.071). The model explained 20.5% of CE variance (Nagelkerke R^2=0.205). Parallel mediation analysis (Hayes PROCESS Model 4, 5,000 bootstrap samples) found no significant indirect effects of MDM on CE via CRM (IE=0.021, 95% BC CI [-0.072, 0.121]) or CKM (IE=0.032, 95% BC CI [-0.061, 0.126]); Hypothesis H4 was not supported. CRM and CKM emerge as the principal drivers of CE within the CRI framework, while MDM functions as a foundational data quality enabler whose strategic value is realised through its enabling effect on CRM execution and knowledge management. Findings should be treated as exploratory given the sample size and cross-sectional design. Future research should replicate with larger sector-specific samples and longitudinal designs, particularly in regulated BFSI contexts where MDM architecture is shaped by data governance mandates.
I. INTRODUCTION
CRM, MDM, and CRI are presented as integrated approaches for managing customer relationships, improving data quality, and strengthening customer engagement. The introduction frames CRM–MDM integration as a response to siloed systems and poor data quality.
- CRM: CRM combines customer acquisition, retention, marketing, sales, delivery, and relationship-management activities.It is described as a broad organisational strategy for creating value for customers and the organisation.
- CRM: CRM is used to acquire, engage, retain, and re-engage customers while reducing marketing and customer-service expenses.Successful implementation requires alignment among strategy, organisational structure, segmentation, technology, production, and management processes.
- MDM: MDM synchronises key customer data across corporate systems to improve data quality and support data-driven decision-making.It provides a dependable source for important organisational data assets.
- CRI integration: Integrated CRM and MDM are positioned as a response to siloed systems and poor data quality in customer-engagement operations.The framework combines CRM’s interactive role with MDM’s data-governance capabilities.
- Study focus: The study examines how each CRI component contributes to effective customer engagement through a structured quantitative approach.The introduction identifies CRM–MDM integration as a strategic focus and frames customer engagement as the outcome of interest.
II. LITERATURE REVIEW
The literature review presents CRM–MDM integration, CKM, and analytics as complementary foundations for customer engagement. It links centralised customer data and customer knowledge with personalisation, prediction, and service improvement.
- CRM–MDM integration: CRM manages customer interactions while MDM centralises and cleans data to create a single customer view.The literature associates their integration with improved analytical capability, retention, and engagement.
- CKM: CKM combines CRM and knowledge management to connect customer-interaction data with actionable insights.Customer feedback, complaints, and suggestions can feed innovation and service improvements.
- Emerging technologies: IoT, real-time analytics, and AI support continual customer-data collection, demand prediction, and proactive retention strategies.The review also describes chatbots and virtual assistants as tools for faster responses and improved customer satisfaction.
- Analytics: Big Data Analytics enables CRM systems to process large volumes of consumer data and develop personalised contact points.Predictive analytics and sentiment analysis are described as tools for tailoring experiences and understanding feedback.
E. Behavioural Analysis in CRM
The behavioural-analysis discussion presents CRM as increasingly data-driven, with CRI extending this orientation through customer insights, prediction, real-time integration, and behavioural scoring.
- Behavioural Analysis in CRM: CRM is moving toward data mining to obtain actionable insights from customer data for engagement and loyalty.The discussion also identifies digital transformation and early disengagement detection as related research directions.
- Customer Relationship Intelligence: CRI collects, analyses, and applies customer-relationship insights to enhance interactions and build stronger relationships.Its stated purpose is to derive meaningful patterns from customer interactions rather than merely store data.
- Behavioural Analysis in CRM: CRI uses multidimensional insights and behavioural scoring to forecast customer behaviour and support personalised engagement strategies.Inputs include purchase history and browsing behaviour for targeted marketing and resource allocation.
- Real-time intelligence: Real-time data integration enables organisations to adapt to customer needs while building trust and satisfaction.The framework is described as relying on IoT and cloud technologies to support this responsiveness.
I. Impact of CRI on Business Performance
CRI is presented as a route to higher customer satisfaction, loyalty, profit, and operational efficiency through analytics, real-time data, and responsive customer management. The section also states the study’s hypotheses for testing component effects and mediation.
- Business performance: CRI can tailor customer experiences, strengthen client relationships, and support customer satisfaction, loyalty, and profit.The discussion links advanced analytics and real-time data with more responsive customer experiences.
- Business performance: CRI-based marketing actions are described as improving resource-investment conversion rates and revenue.The framework is also associated with stronger sales, operational efficiency, and return on investment.
- Business performance: CRI is framed as enabling stronger sales, operational efficiency, and ROI through integrated customer intelligence.These outcomes are presented as consequences associated with CRI-based customer-management practices.
- Research hypotheses: The study tests whether CRM, MDM, and CKM have positive effects on customer engagement.H1–H3 are evaluated using ordinal logistic regression with p<0.05 as the support criterion.
- Research hypotheses: H4 tests whether MDM’s effect on customer engagement is mediated through CRM and CKM in a parallel indirect pathway.Support requires a non-zero bootstrapped indirect effect whose 95% bias-corrected confidence interval excludes zero.
L. India/BFSI Regulatory Context
The selected passages describe the study’s research design and instrument reliability, while the supplied BFSI regulatory-context passages outline a regulated setting without reporting specific findings here.
- The supplied BFSI passages identify regulatory requirements surrounding MDM architecture and consent-based CRM personalisation and CKM analytics.
- The study used a cross-sectional quantitative design to examine CRM, MDM, and CKM effects on customer engagement.
- Cronbach’s α = 0.809 confirmed acceptable internal consistency above the 0.70 threshold.
B. Data Collection
The study collected online survey data from professionals across four sectors and analysed construct relationships, predictive effects, common method bias, and mediation using IBM SPSS.
- A self-administered online survey collected data from professionals in retail, healthcare, information technology, and telecommunications.The a priori minimum sample size was 74, based on power = 0.95, α = 0.05, and a medium effect size f 2 = 0.15.
- IBM SPSS Statistics was used to analyse the data.
- Spearman’s rho examined associations among CRM, MDM, CKM, and CRI, which were generally weak with r<0.20.
- Ordinal logistic regression tested CRM, MDM, and CKM predictive power for customer engagement and business performance.
- Harman’s single factor test assessed whether common method variance could explain the observed relationships.All 16 survey items entered an unrotated one-factor principal components analysis, with less than 50% variance treated as evidence against substantial common method variance.
- Parallel mediation used Hayes’ PROCESS Model 4 with 5,000 bootstrap samples and 95% bias-corrected confidence intervals to test H4.
A. Frequency Distribution of Respondents’ Demographic Details
The respondent pool was demographically diverse, with balanced gender representation, varied qualifications and experience, and participation across multiple industries.
- The 25–34 age group was the largest, representing 24% of respondents.
- Gender representation was 53% female and 47% male.
- High school diplomas and Ph.D.s were the most prevalent qualifications, at 22% and 19%, respectively.
- Respondents with more than 10 years of experience formed the largest experience group at 28%.
- Telecommunications and healthcare were the most represented industries, at 19% and 15%, respectively.
C. Influence of CRM, MDM, and CKM on Customer Engagement
The ordinal regression model significantly improved on the intercept-only model, explained 20.5% of customer-engagement variation, and identified CRM and CKM as significant positive predictors while MDM was not significant.
- χ2(3) = 20.913, p < 0.001 showed that the final model fit significantly better than the intercept-only model.
- Nagelkerke R2 = 0.205 indicated that the model explained approximately 20.5% of variation in customer engagement.
- CRM had a positive significant effect on customer engagement, with β = 0.717 and p = 0.002, supporting H1.
- CKM had a positive significant effect on customer engagement, with β = 0.581 and p = 0.009, supporting H3.
- MDM had a positive but non-significant direct effect, with β = 0.346 and p = 0.071, so H2 was not supported.
D. Influence of CRM, MDM, and CKM on Business Performance
The Business Performance model showed borderline adequate fit and modest explanatory power, with CKM the only statistically significant predictor among the reported coefficients.
- χ2 = 195.567, p = 0.052 for Pearson goodness-of-fit was marginally non-significant, while deviance χ2 = 144.252, p = 0.876 indicated adequate specification.
- Nagelkerke R2 = 0.118 indicated that the Business Performance model explained approximately 11.8% of variation.The result was described as modest explanatory power in an exploratory cross-industry study.
- CKM (β = 0.489, p = 0.021) was statistically significant, whereas CRM (p = 0.085) and MDM (p = 0.066) were marginally significant.
E. Common Method Bias Test and Mediation Analysis
The study found limited evidence of common method variance and no statistically significant indirect effect of MDM on CE through CRM or CKM. These results leave CRM and CKM as stronger direct predictors while highlighting mixed model fit and the need for broader validation.
- Common Method Bias Test: 27.7% of total variance was explained by the single factor in Harman’s test, below the 50% threshold for dominant common method variance.
- Mediation Analysis: MDM’s indirect effect on CE via CRM was 0.021, with 95% BC CI [−0.072, 0.121], which contained zero.
- Mediation Analysis: MDM’s indirect effect via CKM was 0.032, with 95% BC CI [−0.061, 0.126], and the total indirect effect was 0.053, with 95% BC CI [−0.075, 0.177].
- Model Interpretation: The CE model showed mixed fit, with Nagelkerke R2 = 0.205 and Pearson χ2 = 197.018, p = 0.045 indicating modest explanatory power and possible residual misspecification.
- Mediation Analysis: Neither mediation pathway reached statistical significance, so Hypothesis H4 was not supported.
- Implications: Future research should use larger sector-specific samples, longitudinal designs, and objective outcome measures, particularly in regulated BFSI contexts.
VI. LIMITATIONS AND FUTURE RESEARCH DIRECTIONS
The study’s limitations constrain statistical power, measurement validity, generalisability, and causal interpretation. Future research should use larger, more diverse, longitudinal, and better-controlled designs.
- Sample size and statistical power: At n = 100, achieved power was approximately 0.46 for the maximum observed correlation, so findings remain exploratory and require larger-sample replication.Detecting r = 0.187 with 80% power requires n≈223, while detecting r = 0.136 requires n≈423.
- Common method variance: Single-source, same-survey measurement creates potential common method bias despite Harman’s test finding 27.7% variance below the 50% threshold.The authors recommend temporal separation, objective CRM or CE metrics, or multi-informant designs.
- Sample composition and generalisability: Online convenience sampling across four industries limits generalisability, especially to capital-intensive and regulated banking and financial services contexts.Within-industry differences in CRM maturity, MDM investment, and regulatory context were not controlled.
- Cross-sectional design and causal inference: The cross-sectional design establishes predictive relationships but cannot establish causal direction from a single-time-point survey.Longitudinal panels or natural experiments are needed to establish temporal precedence and support causal claims.