CUSTOMER GRIEVANCE REDRESSAL IN E-RETAILING: A DATA MINING APPROACH

Abstract

In the rapidly evolving e-retail sector, efficient customer grievance redressal is crucial for maintaining competitive advantage and customer loyalty. This study introduces a novel approach by leveraging data mining techniques, specifically clustering and causal analysis, to revolutionize the grievance redressal process in e-retailing. Using a comprehensive synthesized dataset of 1,000 customer grievances, we implemented advanced text preprocessing methods and applied clustering algorithms such as K-means to categorize grievances into meaningful clusters. Our innovative integration of causal analysis with clustering results revealed systemic issues, providing actionable insights into the underlying causes of customer dissatisfaction. This research not only uncovers critical patterns in customer complaints, but also proposes targeted interventions to address these issues effectively. The expected contributions of this study extend to various stakeholders, including e-retailers, who can enhance their customer service strategies; data scientists, who gain insights into the application of advanced data mining techniques; and customers, who benefit from improved service quality and satisfaction. By addressing common grievances related to delivery, product quality, and service efficiency, this research offers a transformative framework for elevating the standard of customer grievance redressal in the e-retailing industry. In future work, the researcher plans to employ machine learning to develop a hybrid predictive model, further enhancing the capability to identify and address potential grievances preemptively.