CUSTOMER CLUSTERIZATION AS A SUPPORT OF CUSTOMER RELATIONSHIP MANAGEMENT AT PT. SIP (PRIME INSTALLATION CERTIFICATION) WITH K-MEDOIDS ALGORITHM

Abstract

This thesis aims to investigate the implementation of customer clustering techniques as a crucial component in developing Customer Relationship Management (CRM) strategies at PT. SIP (Sertifikasi Instalasi Prima). CRM is a highly important approach in managing customer relationships, and customer clustering can aid in understanding different customer behavior patterns. In this study, the K-Medoids algorithm is employed to group customers based on their characteristics and purchasing patterns. The research methodology involves collecting customer data from PT. SIP and processing the data for clustering analysis preparation. The subsequent steps involve applying the K-Medoids algorithm to form homogeneous customer clusters. The clustering results are integrated into the existing CRM strategy, enabling PT. SIP to provide services more effectively tailored to the needs and preferences of each customer group. The findings of this study offer valuable insights into various customer purchasing patterns and preferences. The clustering outcomes assist in identifying the most valuable and potential customer groups, providing a better understanding of how PT. SIP can enhance their interactions and services. By incorporating clustering approaches into the CRM strategy, the company can improve customer retention, enhance customer satisfaction, and optimize overall customer value.