<![CDATA[Customer segmentation is applied to sales transaction data from a sheet plastic manufacturing company covering the 2019–2025 period and obtained from the company’s Enterprise Resource Planning (ERP) system. This study aims to identify heterogeneous customer characteristics and generate actionable market segments based on Recency, Frequency, and Monetary (RFM) values using the K-Means Clustering algorithm. The methodology comprises systematic data cleaning, transaction aggregation, RFM calculation, feature normalization, cluster modeling, and determination of the optimal number of clusters through the Elbow Method and Silhouette Score. The final dataset consists of 46,372 transactions involving 1,223 active customers, with a cumulative transaction value of IDR 722.4 billion. The findings reveal five optimal clusters, validated by a Silhouette Score of 0.513, indicating reasonably good clustering quality and meaningful separation among customer segments. The segmentation identifies five distinct customer groups: Low Engagement (50%), characterized by limited transaction activity and requiring targeted reactivation strategies; Churn (37%), representing long-inactive customers at significant risk of disengagement and requiring structured re-engagement programs; Potential (11%), comprising active customers with substantial transaction values and strong development opportunities; Key Account (less than 1%), representing strategically important customers with the highest business contribution and requiring prioritized relationship management; and High Value (2%), consisting of loyal, profitable customers who should be retained through personalized loyalty initiatives. These findings demonstrate that integrating RFM analysis with K-Means Clustering provides a practical, data-driven approach to understanding customer heterogeneity in manufacturing markets. The resulting segmentation framework offers a strategic foundation for targeted retention initiatives, personalized promotional campaigns, improved customer relationship management, optimized allocation of sales resources, and more effective managerial decision-making based on measurable customer behavior and long-term value]]>
