Analisis Segmentasi Pelanggan Menggunakan K-Means Untuk Personalisasi Penawaran Produk Dalam Konteks E-Commerce

Welmi Simanjuntak
Nova Magdalena Br Hombing
Andri Wijaya

Jurnal Sains dan Teknologi
Vol. 2 No. 2 (2025): Oktober - Desember

The rapid growth of e-commerce requires companies to gain deeper insights into customer behavior in order to enhance marketing effectiveness. This study aims to segment e-commerce customers using the Recency, Frequency, and Monetary (RFM) model combined with the K-Means algorithm. The research utilizes an Online Retail Dataset, with stages including preprocessing of 10,001 initial records into 2,311 valid data points, RFM calculation, and Min-Max normalization. The results indicate that the K-Means algorithm successfully groups customers into five clusters with distinct transaction behavior characteristics. Each cluster represents different levels of customer activity, ranging from high-value customers (Cluster 3) to at-risk customers (Clusters 0, 2, and 4). These segments serve as a basis for designing personalized product offerings and improving customer retention strategies in the e-commerce context.