Segmentasi Pelanggan Menggunakan Algoritma K-Means untuk Customer Intelligence dan Strategi Pemasaran pada Perusahaan Retail Online (Studi Kasus: Dataset Online Retail II)

Muhammad Naufal Skha Yusfa (Universitas Pamulang)
Ahyat Musyawwa (Universitas Pamulang)
Aldy Bifal Pratama (Universitas Pamulang)

Journal of Information Systems and Business Technology
Vol 2 No 3 (2026): Journal of Information Systems and Business Technology

Customer intelligence is crucial for online retail companies to design targeted marketing strategies and increase customer retention. This study aims to segment customers based on purchasing behavior using RFM analysis and K-Means clustering. The Online Retail II dataset containing more than 525,000 transactions from December 2010 to December 2011 was used. After data cleaning, outlier handling, and RFM feature engineering, the Elbow Method and Silhouette Score determined the optimal number of clusters (K=3). The results produced three distinct customer segments: Lost/At Risk (25.17%), Regular Customers (74.32%), and High-Value Loyal (0.51%). Actionable business recommendations were formulated for each segment. This segmentation provides deeper customer intelligence and supports more effective marketing strategies for online retail businesses.