<![CDATA[Perkembangan bisnis retail online yang semakin pesat menuntut perusahaan untuk memahami perilaku pelanggan secara lebih mendalam agar dapat merancang strategi pemasaran yang efektif. Penelitian ini bertujuan untuk melakukan segmentasi pelanggan berdasarkan pola transaksi dengan menggunakan metode K-Means Clustering. Data yang digunakan merupakan data sekunder dari Online Retail Dataset yang diperoleh melalui UCI Machine Learning Repository, yang berisi catatan transaksi 4.338 pelanggan dari sebuah toko online di Inggris. Tahapan penelitian meliputi data preprocessing, pembentukan variabel Recency, Frequency, Monetary (RFM), standarisasi data, dan penerapan algoritma K-Means dengan jumlah cluster (k) = 3. Hasil penelitian menunjukkan bahwa pelanggan terbagi ke dalam tiga kelompok utama: pelanggan loyal (0,3%), potensial (74,8%), dan pasif (24,9%). Validitas clustering dikonfirmasi melalui tiga metrik evaluasi dengan Silhouette Score 0,602, Davies-Bouldin Index 0,756, dan Calinski-Harabasz Score 3.124,58. Cluster loyal berkontribusi 18,4% dari total revenue meskipun hanya 0,3% populasi. Penerapan metode K-Means terbukti efektif dalam mengidentifikasi pola perilaku pelanggan yang dapat dimanfaatkan untuk menentukan strategi retensi dan promosi yang lebih tepat sasaran. The rapid growth of online retail businesses requires companies to deeply understand customer behavior in order to design effective marketing strategies. This study aims to perform customer segmentation based on transactional patterns using the K-Means Clustering method. The dataset used is secondary data obtained from the Online Retail Dataset available in the UCI Machine Learning Repository, containing transaction records of 4,338 customers from a UK-based online store. The research stages include data preprocessing, construction of Recency, Frequency, Monetary (RFM) variables, data standardization, and implementation of the K-Means algorithm with the number of clusters (k) set to three. The results show that customers are grouped into three main segments: loyal customers (0.3%), potential customers (74.8%), and passive customers (24.9%). Clustering validity is confirmed through three evaluation metrics with Silhouette Score of 0.602, Davies-Bouldin Index of 0.756, and Calinski-Harabasz Score of 3,124.58. The loyal cluster contributes 18.4% of total revenue despite representing only 0.3% of the population. The application of the K-Means method proves effective in identifying customer behavior patterns that support management in developing more targeted retention and promotional strategies.]]>
