| Summary: | <p><em>This study aims to measure customer </em><em>purchase</em><em> behavior by using </em><em>RFM</em><em> model with K-Means, silhouette Index method. Using historical data by </em><em>Cendana</em><em> </em><em>Perdana</em><em> </em><em>Perkasa</em><em> Distributor to data this study. The company has never to measured customer buy behavior by this model. The results of this study showed that there 3 cluster with 0,589 silhouette Index. Based on the highest silhouette Index then the cluster is divided into 3 segment with some segments 1 is 57 customers, segment 2 is 257 customers and segment 3 is 8 customers. The best </em><em>purchase</em><em> value is segment 3 with symbolized byR↓F↑M↑ that means this segment has high frequency and monetary as well as low regency. The segment has potential customers for the future. The worst </em><em>purchase</em><em> value is segment 1 withv </em><em>symbolised byR↑F↓M↓ that means this segment has low-frequency and monetary as well as high regency. The segment has an unclear or uncertain customer will be a potential customer.</em><em></em></p><p><strong><em> </em></strong><strong><em>Keywords</em>: </strong><em>Customer Purchase Behavior, K-Means, RFM and Silhoutte Index</em></p>
|
|---|
