Customer Value Assessment Methodology Using DM Approach

The purpose of this study is to segment customers by value and observe different various characteristics in different clusters. We propose a new segmentation method based on DM and most commonly used CRM models; RFM, and demographic variables. The method is based on two-phases clustering model by k-means and SOM techniques and has been implemented in chain stores B in Iran. The descriptive findings of the study were rated clusters and pattern types of these customers to identify the target customers’ positions. The existing customers were divided into 35 groups.In this chain store, each customer has a transaction record that stored in the store’s database but for Demographic data they were asked telephonic. Beyond simply understanding customer value in each cluster, the chain store would gain the opportunities to establish better customer relationship management strategies, improve customer loyalty and revenue and find opportunities for up and cross selling.