Customer Clustering Based on Customer Lifetime Value: A Case Study of an Iranian Bank

Customer lifetime value (CLV) as a quantifiable parameter plays an important role in customer clustering. Clustering based on CLV helps organizations to form distinct customer groups, reveal buying patterns, and create longterm relationships with their customers. Our research aims at the synthesis of a CLV model and a clustering algorithm in a new comprehensive framework. First, a model for calculation of CLV is suggested, which is called Group LRFM or GLRFM briefly. In this model, four parameters, Length, Recency, Frequency, and Monetary, are determined according to the products/services used by customers. Then, a novel framework based upon the model is presented in eight steps for customer clustering. In traditional methods, the customers of valuable cluster are treated the same. But in proposed framework, company can design different and proper strategies for each cluster based on the use of products/services. The experimental results in banking industry verify that proposed approach allows an accurate and efficient cluster analysis; it provides appropriate information to create clear sales and marketing policies for three identified segments.