A case study of applying LRFM model and clustering techniques to evaluate customer values

Abstract

Abstract A case study of applying LRFM (length, recency, frequency, and monetary) model and clustering techniques in evaluating an outfitter’s customer values is presented. Self-organizing maps is first used to determine the best number of clusters and then K-means method is applied to classify 551 customers into twelve clusters when L, R, F, and M are the segmenting variables. The results show that Cluster 5 might be the most important cluster because the average values of L, R, F, and M are well above the averages. In contrast, the customers in Clusters 7 and 10 have low contributions since L, R, F, and M values are below the average values. As a result, with the applications of LRFM model and clustering techniques, the outfitter can allocate and utilize resources effectively and efficiently to first identify high-value and profit potential customers and then design different marketing strategies to maximize its profits for different types of clusters.