New RFM-D classification model for improving customer analysis and response prediction

Abstract

Customer segmentation is seen as one of the pillars of a successful advertising campaign. Marketers give great importance to this flagship phase in the process of marketing new products. Successful segmentation will involve successful “Customer Targeting” and therefore a profitable customer marketing campaign. Many works have dealt with customer segmentation using unsupervised Machine Learning algorithms such as K-Means by applying the famous Recency, Frequency and Monetary model. That model suffers from insufficiency by ignoring other important parameters according to the field of application. In this paper, we have modified the model by adding diversity “D” as a fourth parameter, referring to the diversification of products purchased by a given customer. The segmentation based on RFM-D is applied in a retail market in order to detect behavior patterns for a customer. The proposed model increases the quality of prediction of customer behavior; Companies could predict, customers who will respond positively.

Keywords

• RFM & RFM-D models
• Diversity
• K-Means algorithm
• Customer segmentation
• Customer response prediction and targeting optimization