Using RFM-R Analysis for Effective Customer Segmentation in Bank Marketing

In present time, people are increasingly conducting financial transactions via mobile banking. Online marketing is crucial in the banking industry to attract customers and maintain existing customer loyalty. Choosing customer group targets is an important part of developing a marketing plan using a customer-centric approach. In addition to focusing on customer needs, it is also important to consider the revenue that the bank will receive from customers. Recency, Frequency, and Monetary (RFM) approach is generally employed for customer segmentation. This paper focuses on segmenting mobile banking customers using the RFM-R model. The RFM-R model is an extended version of RFM model that introduces the “R” variable, which “R” represents the revenue generated by customers for the bank in each transaction. We use the K-Means algorithm for customer segmentation and the Silhouette Index and Elbow method to determine the optimal K value. The dataset consists of 685,205 transactions from 50,497 unique customers of one of commercial banks in Thailand starting in January 1, 2022 to December 31, 2022.