The distribution-based recency, frequency and monetary (RFM) score segmentation method: A novel RFM model for enhanced marketing strategies
Abstract
Customer segmentation is important for marketing and campaign management, and the RFM (recency, frequency and monetary) model is widely used for this purpose. However, existing algorithms typically segment customers based on overall similarity in R, F and M scores, which may not provide the most meaningful insights for retailers. A more effective approach involves grouping customers according to the distribution of their R, F and M scores rather than clustering those with similar values. To address this limitation, this paper introduces the DbRFMSS (distribution-based RFM score segmentation) method. The proposed method categorises customers based on the distribution of R, F and M scores, offering more actionable segmentation. While many tools, including Excel and other applications, can perform RFM analysis, the DbRFMSS method leverages machine learning algorithms to handle large datasets more precisely. Comparative analysis with the traditional RFM approach demonstrates that the proposed method performs better in terms of computation time, iteration efficiency and silhouette measures. This article is also included in The Business & Management Collection which can be accessed at https://hstalks.com/business/.
The full article is available to subscribers to the journal.
Author's Biography
A. Joy Christy is an assistant professor in the SASTRA Deemed to be University’s School of Computing. Her research areas include data analysis and data mining, Big Data, image segmentation and convolutional neural networking. She has written numerous articles on statistical analysis and data mining, machine learning, medical imaging, image segmentation and software metrics.
A. Umamakeswari is a professor and Dean of the School of Computing at SASTRA Deemed to be University. Her research focuses on wireless sensor networks, cloud computing, Internet of Things, machine learning, embedded systems and blockchain. She has contributed to various projects funded by Indian government institutions, including ones focused on content development for software testing, microprocessors and interfacing and the design of logic tools for process modelling. She has guided research on the development of intrusion detection systems for wireless sensor networks and industrial cyber-physical systems, computation offloading for edge environments, cloud security, blockchain-based solutions for the Internet of Things and the use of machine learning to predict air quality in urban environments. She has written numerous papers on the Internet of Things, machine learning and cyber-physical systems.
S. Sri Laxmi Narasimhan Sri Laxmi Narasimhan Sridharan is a post-graduate researcher, having completed his Bachelor's degree in Computer Science Engineering. His areas of research include AI/ML models, game theory approaches, feature engineering and combinatorial topology.
Citation
Christy, A. Joy, Umamakeswari, A. and Sri Laxmi Narasimhan, S. (2025, December 1). The distribution-based recency, frequency and monetary (RFM) score segmentation method: A novel RFM model for enhanced marketing strategies. In the Applied Marketing Analytics: The Peer-Reviewed Journal, Volume 11, Issue 3. https://doi.org/10.69554/YQPB8103.
Publications LLP