Customer Segmentation For Online Retailers Using RFM Modeling To Adopt Profitable Customer-Centric Marketing Strategies

Abstract

Customer Segmentation is the process of segregating customers based on some prominent features that could help online retailers sell more products with less marketing expenses. This process's rationale is the belief that customers exemplify differences in their attitude, behaviour, and demography. An unsupervised machine learning techniques like cluster modelling can develop groups or segments of a customer population. The goal is to create separate groups, but the groups themselves have closely related features, and this can be a powerful means to identify unsatisfied customer problems. In keeping with this data, companies can outperform the competition by developing customized marketing campaigns, designing an optimal distribution strategy, choosing specific product features for deployment, and prioritizing new product development efforts. Thereby, customer segmentation enables a company to customize its relationships and deliver personalized experiences to customers.
This research aims to develop a good understanding of the company's customer base and its behaviour, using the RFM (Recency, Frequency, Monetary) model to accurately classify existing customers' to underscore its most profitable customer groups.
The purpose of this research is to help the company invest in these customers' segments to generate revenue, reduce costs and be profitable.
This research will also predict the items customers will buy in the future using a collection of machine learning techniques: SVM, Logistic Regression, Decision Tree, Random Forest, Adaboost, Ensemble Classifier. This research will essentially categorize the selected customer base of a UK-based online retailer into appropriate customer segments and predict future purchases based on the customer segmentation.