The efficient segmentation of customers of an enterprise is categorized into groups of similar behavior based on the RFM (Recency, Frequency and Monetary) values of the customers. The transactional data of a company over is analyzed over a specific period. Segmentation gives a good understanding of the need of the customers and helps in identifying the potential customers of the company. Dividing the customers into segments also increases the revenue of the company. It is believed that retaining the customers is more important than finding new customers. For instance, the company can deploy marketing strategies that are specific to an individual segment to retain the customers. This study initially performs an RFM analysis on the transactional data and then extends to cluster the same using traditional K-means and Fuzzy C- Means algorithms. In this paper, a novel idea for choosing the initial centroids in K- Means is proposed. The results obtained from the methodologies are compared with one another by their iterations, cluster compactness and execution time.

INTRODUCTION

In recent years, there has been a massive increase in the competition among firms in sustaining in the field. The profits of the company can be improved by a customer segmentation model. Customer retention is more important than the acquisition of new customers. According to the Pareto principle [4], 20% of the customers contribute more to the revenue of the company than the rest. Customer segmentation can be performed using a variety of unique customer characteristics to help business people to customize marketing plans, identify trends, plan product development, advertising campaigns and deliver relevant products. Customer segmentation personalizes the messages of individuals to better communicate with the intended groups. The most common attributes used in customer segmentation are location, age, sex, income, lifestyle and previous purchase behavior. Here, segmentation is done using behavioral data since it is commonly available and continuously evolving with time and purchase history. RFM (Recency, Frequency, and Monetary) analysis is a renowned technique used for evaluating the customers based on their buying behavior. A scoring method is developed to evaluate scores of Recency, Frequency, and Monetary. Finally, the scores of all three variables are consolidated as RFM score ranging from 555 to 111 [2] which is used to predict the future patterns by analyzing the present and past histories of the customer. In this context, it has been observed that the scores of three factors Recency, Frequency and Monetary directly proportional to customer’s lifetime and retention.