As information technology advances, too much must be generated, stored, and pre-processed to meet the demands of various companies. One marketing tool that might assist a company in promoting and gaining from sales efforts is segmentation. Understanding the idea of predictive modelling and knowing a way to leverage BD (Big Data) for segmentation is crucial for marketing professionals. The boundaries between various fields are becoming less defined, and there is an increasing amount of field interlinking. The goal of the project is to develop market and customer segments based on predictive modelling with BD that is already there within the organisation. A complex prediction model can be used to construct client segments, according to the findings of an empirical investigation. Market segmentation recognises that not all customers have the same interests, buying habits, or preferences. Market segmentation seeks to increase the strategic and targeted nature of a company's marketing initiatives in contrast to general coverage of all potential customers. A company can improve its chances of making sales and better manage its resources by creating specialized strategies for certain commodities in relation to target groups. Customer segmentation, which is the procedure of making cluster of same type thinking people. The clustering technique aids in improving comprehension of customers in terms of both static demographics and dynamic behaviour. The simple yet effective RFM approach may be used to segment the market. RFM analysis is used to examine consumer behaviour, including how lately (recency), frequently (frequency), and financially (spend) the customers make purchases (monetary). In this study, data mining has been used to group products into categories based on recent sales, frequent sales and total amount spent. This study has put out a brand-new k-Means methodology for RFM analysis. For e-commerce platforms, consumer segmentation may successfully save marketing costs while also increasing customer happiness. The output measure is compared with the existing RFM models.