The online retail industry is the world's largest growing sector having a vast amount of online sales data to promote and sell their products. However, overloaded data emerge on the customers' side. Customers today expect personalized, relevant information and offers based on their preferences, recent interactions, and the latest product and support experiences. They are ready to take off their journeys with a single poor experience.Companies cannot risk faltering or even provide a below an average interaction at any step along the entire customer journey to affect their revenue. Effective segmentation will help to identify identical and differentiated customers, as proper segmentation will be developed insights inform to the strategic road map which intends to take benefit of profit-driving opportunities with each segment of customers. This could shorten the customer purchase cycle, building customer loyalty, driving higher spend, lowering service, or deepening cross-product and up-selling penetration and support costs. Customer segmentation will target potential customers, by analyzing cross-selling and up-selling product bundling techniques, preferring geographic segmentation, predicting customer lifetime value, and considering historic values for future prediction. The paper discusses four classification algorithms that are comparatively tested to find the best fit classifier to build a personalized recommendation system, giving the customer their personalized products. The study proposes data segmentation modeling and optimizes classification algorithms for driving continuous e-commerce growth. Optimizing the gained segmented values for classification to make the model more precise and accurately segment customers for achieving more effective customer marketing through personalization, to enhance sales by boosting potential customers.