Customers are the most valuable assets of a company. As Internet becomes popular, web personalization for E-commerce has gained much attention. Building and maintaining customer loyalty are important issues in electronic business. By providing customer services, sharing cost benefits with online customers, and rewarding the most valued customers, customer loyalty and customer equity can be improved. Attempt has been made to propose Adaptive Neuro Fuzzy classification as a multidimensional data analysis for establishing profitable customer relationships. With the advent of Internet technology, number of online customers and number of attributes of these customers keeps on varying with time and situation which is very difficult to predict. This dynamic ever changing behavior of customers can be modeled by hybridization self learning, adaptive neural network along with fuzzy logic. Customer performance indicators for revenue and profitability, customer investment, relationship, recommendation, information and cooperation indicators allow to segment customers precisely to define customer strategies in order to increase corporate profits and growth.