Abstract: This paper presents methodologies for creating profiles of typical customers for certain types of services and products in user-oriented businesses. Because of the specific characteristics of each of the domains where these methodologies would be applied creating one solution for all purposes does not solve the problem. The emphasis of this paper is on experimenting with real data from telecommunication sector and on the construction of clustering system for user profiling. The approach for solving this problem involves analysing the characteristics of the observed data and adjusting the process of intelligent data analysis according to its specifications.