Stuty On Model Of Customer Classification Based On The Customer Value

With globalization of economy, development of network and easier information exchange within people, market competition of enterprises become more and more severity. The decisive factor for survival and development of an enterprise is the capability of the enterprise for contending for customers, and that the crucial factor of customer acquisition successfully for the enterprise is to know the information clearly of the key customers of the enterprise. Therefore, a reasonable customer classification is a premise condition for improving the relation between the enterprise and its customers. The effective classification of the customer groups redounds to distribute exiguous resource and strengthen the contact between the enterprise and the customer. A competitive advantage is hence obtained actually. Based on the theory, the classification methods and models are investigated in detail in this paper. In the paper, the theory of management decision-marketing and the technologies of data mining and statistical data are used to analyze the variables and models of the customer classification. The maincontent is summarized as follows: (1) Customer value is used as a criterion for the customer classification in the paper. According to the connotation of the customer value, the constitution of the customer value is first analyzed in detail, including the relations between the customer value and the customer current value, the customer potential and the customer loyalty. The theory of constructing model based on the customer current value, the customer potential and the customer loyalty is then provided. The customer current value is expressed using RFM (Recency, Frequency and Money value), while the customer potential value is expressed using the cross-selling capability of the customer.(2) For the analysis of the customer current value using RFM, the customer classification model based on Self Organization Map (SOM) is established. The optimal export frame of the SOM customer classification model is 4×4 through analyzing the SOM model with the export frame of 2×2,3×3,4×4. Each kind of customer is thought having different current value by the analysis of the 16 kinds of customer classified on RFM. Furthermore, the contributing right of the customer current value is analyzed using analytic hierarchy process (AHP). The order of the current value of every kind of customer is also acquired through arrangement.(3) For the cross-selling capability of the customer, which denotes the customer potential value, two understanding are provided. Two cross-selling models are then established based on the Counter Propagation Network (CPN). The investigation indicates that both models can forecast the cross-selling activity efficiently. The numerical analysis methods which can denote the cross-selling capability of the customer are put forward through estimating the two cross-selling models. The numerical value forecast of the cross-selling capability can be acquired through analyzing the cross-selling capability based on the customer maturity and the product degree of the latest product bought by the customer. Combined with the result of the customer classification based on RFM, the customer is classified again and the current value and potential value of each kind of customer are also explained.(4) The customer loyalty is analyzed from the two aspects of sensibility and behavior in the paper and a guideline system of the customer loyalty is set up. The guideline system can be used to count numerical value of the customer loyalty. Finally, a mode based on the CPN is established to predict the customer loyalty.(5) To sum up the theories of the RFM, the cross-selling and the customer loyalty in the customer classification, a customer classification system can be set up base on the customer values (the RFM, the cross-selling and the customer loyalty) and a model is hence established. The value order of each kind of customer classified according to the model can be used to provide a direction for distributing resource of the enterprise reasonable.