Research On RFM Multi-level Customer Value Model Based On Product Classification And Its Application

Nowadays,In the buyer's market,retail enterprises are facing fierce competition in the market.It's urgently in need that retail enterprises have to identify valuable customers and meet their variety requirement.In the period of large data,enterprises gets much data of customer's transactions.Based on these data,we can use data mining technology to give reasonable customer segmentation and to identify group behavior feature of different customers,It will be a trend to provide more detailed and specific guidance for the management decision-making.The RFM model,which was originally used in direct sales area,is now widely used in retail,banking,and communications.When the traditional RFM model used in customer segmentation of retail enterprises,all products are treated as a whole to evaluate customer value and customer loyalty.Then its classification results can not distinguish different customers for different series of products of customer value.Thus it cannot provide specific guidance for the precise marketing.Customer value types are different from different customers.There are also differences in customer value types for the same customer.The nature of the series of products is different.However,we generally regard all the products of retail enterprises as a whole to divide the customer value types.This will miss some important customer information.At the same time,we can provide clear guidance for management decision-making only if we know the customer value types of various products.For these two reasons,this paper creates a RFM multi-level customer value model based on product classification.In the model,The problem of product classification standard is solved,and the application significance of the model is expounded.In this paper,we first establish a mathematical model that verifies the necessity and usefulness of the product classification when measuring customer value.Then,the product classification standard in the RFM multi-level customer value model is studied by referring to the product classification criteria of the telecommunication industry and the bank in the evaluation of customer loyalty,and combining the product classification basis of the mathematical model.Finally,combined with a large cosmetic retail company for one year of customer sales data,validated the established model to measure customer value of the effectiveness and effectiveness.The results of the application of the model show that customers with the same customer value may have a large difference in the customer value of different products.The model can identify the customer value types of different series for the customers with the same customer value type.The underlying RFM multi-level customer value model can identify 20% of high-value customers 'different behavioral characteristics,namely,the high value customers' customer value types are different from products;the model can provide more specific guidance for the company's managers to maintain high-value customers.