The Design And Implementation Of Customer Churn Management System In Distribution Industry Based On Decision Tree Technology

Among all CRM activities in modern enterprises, customer churn analysis plays a more and more important role, as customer value is more recognized by the enterprises. The customer churn analysis, on the one hand, provides a clear understanding of customers, which is very important to decide different customer management strategies for different customer types or different customer maturity statuses. On the other hand, the customer churn analysis provides a solid base for other CRM activities like cross-selling and marketing campaign promotions.This article firstly discusses how to use data mining decision tree technology and tools, to analyze customer churn issue and predict customer churn trend in distribution industry, in turn to build a customer churn prediction decision tree. Then, apply different input parameters to conduct a quantitative analysis on the generality and accuracy of the customer churn prediction decision tree. Afterwards, study how to build a customer churn management system, to firstly automate all data mining activities of customer churn analysis and prediction processes, secondly to manage all the tracking processes of customer retainment activities. How to resolve and implement several key technical puzzles of customer churn management system is discussed as well, including the storage of decision tree result, dynamic update of decision tree, and the integration with workflow engine system.At last, the article introduces the operational statistics of the up-and-running customer churn management system, and summarizes the improvement of customer churn management with the system, and also provides a visional insight over further research aspects in customer churn analysis and prediction.