Applying and Evaluating Models to Predict Customer Attrition Using Data Mining Techniques

As competition intensifies, retaining customers becomes one of the most serious challenges facing customer service providers. Customer attrition prediction models hold great promise as power-ful tools for enhancing customer retention. Several statistical meth-ods have been applied to develop models predicting customer attri-tion. Yet little research is done on the relative performance of mod-els developed by different methods. The lack of knowledge about the performance of various prediction models is more pronounced due to the nonlinear nature of the combined causes of attrition (such as switching to another provider or canceling a service). The development of data mining techniques has made the comparison of prediction power of different models more efficient and easier. In this article we demonstrate how to use data mining techniques and software to fit and compare different customer attrition prediction models, using data from a major telecom service provider