Customer Churn Analysis using Bank's Satisfaction Survey

This case study applied CHAID to a survey of bank customers (CCB). The data file contains demographic information and attribute ratings ( i.e., customer statisfaction) from a sample of 2,200 cutomers from China Construction Bank, Hubei branch. The objective of the study is to examine whether the customer was likely to change banks ( intended churn ) and there is interest in building a model to predict this reponse. An accurate prediction model could be used to target interventions designed to retain customers and would provide descriptions of the customer groups likely to leave. The survey was conducted in Febuary, 2001. One limitation of this study is that the outcome variable is intention to churn, rather than a direct measure of wheather or not the customer actually left the bank, which would be the preferred measure. Another issue is that while demographic and attribute information are available for all customers and, in practice, account activity over time is often used when building churn models.

Questionnaire; Data;

Report is incoming.