As markets have become increasingly saturated, companies have acknowledged thattheir business strategies need to focus on identifying those customers who are mostlikely to churn. To address this, a method is required that can identify these customers,so that proactive retention campaigns can be deployed in a bid to retain them. Tofurther complicate this, retention campaigns can be costly. To reduce cost andmaximise effectiveness, churn prediction has to be as accurate as possible to ensure thatonly the customers who are planning to switch their service providers are being targetedfor retention.Current techniques and research as identified by literature focus primarily on theinstantaneous prediction of customer churn. Much work has been invested in thismethod of churn prediction and significant advancement has been made. However oneof the major drawbacks of current research is that the methods available do not provideadequate time for companies to identify and retain the predicted churners. There is alack of time element in churn prediction. Current research also fails to acknowledge theexpensive problem of misclassifying non-churners as churners. In addition, mostresearch efforts base their analysis on customer demographic and usage data that canbreach governing regulations. It is proposed in this research that customer complaintsand repairs data could prove a suitable alternative.The doctoral research presented in this thesis aims to develop a customer profilingmethodology for predicting churn in advance, while keeping the misclassification levelsto a minimum. The proposed methodology incorporates time element in the predictionof customer churn for maximising future churn capture by identifying a potential loss ofcustomer at the earliest possible point. Three case studies are identified and carried outfor validating the proposed methodology using repairs and complaints data. Finally, theresults from the proposed methodology are compared against popular churn predictiontechniques reported in literature. The research demonstrates that customers can beplaced into one of several profiles clusters according to their interactions with theservice provider. Based on this, an estimate is possible regarding when the customercan be expected to terminate his/her service with the company. The proposedmethodology produces better results compared to the current state-of-the-art techniques.
展开▼
