The emergence of electronic commerce has provided customers with more information about market opportunities and choices. This has made customers to become more demanding and likely to switch between suppliers giving birth to the notion of 'churn' (Lejeune 2001). In such an environment with intense competition, customer retention becomes the focal concern ofbusinesses (Colgate and Danaher 2000). This emphasis on retention aspect of CRM comes from the fact that long-term customers buy more and cost less to serve (Ganesh, Arnold, and Reynolds 2000). Additionally, replacing existing customers with 'new' ones is known to be more expensive (Bhattacharya 1998). Regarding this, one straightforward way to handle the customer churn issue is to identify customers who are likely to defect in a close future and persuade them to stay by providing incentives (Neslin, Gupta, Kamakura, Junxiang, and Mason 2006). This calls for models capable of making accurate predictions about consumers' behavior in future. Such models should be able to specify which customers in a dataset have a higher probability to churn in a given future time period.
展开▼
