Correspondence: Mary M. Walker, Williams College of Business, Xavier University, 3800 Victory Parkway, Cincinnati, OH 45207, US. Tel: +1 513 745 2980; e-mail: walkerm@xavier.edu

¹is owner-operator of S2 Statistical Solutions, Inc. founded in 1992, and performs statistics consulting and training seminars. Dr Gunnarsson has taught statistics courses for more than ten years in the fields of education, psychology, and business at Xavier University, The Union Institute, and University of Cincinnati.

²is Professor of Marketing at Xavier University in Cincinnati, Ohio, and in this role was the recipient of a grant sponsored by the General Electric Company to develop a course on data mining. She has served as Xavier's Acting Vice President for Information Resources, overseeing all information technology during 2004–2005.

³is a chemical engineer and chemist with experience as a SAS programmer and data analyst. He currently consults on topics of survey analysis, GIS, and mapping software.

⁴is Senior Data Analytical Manager with a background in software management and market research. He has organised three successful annual SAS one-day conferences for statisticians and researchers.

Received 19 July 2007; Revised 19 July 2007.

Abstract

Many organisations across a variety of industries are engaging in the process of data mining as part of an overall strategy for business intelligence, customer relationship management (CRM), including churn prevention. This paper provides an overview of the data mining process and illustrates a case study in which data mining is utilised as a churn prevention tool for a major Midwest USA newspaper. For this case study, a decision tree, a common modelling technique, was the analytical tool of choice. Lessons learned throughout the data mining process are provided to offer insight and to promote the sharing of information. Strategies for getting started in the data mining process are presented to encourage organisations to embrace a data-driven strategy for business intelligence, CRM and churn prevention.