Cluster analysis segments a customer database so that customers within segments are similar, and collectively different from customers in other segments. Similarity is in terms of the “clustering variables,” which may be psychographics, demographics, or transaction measures such as RFM. The clusters often have rich interpretations with strong implications for which customers should be targeted with a particular offer or marketed to in a certain way. This chapter discusses the details of cluster analysis, including measures of similarity, the major cluster methods, and how to decide upon the number of clusters and their interpretation.
KeywordsCluster Analysis Output Layer Output Unit Cluster Variable Cluster Membership
Unable to display preview. Download preview PDF.