The Comparison of Customer Segmentation Methods in the Perspective of Manager's Aims

The paper includes the analysis of different methods of customer segmentation form the perspective of two manager’s aims: to segment new unknown market and to predict to which cluster the customer should be attributed to. The topicality of the current issue might be explained in the following way: despite the fact that there are more and more methods of segmentation that appear and market is quite dynamic, there is no clear understanding which exact method manager should use to solve two different issues. The aim of the research is to analyze existing methods of customer segmentation, to evaluate how they can be applied in solving two main manager’s issues and to compare and contract chosen methods empirically using real data of coffee industry. Cluster analysis and latent-class analysis as well as discriminant analysis and CHAID are compared. The first chapter includes the definition of the segmentation, its assumptions, school of thoughts and its influence on company’s performance. Proceeding chapter contains the description of segmentation methods and bases as well as classification of those methods in the perspective of manager’s issues. Empirical comparison of chosen methods using coffee customers database is provided in the third chapter. Finally, the author gives some weaknesses of the current paper which might be improved in the next research. The paper gives recommendations to the manager of what methods to use in which situation and provides empirical base for further research in the current field of investigations.