Stability of market segmentation with cluster analysis – A methodological approach

• We analyse the stability of market segments based on cluster analysis over time.
• We focus on the analytical process of decision making in cluster analysis.
• We determine cluster stability over time regarding number, size and properties.
• Cluster structure may change within a few years.
• Stability of market segments should not be taken for granted.

Market segmentation is a very popular marketing tool. In the food sector, the characteristics of different consumer attitudes and consumption habits are often used as the basis for segmentation. However, the success of a target-oriented marketing approach to selected groups of consumers depends on the results of the methodology applied. So far, relatively little attention has been paid to the reliability of the analysis used for attitude-based market segmentation, to the validity or internal stability of results or to the dynamic stability over time with regard to number, size and properties of the segments.In our study, we used data from a panel of more than 10,000 German households. The participants were segmented using a statement battery and the application of cluster analysis. In order to ensure an internally stable cluster solution, our focus was on the analytical and technical process of decision making when clustering a large dataset. A combination of various statistical measures was applied in order to enable objective decision making in the determination of the optimal number of clusters. The dynamic stability of the resulting segments was determined by confirmatory cluster analyses using data from the same individuals in three subsequent years.The results of the analyses show that neither the internal nor the dynamic stability of market segments should be taken for granted. Therefore, marketers face the challenge of designing segment-specific marketing strategies in a way that allows changes in consumer preferences to be integrated.