A new taxonomy of differences between consumers is presented. It is argued that there are at least five generic types of differences between consumers: preferences for product benefits, consumer interaction effects, choice barriers, bargaining power, and profitability. Recognition of these distinct types of consumer heterogeneity has implications for the identification of relevant segmentation variables, the methods used to form segments, and the appraisal and optimization of existing segmentations.

Different methods of analysing contingency table data are discussed and compared. Using car switching data from Britain and France, the relative strengths of labour-intens ive and automated methods are investigated. It is shown that there is considerable value in procedures that mimic and refine the labour-intensive methods.