Observed and Unobserved Preference Heterogeneity In Brand Choice Models

In deciding what brand to buy consumers trade off between how valuable each brand is to them and its price. Scanner data based brand choice models that evaluate this trade off and allow for unobserved heterogeneity are a very popular topic. We extend this line of research by explicitly incorporating brand preferences rather than just utilizing brand specific constants and past purchase behavior to proxy for them. We illustrate our model using a unique dataset that combines survey and scanner data collected from the same individuals. The addition of individual specific brand preference information to a standard scanner choice model that incorporates brand specific constants, loyalty, promotion and price significantly improves how well choices are explained and predicted. Moreover, this “observed” heterogeneity via inclusion of individual specific brand preferences better explains choice than does “unobserved” heterogeneity in the standard scanner model parameters. This is largely due to the inability of the researcher to specify unobserved heterogeneity distributions that fully capture the heterogeneity underlying consumer preferences. We find that a model which does not include the individual level preference information overestimates both brand loyalty and price sensitivities. The brand loyalty coefficient is overestimated because models without preference information confound state dependence, heterogeneity and preference effects. Price sensitivities are reduced in the more complete model because the “real” preference based consumer is less ready to switch from his most preferred brand than is the “average” preference based consumer implicitly used in standard models. Furthermore, we find that, the lack of cross-sectional survey information on the differences in preferences across households causes the standard models to overestimate the heterogeneity in price and loyalty sensitivities. This also results in these models commonly misidentifying both high and low price and loyalty sensitive consumers. Our more accurate loyalty and price sensitivity estimates should help managers set prices and target promotions better.