Finding Choice Alternatives in Memory: Probability Models of Brand Name Recall

Recalling brand names is an important aspect of consumer choice in many situations. The authors develop a general Markov model that relates probabilistic aspects of recall to consumer and marketing mix variables. Then they illustrate how parameters can be estimated from recall data for three special cases of the model: The first is a “baseline” version of the model that can be used to assess the extent to which categories of brands can be automatically included in or excluded from memory search; the second is a zero-order model that can be used to estimate the effects of brand variables and individual usage rates on recall latency; and the third is a first-order model that uses aggregate recall data to assess market structure. Finally, the authors demonstrate how the model can be used in numerical analyses to evaluate awareness-building strategies.