A Model for Relating Advertising Media Exposures to Purchase Incidence Behavior Patterns

A stochastic model is proposed to study the effect of advertising on consumer purchase dynamics by specifically relating the distribution of exposures, from a brand's media schedule, to brand purchase incidence behavior patterns over time. The approach is based on underlying individual consumer behavior assumptions that reflect the dynamics of exposures to advertising and brand purchase behavior, conditional on the level of advertising exposure received. Moreover, a population-based model is derived by aggregating over individual consumers who may be heterogeneous with respect to exposure probability, purchase rate, and sensitivity to advertising. The approach can be implemented with experimentally-based data and estimation procedures that seek to isolate the effect of advertising and also permit the evaluation of advertising influences on purchase behavior patterns. For the purpose of model implementation, parameter estimation procedures are outlined and applied to empirically evaluate three alternative model versions that reflect different advertising exposure behavior assumptions. Both the goodness of fit results presented and prior empirical experience, in connection with past applications to TV media schedule analysis, lend to the model's validity. The paper concludes with an appraisal of the potential strengths and limitations of the proposed model as a tool for advertising media analysis and planning in practice.