A Household-Level Television Advertising Exposure Model

Abstract

The author proposes a household-level television advertising exposure model for media planning that is based on household exposure records. Consistent with the industry practice, managerial input of advertising for the model is gross rating points (GRP) for each “daypart” and week. Exposures of each household within each daypart during each week follow a Poisson process with a distinct rate, which is inferred from the household's past exposure history after accounting for the advertising intensity. Because exposure pattern is computed at the household-level, the model accounts for audience accumulation of dayparts combination, which has a considerable managerial importance in media planning. The model is shown to fit the observed exposure pattern well at both disaggregate and aggregate levels. Applications illustrate the effect of changing the level of GRP on reach and frequency and advertising efficacy of different day-parts. The author implements reallocation of the current level of GRP among dayparts using the “greedy algorithm” to improve reach.