Diagnosing Brand Performance: Accounting for the Dynamic Impact of Product Availability with Aggregate Data

The extent to which a brand's individual products (relative to competing products) are available to consumers for purchase in a retail store can critically affect the brand's overall performance. However, store-level product availability information is lost in aggregate market-level data sets and has been ignored by extant demand studies in general, which can create the risk of misinformed managerial decision making. In this research, the authors propose a unique methodology to enable manufacturers to infer retailers' joint stocking probability of products from aggregate data and, thus, enable consumers' choices to be contingent on the assortment of products available in retail stores. The application of the proposed framework in the context of an emerging market results in unbiased demand parameter estimates, a significantly better model fit, and richer managerial insights (compared with conventional approaches) pertaining to how brand performance is affected by the (1) dynamics of retailers' stocking preferences, (2) assortment of products that retailers are more (vs. less) likely to jointly stock, and (3) cannibalization of retailers' shelf space resulting from product line extensions.