Estimating Search Benefits from Path-Tracking Data: Measurement and Determinants

We study consumer search behavior in a brick-and-mortar store environment, using a unique data set obtained from radio frequency identification tags, which are attached to supermarket shopping carts. This technology allows us to record consumers’ purchases as well as the time they spent in front of the shelf when contemplating which product to buy, giving us a direct measure of search effort. We estimate a linear regression of price paid on search duration in which search duration is instrumented with a search cost shifter. We show that this regression allows us to recover the marginal return from search in terms of price at the optimal stopping point for the average consumer. Our identification strategy as well as coefficient interpretation are valid for a broad class of search models and we are hence able to remain agnostic about the details of search process, such as search order and search protocol. We estimate an average return from search of $2.10 per minute and explore heterogeneity across consumer types, product categories and category location in the store. We find little difference in the returns from search across product categories, but large differences across consumer types and locations. Our findings suggest that situational factors such as the location of the category or the timing of the search within the shopping trip are more important determinants of search behavior than category characteristics like the number of UPCs.