Abstract: Data from disparate sources including panel data, display audit data, point of sale data, trade area data, and store data are combined and analyzed to derive a model that relates in-store display attributes to individual shopping behavior. In particular, techniques are proposed for linking information about in-store displays with shopping data for individual consumers in order to create a model characterizing whether purchasing behavior for a particular shopper, trip type, and/or the basket contents are influenced by a particular in-store display. These inferences can further be projected onto a larger population or trade area to predict impact of different promotional strategies using in-store displays.