The authors propose a new, exploratory approach for analyzing market structures that leverages two recent methodological advances in natural language processing and machine learning. They customize a neural network language model to derive latent product attributes by analyzing the co-occurrences of products in shopping baskets. Applying dimensionality reduction to the latent attributes yields a two-dimensional product map. This method is well-suited to retailers because it relies on data that are readily available from their checkout systems and facilitates their analyses of cross-category product complementarity, in addition to within-category substitution. The approach has high usability because it is automated, is scalable and does not require a priori assumptions. Its results are easy to interpret and update as new market basket data are collected. The authors validate their approach both by conducting an extensive simulation study and by comparing their results with those of state-of-the-art, econometric methods for modeling product relationships. The application of this approach using data collected at a leading German grocery retailer underlines its usefulness and provides novel findings that are relevant to assortment-related decisions.
Personalized marketing in retail requires a model to predict how different marketing actions affect product choices by individual customers. Large retailers often handle millions of transactions daily, involving thousands of products in hundreds of categories. Product choice models thus need to scale to large product assortments and customer bases, without extensive product attribute information. To address these challenges, we propose a custom deep neural network model. The model incorporates bottleneck layers to encode cross-product relationships, calibrates time-series filters to capture purchase dynamics for products with different interpurchase times, and relies on weight sharing between the products to improve convergence and scale to large assortments. The model applies to loyalty card transaction data without predefined categories or product attributes to predict customer-specific purchase probabilities in response to marketing actions. In a simulation, the proposed product choice model predicts purchase decisions better than baseline methods by adjusting the predicted probabilities for the effects of recent purchases and price discounts. The improved predictions lead to substantially higher revenue gains in a simulated coupon personalization problem. We verify predictive performance using transaction data from a large retailer with experimental variation in price discounts. This paper was accepted by Gui Liberali, Management Science Special Issue on Data-Driven Prescriptive Analytics.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.