A Feature-based Consideration Set Choice Model for Online Retailing

Abstract

We propose a feature-based consideration set model (FCM) motivated by customers’ purchasing behavior in online retail platforms. In our model, customers first form a consideration set by including products with the largest utility based on a subset of product features that are visible on the search results page. The customers then make a purchase decision from the consideration set by accounting for all product features available on the product page. The FCM incorporates heterogeneity in customer preferences across both the consideration and purchase stages. We study joint assortment and price optimization under the FCM. For a homogeneous customer population, we show that the problem can be solved in polynomial time (in the number of products). When the population is heterogeneous comprising multiple customer types, we establish that the joint optimization problem is NP-hard. For tractability, we consider a special case where the heterogeneity in the consideration stage is limited to the price sensitivities of the customer types. Our solution approach again results in a polynomial-time algorithm, when the number of customer types is fixed. We also present an extension of the FCM that facilitates the estimation of model parameters from customers’ click and purchase data, and develop an efficient MLE procedure. Finally, our numerical results on real and synthetic data demonstrate the superiority of the proposed model over common benchmarks both in terms of out-of-sample prediction accuracy and decision performance.

Keywords: Choice Models, Consideration Set, Pricing, Online Retail