A Structural Model of Channel Choice with Implications for Retail Entry

In this paper, we propose a structural model of channel choice and product purchases, in which consumers base their channel choice decision on both the utility inherent in visiting a channel and the expected utility from purchasing products upon visiting. Specifically, we allow channels to have different abilities to resolve consumer uncertainty about product fit that varies by product category. We use this model to evaluate and predict the demand impact of retail entry for a seasonal goods provider that uses both retail and online channels to sell directly to consumers. Both online and retail sales will depend on retail outlet proximity through the rate of new customer arrivals and the shopping trip frequency of existing customers, as well the tradeoffs inherent in choosing between channel formats and the outside option. We find evidence of channel complementarity through increased arrivals of new customers and shopping frequency as the distance to retail outlets decreases. While the net effect of retail entry is demand-stimulative, we find these market expansion effects are partially offset by increased switching from online to retail formats in areas close to store locations. In a series of counterfactual experiments, we demonstrate how our estimates may be used to identify promising locations for new physical stores and to explore channel-based price discrimination policies. ⇤Shriver: Columbia Business School, Columbia University, [email protected]. Bollinger: The Fuqua School of Business, Duke University, [email protected]. We thank the Wharton Customer Analytics Initiative (WCAI) for access to the data and seminar participants at the Marketing Science conference for their comments. We are also grateful to Avi Goldfarb, Kitty Wang, and Hema Yoganarasimhan for useful discussions regarding the paper. All errors are our own.