Events

Dr. Xitong Li
Associate Professor
Department of Information Systems and Operations Management
HEC Paris, France

Although conventional wisdom suggests that product recommendations can help consumers find products that are lower priced and fit their tastes better, the literature lacks empirical evidence of these mechanisms. We separately estimate the benefits of recommendations to consumers due to these mechanisms through a randomized field experiment on an apparel retailer’s website in the US. We collect unique data on the affinity scores computed by the recommendation algorithm to estimate the effect of product recommendations in helping consumers discover larger-value products, which are lower-priced, fit their tastes better, or both. The discovery of higher-value products results in a higher consumers’ purchase probability (lower likelihood of failed search efforts) and purchase of lower-priced and/or better-fit products. We provide additional evidence of these mechanisms by showing a larger benefit of product recommendations to consumers in product categories with higher average prices, larger relative price dispersions, and higher heterogeneity in consumers’ tastes. Finally, we find that consumers substitute other search tools on the website with product recommendations when available. Our findings have implications for the design and deployment of algorithmic product recommendations on digital platforms.