How Beneficial are Recommendations to Consumers? Estimates of Relative Benefits of Product Recommendations to Consumers and Retailers

31 Pages Posted: 17 Nov 2020 Last revised: 8 Jan 2021

See all articles by Xiang (Shawn) Wan

Xiang (Shawn) Wan

University of Florida - Warrington College of Business

Anuj Kumar

University of Florida - Warrington College of Business

Xitong Li

HEC Paris - Information Systems and Operations Management

Date Written: October 1, 2020

Abstract

Do personalized offerings enabled by big data only bring economic benefits to firms but not to consumers? In this paper, we seek to address this question by examining a particular category of personalized offerings provided by algorithmic product recommendation systems. While the economic value of algorithmic product recommendations to firms is mostly understood, their benefits to consumers are not well quantified in the literature. We conduct a randomized field experiment on an apparel retailer’s website in the US to estimate the benefits of recommendations to consumers. We find that recommendations helped consumers discover lower-priced products on the website, which results in a higher probability of product purchase (lower likelihood of failed search efforts) and a lower price of purchased products. Specifically, an additional recommended product page view due to recommendations leads to a 15 percent increase in purchase probability and a $1.59 decrease in the purchase price. The product recommendations are particularly beneficial to consumers for product categories with a higher relative price dispersion and a higher proportion of niche products. Due to these benefits, consumers substitute 0.14 keyword search page views and 0.63 product category page views for each recommended product page view under recommendations. The recommendations generated a surplus of $56,631 (3.8 percent of the total sales), two-thirds of which the retailer captured as additional revenue, and the remaining one-third the consumers as price savings.

Keywords: Product recommendations; Economic value of recommendations; Search costs; Consumer surplus; Randomized experiment

Suggested Citation

Wan, Xiang (Shawn) and Kumar, Anuj and Li, Xitong, How Beneficial are Recommendations to Consumers? Estimates of Relative Benefits of Product Recommendations to Consumers and Retailers (October 1, 2020). HEC Paris Research Paper No. MOSI-2020-1403, Available at SSRN: https://ssrn.com/abstract=3702762 or http://dx.doi.org/10.2139/ssrn.3702762

Xiang (Shawn) Wan

University of Florida - Warrington College of Business ( email )

Gainesville, FL
United States

Anuj Kumar (Contact Author)

University of Florida - Warrington College of Business ( email )

337 STZ WARRINGTON COLLEGE OF BUSINESS
GAINESVILLE, FL 32611-0001
United States
3522730587 (Phone)

Xitong Li

HEC Paris - Information Systems and Operations Management ( email )

1 rue de la Liberation
Jouy-en-Josas Cedex, 78351
France

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
38
Abstract Views
353
PlumX Metrics