48 Pages Posted: 11 Oct 2016 Last revised: 23 Jun 2017
Date Written: May 24, 2017
Online product reviews greatly impact sales. Although aggregate data suggests that customers read reviews rather than relying only on summary statistics, little is known about consumers’ review-reading behavior and its impact on conversion at the granular level. To fill this research gap, we analyze a comprehensive dataset that tracks individual-level review reading, search, as well as purchase behaviors, and we achieve two objectives. First, we describe consumers’ review-reading behaviors. In contrast to what has been found with aggregate data, individual level consumer online journey data shows that around 70% of the time, consumers do not read reviews; they are less likely to read reviews for products that are inexpensive and have many reviews. Second, we quantify the causal impact of quantity and content information of the “read” reviews on sales. The identification relies on the variation in the reviews seen by consumers due to newly added reviews. To extract content information, we develop two deep learning models. The full deep learning model predicts conversion directly and the partial deep learning model identifies six dimensions of content in the reviews. The comparative advantage of deep learning models is that they let us sift category-specific content features across a wide range of product categories without hand-coding features with human intervention or domain knowledge. Across both models, we find that aesthetics and price content in the reviews significantly affect conversion across a wide range of product categories. Counterfactual simulation suggests that re-ordering review content can have the same effect as a 1.6% price cut for boosting conversion.
Keywords: Consumer Purchase Journey, Product Reviews, Review Content, Deep Learning
Suggested Citation: Suggested Citation
Liu, Xiao and Lee, Dokyun and Srinivasan, Kannan, Large Scale Cross Category Analysis of Consumer Review Content on Sales Conversion Leveraging Deep Learning (May 24, 2017). NET Institute Working Paper No. 16-09. Available at SSRN: https://ssrn.com/abstract=2848528 or http://dx.doi.org/10.2139/ssrn.2848528