AI and Analytics for Business

Research Paper Series

Online Product Opinions: Incidence, Evaluation and Evolution

While recent research has demonstrated the impact of online product ratings and reviews on product sales, we still have a limited understanding of the individual’s decision to contribute these opinions. In this research, we empirically model the individual’s decision to provide a product rating and investigate factors that influence this decision. Specifically, we consider how previously posted opinions in a ratings environment may affect a subsequent individual’s posting behavior, both in terms of whether to contribute (incidence) and what to contribute (evaluation), and identify selection effects that influence the incidence decision and adjustment effects that influence the evaluation decision.

Our results indicate that individuals vary in their underlying behavior and in their reactions to previously posted product ratings. While less frequent posters exhibit bandwagon behavior, more active customers reveal differentiation behavior when posting online opinions. Systematic patterns in these behaviors have important implications for the evolution of online product opinions, which we illustrate through the use of simulations. Our simulations also show that posted product opinions can be affected substantially by the composition of the underlying customer base. Specifically, when a product’s customers are polarized, posted opinions are more negative and exhibit a stronger downward trend when compared to a homogeneous, neutral customer base with the same median opinion. This is a result of a core group of “activist” customers posting increasingly negative opinions in an effort to differentiate themselves from others in the community.

Keywords: User-Generated Content, Online Word-of-Mouth, Product Ratings and Reviews, Opinion Dynamics