When Positive Reviews Hurt and Negative Reviews Help: The Role of Review Novelty, Volume, and Valence

The popularity of electronic word-of-mouth has increased dramatically over the years, and it
has become important to understand how online reviews impact stakeholders such as platforms, businesses, and consumers. While prior works have examined the impact of different review and reviewer characteristics, the impact of novel information in reviews (which is information not present in other reviews) is not well understood. On one hand, novel information can be perceived as valuable by consumers when making decisions. On the other hand, such information can be perceived as inconsistent with existing reviews and, therefore,
not credible. These perceptions can be influenced by the volume of available reviews as well as the valence of reviews. The Elaboration Likelihood Model with its characterization of levels of elaboration provides important insights to understand the interactions between novelty and volume while accounting for the valence of reviews. We use a dataset from Yelp that provides reviews and related information for restaurants to investigate whether the influence of novel information in reviews changes as the review volume and valence change. We show that novel information positively impacts the helpfulness of reviews and restaurant check-ins when the review volume is low—this holds for novel information whether it is in positive or in negative
reviews. However, the impact of novel information becomes negative when the review volume becomes high; surprisingly, this holds even for positive reviews. Our findings have practical implications for review sites (platforms), businesses, contributors (reviewers), and consumers.