Online product ratings drive sales, prompting managers to constantly seek ways to improve them. While traditional research compares high-rated reviews against low-rated ones (e.g., 5 stars vs. 1 star), this study investigates the subtler, yet critical, distinction between “best” (5-star) and “second-best” (4-star) reviews. We provide novel insights for managers striving to elevate “good” products to “great” ones. Our framework highlights two dimensions of evaluation: the product (hedonic vs. utilitarian) and the consumer (affect vs. cognition). We hypothesize that achieving the highest rating is significantly harder for products perceived as high-risk, low-affect, or high-cognition. Furthermore, we examine how these effects are moderated by three key interactions: Affect × Cognition, Product Type × Cognition, and Product Type × Risk. Empirical analysis of a 15-year dataset of Amazon reviews across 19 product categories supports our hypotheses. A secondary application in the hotel industry further identifies specific amenities that can help managers raise their product ratings from 4 to 5.
Kim et al. (Tue,) studied this question.