Standing out from adjacent reviews: How content similarity affects review helpfulness

Prior research and practice have largely examined review content characteristics in isolation to enhance review helpfulness, overlooking that perceived helpfulness is context dependent. This oversight poses a broader platform design challenge: how to structure information environments as consumers browse reviews sequentially. Drawing on the information diagnosticity and variety-seeking perspectives, we introduce adjacency-based content similarity and argue that platforms can enhance overall helpfulness by reducing local redundancy. Using around 3 million TripAdvisor hotel reviews and Doc2Vec to measure this similarity, we find that reviews are perceived as less helpful when highly similar content appears adjacently; this effect is attenuated for high-reputation reviewers and amplified for hedonic (vs. utilitarian) products, and is driven primarily by similarity in content on experience-related attributes. Complementary online experiments establish causality and identify perceived uniqueness as the mediating mechanism. We further replicate the main pattern in other contexts (e.g., restaurants, video games, and consumer goods). This study advances understanding of review helpfulness by highlighting the role of sequential information structures and offers direct implications for platforms’ algorithmic feed design, particularly through review reordering and curation to mitigate local redundancy.