Online review structure, not just sentiment, predicts what readers find helpful
Why it matters: Companies and third-party review platforms can design pages to prompt more helpful reviews for potential customers.
- Universities of Cambridge and Queensland conducted a study of 195,675 Amazon reviews across 5,487 products to understand the impact of review structure on perceived helpfulness.
- Dr. Yeun Joon Kim from Cambridge Judge Business School highlights the challenge of crafting evaluative messages with both positive and negative aspects, emphasizing the overlooked structural dimension.
- The study identified nine possible review structures, ranging from consistently positive to consistently negative, and various combinations in between.
- Review helpfulness is enhanced by increasingly positive structures for highly-rated products, progressively negative structures for average-rated products, and constructively critical openings for low-rated products.
- Dr. Luna Luan from the University of Queensland noted that the results, though nuanced, clearly demonstrate the importance of tailoring review structure to the product's rating.
A study of nearly 200,000 Amazon reviews reveals that the structure of an online review, not just its sentiment, significantly impacts how helpful readers perceive it to be. Researchers from the Universities of Cambridge and Queensland found that the optimal review structure varies depending on the product's rating, with different trajectories of positive and negative feedback proving more effective in different scenarios.
