Review structure shapes helpfulness, 200K-review study finds
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- Researchers from the Universities of Cambridge and Queensland analyzed 195,675 Amazon reviews spanning 5,487 products and concluded that how information is organized through a review shapes helpfulness as much as the content itself.
- Published in Scientific Reports, the study identified nine review structures (Type A through Type I) ranging from 'starts positive, grows more positive' to 'starts negative, grows more negative.'
- Optimal structure depends on product rating: highly-rated products benefit most from increasingly positive reviews; average-rated products get a helpfulness boost from progressively negative trajectories; low-rated products score highest when they open constructively before introducing criticism.
- The most commonly used review styles are not the most helpful, particularly for average- and low-rated products—suggesting reviewers often prioritize venting or self-expression over reader usefulness.
- Co-author Dr. Yeun Joon Kim of Cambridge Judge Business School proposed that review platforms replace blank 'Write your review here' fields with micro-prompts guiding reviewers toward structures readers find most informative for a given product rating.
- Co-author Dr. Luna Luan of the University of Queensland, who led the research during her Ph.D. at Cambridge, said the findings show overall sentiment is insufficient: 'It is the broader structure of sentiment—how positivity and negativity evolve throughout the review—that shapes how readers interpret online reviews.'
Why it matters: The study hands review platforms and retailers a concrete design lever: instead of blank text boxes, structured micro-prompts could nudge reviewers toward trajectories proven to boost helpfulness by product rating tier. The mismatch between popular review styles and the most useful ones—driven by reviewers prioritizing self-expression over information transfer—means platforms have measurable room to improve review quality without collecting more data.
Ask SkimNews
