Hi, I am Nalin Shani,

Research

Working Papers

• Too Experienced to Be Helpful? How Product Experience Relates to the Helpfulness of Negative Reviews on Steam
  with Achal Bassamboo and Maria Ibanez
  Under Review at Marketing Science
  Covered by Kellogg Insight (link)
  Abstract

  As user reviews grow on online platforms, separating useful feedback from noise becomes important for helping consumers navigate available reviews. Yet the most direct indicator of review helpfulness, readers' votes, arrives later, so platforms must assess the expected helpfulness of new reviews based on cues available at submission. Many platforms display reviewer experience cues, but it is unclear how such experience translates into perceived review helpfulness for negative reviews, where sustained use and a decision to reject the product must be reconciled. Using 26.8 million video-game reviews on Steam (4.4 million negative), where reviews display the reviewer's playtime at the time of writing and a binary verdict (recommend/not recommend), we estimate models with rich controls and fixed effects. We find that for negative reviews, helpfulness is inverted U-shaped in reviewer experience, peaking at moderate experience. This contrasts with the well-established aggregate U-shaped relationship (Zhang et al. 2026). Exploratory extensions show that observable cues (e.g., review length, product maturity) shift this relationship. Taken together, these findings show that reviewer experience is a verdict- and context-dependent helpfulness signal rather than a uniform quality signal.

  

  Panels show the binned mean log(helpful votes + 1) by percentiles of log(playtime) for (left) recommended and (right) not-recommended reviews. Shaded bands indicate point-wise 95% confidence intervals for the binned mean. x-axis tick labels show representative playtime values (in hours).