Illustration showing screens with stars, indicating online reviews

How Power Imbalances Can Make Online Reviews Harmful

By Charly Haley

 

Before doing almost anything — from picking a restaurant to applying for a job — people usually check the online reviews first. Reviews can increase accountability, providing an incentive for companies to offer a five-star experience. But when might review systems do more harm than good?

New research from Minnesota Carlson Associate Professor Jiao Luo reveals that when a societal power differential is present between two parties on a review platform — such as employees rating their employers — the platform can become ineffective.

“We almost automatically implement these rating systems, at this point, on any platform because we want to provide better information for both sides,” says Luo, who holds the Andrew Van de Ven Faculty Fellowship in Carlson’s Strategic Management & Entrepreneurship Department. “But the challenge is, if we don’t think about some of the fundamental power differences originating from the societal conditions, then we run the risk of the more disadvantaged population side actually not receiving a benefit from the rating system.”

Luo’s findings, which will be published in Organization Science, come from an experiment she did in 2021 in partnership with an online platform in Singapore that connects in-home nannies with employers. The nannies on this platform were all foreign workers, which created an even greater power differential due to their reliance on work for their visas.

Key Takeaways

  • Power Imbalances Hurt: Carlson research reveals rating platforms fail when large societal power differentials exist.
  • Fear Halts Feedback: Vulnerable or disadvantaged people may disengage from review systems due to a fear of retaliation from those with more power.
  • Evaluate User Impact: Leaders must assess societal power dynamics before implementing platform rating systems.
Jiao Luo
Associate Professor Jiao Luo

Luo’s experiment introduced a rating system on the platform. What she found was that the rating system did not increase accountability for the employers — instead, it resulted in disengagement from the workers for fear of retaliation, even when the reviews were anonymous.

“That was the surprising part,” Luo says, “but when we did our post interviews, it made sense. The workers were hesitant to upset their employers. They said they’d rather take mistreatment than have the employers leave the platform because then they would have no work.” Because of Luo’s study, the company she partnered with ultimately decided not to permanently implement a review system.

While this experiment was conducted in a specific industry in Singapore, Luo says the findings about the impact of power differentials on rating systems can be applied broadly. Any company leaders who are considering implementing a review system should think about the impact on users with less power and the consequences of that, she says. Beyond the example of employees rating employers, this research could also apply to any platform where service providers and consumers rate each other.

“The circumstances we’re describing are very specific — and that’s the nature of field experiments, they’re very localized — but the idea of power differential is not specific at all,” Luo says. “When we set up these kinds of rating systems, we often don’t think about or don’t even ask the side that has less power what their motivations are or how they will react. If we don’t think about that, then we have the potential to really harm the ones that we’re trying to help.”

This article appeared in the Spring 2026 Discovery magazine

Discover how faculty research is going virtual in the classroom, using AI to make the medical device industry safer, revealing a cost-effective way to retain volunteers, and more.

Spring 2026 table of contents