When Feedback Backfires
Instant evaluations, in some cases, can worsen performance
In these data-driven times, we increasingly believe that feedback is the secret to any success. Tell us what we are doing right or wrong, and we’ll be able to improve.
But it turns out that we humans are not always so good at making corrections based on feedback, no matter how compelling the data.
That’s the upshot of new research by Masha Shunko, an assistant professor of operations management at the University of Washington Foster School of Business.
Dr. Shunko and co-authors Vivek Choudhary and Serguei Netessine analyzed the records of drivers who allowed a cellphone app to monitor and rate their performance in hopes of getting a “good driver” insurance discount. Drivers who were given the opportunity to review the rating of their last trip actually performed 13 percent worse on their next drive, compared to those who had no access to prior performance reports. And those who reviewed prior ratings drove more dangerously. They tended to speed, accelerate too quickly and break harshly.
In other words, the feedback backfired.
Why feedback can backfire
Why does this happen? Dr. Shunko explains that our tendency to overestimate our abilities can lead us to ignore negative feedback. And feedback that is far below a targeted goal can make that goal seem unattainable, leading us to give up.
Scoring below but near a target goal can motivate improvement, but feedback on performance exceeding a goal—especially by a substantial margin—can result in complacency.
“Not all negative feedback motivates people to work harder,” she explains. “Moreover, positive feedback tends to reinforce an ‘optimism bias,’ which can result in riskier behaviors and reduced efforts.
“Hence, we find that the overall average effect of instant feedback is negative.”
She notes that the net negative effect of instant feedback is also short-lived.
Making evaluations work
Dr. Shunko suggests that the findings might extend to other contexts of real-time performance assessments. Driving apps are similar to ones tracking fitness and diet, for example, and might have some of the same problems.
“I think our general conclusion—that not all feedback is equal—will translate to other settings,” she told Harvard Business Review.
So, what’s the best way to give feedback? Dr. Shunko suggests that evaluations should be individualized, whenever possible. Some people prefer immediate feedback while others like to wait. And some respond better to social comparisons while others are motivated by comparisons to their own past performance.
Finally, how can you use feedback to perform better, not worse?
“Focusing on what you’ve done well in the past—your best previous performance—is a good way to understand what you need to do to succeed, while thinking about an average performance would have less impact,” she said. “You want to set a high reference point for yourself.”
“Does Immediate Feedback Make You Try Less Hard? A Study of Automotive Telematics” is the work of Masha Shunko, Vivek Choudhary of INSEAD and Serguei Netessine of The Wharton School.
Read the full Q&A with Dr. Shunko in Harvard Business Review.