One in three times, feedback makes people worse

The further feedback moves from the work towards the person, the more likely it is to backfire.

Feedback is the subject I’ve taught more than any other. Since I started coaching in 2017, it is the number one request, from first-line managers to CEOs. And when I come back six months later, most of them are still bad at it. The same vague sentence. The same compliment stapled to a criticism. The same “I just want to help you grow” that lands like a verdict.

That left me two options. Decide people are hopeless, or ask whether what I’d been teaching was aimed at the wrong thing. I picked the second one. The research had been waiting for thirty years.

The number nobody puts on a slide

In 1996, two researchers, Avraham Kluger and Angelo DeNisi, pulled together every experiment on feedback and performance they could find. Not opinions: controlled studies where some people get feedback, some don’t, and performance is measured afterwards. Six hundred and seven effects, covering 23,663 people.

On average, feedback helped. A moderate effect (d = 0.41), the kind of number that lets you keep the training catalogue exactly as it is.

Then they looked at the spread. In more than one case out of three, feedback made performance worse. People who received feedback did less well afterwards than people who received nothing.

The detail that should bother anyone who teaches this: the same negative effects had been showing up in studies for most of the twentieth century. They were ignored, because they didn’t fit the story. The story is that feedback is a gift: say it well, say it kindly, people improve. The data says a third of the time, no.

So the real question was never “how do I give feedback”. The real question is what separates the two thirds that help from the third that hurts.

It was never about how you say it

Look at what most feedback trainings teach, mine included. Structure. Timing. Tone. Situation, behaviour, impact. Make it a dialogue. Care about the person and be direct about the problem.

All of that is about how you say it. And Kluger and DeNisi found that how you say it is not what decides whether feedback helps or hurts.

What decides is where the attention lands.

Feedback that keeps someone’s attention on the task works. “The number on page three doesn’t match page seven.” The brain goes to the pages. Feedback that pulls attention towards the person, towards who they are, fails. And it fails even when it’s positive.

“You’re really rigorous, but...” The brain doesn’t hear the “but”. It hears “you’re”. Now the person is thinking about themselves: their reputation, their standing in the room, whether you like them, whether they’re safe. All the capacity that should go to the work goes to managing the self.

That’s the whole mechanism. The further feedback moves from the work towards the person, the more likely it is to backfire. Which means the sandwich isn’t a kindness. It’s two slices of self-attention around a criticism nobody can use.

Praise is the worst of it

Ten years later, John Hattie and Helen Timperley sorted the feedback research into four levels: feedback about the task, feedback about the process, feedback about how someone steers their own work, and feedback about the self. The praise.

The self level is the least effective of the four. It is also the one managers use most. “You’re so good at this.” “I can always count on you.” “Great job.”

This isn’t old research gathering dust. In 2020, a new meta-analysis of 435 studies and more than 61,000 participants redid the exercise. Same ranking. Praise at the bottom.

So when a manager opens with a compliment “to make it safe”, here is what actually happens: they pull the attention to the self first, then expect the person to switch back to the task when the real message arrives. Nobody switches back.

Kind managers aren’t failing because they’re kind. They’re failing because they’re aiming at the wrong thing.

What I added, and what stays

My method wasn’t wrong. Structure, dialogue, care: all of that stays, and the full version is in the feedback formula episode and article. What was missing was a step before it. A check on what the feedback is actually about. That’s the step I added.

Three things change in my sessions.

One. The first thing we practise is not saying it well. It’s choosing the object. Before anyone opens their mouth: is this about the work, or about the person? If it’s about the person, rewrite it until it’s about the work. “You’re not a team player” becomes “The decision went out before procurement saw the draft.” Same message, different target.

Two. Praise gets the same test. Not banned, redirected. “Great job” goes. “The way you sequenced the three pilots meant nobody hit saturation” stays. The person still gets the credit, and their attention stays on what they did, so they can do it again.

Three. The sandwich is out. Not softened. Out. If a message needs a slice of bread to be said, it isn’t ready to be said.

And the line I now open every session with: we are not here to make you kinder. Kind isn’t the lever. We are here to make you boring. Boring feedback is about the work. Boring feedback is the two thirds that don’t do damage.

The exercise

One time in three, feedback makes people worse. We’ve had the data for thirty years. I had it too, and now it’s in the training.

So the next time someone says “can I give you some feedback”, the question isn’t whether they’ll say it nicely. It’s whether they’re about to talk about the work, or about you.

And if you give feedback for a living: pick one sentence you said this week. Check what it was about. That’s the whole exercise.

---

This continues every week in The Nerve.

Sources

• Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on performance: a historical review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulletin, 119(2), 254-284. https://doi.org/10.1037/0033-2909.119.2.254
• Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81-112. https://doi.org/10.3102/003465430298487
• Wisniewski, B., Zierer, K., & Hattie, J. (2020). The power of feedback revisited: a meta-analysis of educational feedback research. Frontiers in Psychology, 10, 3087. https://doi.org/10.3389/fpsyg.2019.03087