In another essay on psychological safety I talked at length about the pre-requisite for good feedback: people need to know you care. What else makes good feedback?
Focus on the future.
The goal of giving feedback is to encourage effective future behavior. Thus, when giving feedback, don’t get hung up on the past.¹ Focus on what your teammate can do differently or should keep doing in the future.
When somebody pushes back on feedback, just let it go. Don’t let yourself get sucked into an argument about the past, when the whole point of feedback lies in the future. After all, your interpretation of the past might be wrong.
You might be wrong.
Good feedback works bidirectional. It helps clear up misconceptions on both sides. As the person giving feedback you might be missing some crucial information that puts the feedback-provoking behavior in a different light.²
Thus, replace blame with curiosity. Do not claim superiority over the facts. It is better to talk about the impact a certain behavior had on you (or how it made you feel) as opposed to postulating truth.³ A tinge of solipsism makes all feedback taste better.
Criticize behavior, not character.
So is talking about your feelings, instead of (supposed) facts enough?
“When you gave your presentation today, I couldn’t follow and it made me feel like you didn’t know what you are talking about.”? — “Well, maybe you’re the one who’s stupid, not me!”
If you don’t watch out, feedback can become personal really fast. Just as bad, some feedback is never given, because people are afraid the other person will not take it well.
The solution is to stop committing the fundamental attribution error.⁴ People are not inherently “aggressive”, but often display behavior that is. Character is hard to change.⁵ Behavior less so. Don’t confuse the two. It is the latter that good feedback is about.
Imagine how much harder it would be to learn how to drive if your car only ever budged 30 seconds after you move the wheel. Or how hard it is to build a useful product if you only ever get feedback weeks after launching a feature.⁶
Guidance has a short half-life — a few weeks later your teammate will not have the chance to fix the problem anymore. They might not even remember the problem in the first place. And wouldn’t you rather hear that something you’re doing is bad the first time you do it and not when it has become a habit?
“Whether professionals have a chance to develop intuitive expertise depends essentially on the quality and speed of feedback, as well as on sufficient opportunity to practice.” — Daniel Kahneman, Thinking, Fast and Slow⁷
To help your teammates become experts at what they do, give timely feedback and give it often. Build a habit to speak up, right when you notice a behavior. Don’t wait for the next 1:1 (you might have forgotten it by then), instead take two minutes after a meeting or at the end of a conversation to share what you observed.
Remember, the goal of good feedback? Encouraging effective future behavior. If it is not clear which behavior your feedback is trying to address, you are not being specific enough.
Fortunately, the more timely you give feedback, the easier it is to be specific about the behavior you want to address. Fresh in memory, there will be less discussion about what happened and more time to focus on the future. Being specific also helps to avoid confusing behavior with character and makes it easier to act on your feedback.
“Often you’ll be tempted not to describe the details because they are so painful. You want to spare the person the pain and yourself the awkwardness of uttering the words out loud. But retreating to abstractions […] can actually unintentionally signal that the behavior in question was so bad/shameful that you can’t even talk about it, thereby making it hard for the person to move on.” — Kim Scott, Radical Candor
Praise in public, criticize in private
Feedback isn’t just about being critical. Making sure good behavior gets repeated in the future is just as important as avoiding bad behavior.⁸
“By explicitly describing what was good or what was bad, you are helping a person do more of what’s good and less of what’s bad — and to see the difference.” — Kim Scott, Radical Candor
Giving praise is also a great way to show appreciation. But remember to put as much effort into your praise as you put into your criticisms. Vague criticism is unhelpful and demotivating. Vague praise can be just as counterproductive.⁹
Good feedback focuses on the future, recognizes you might be wrong and criticizes behavior, not character. It is timely and specific.
Voilà. We’ve got the lemons. Time to make some lemonade.
¹ There is an interesting article in the Harvard Business Review that propounds the idea that instead of asking for feedback, a word that is often associated with an evaluation of the past, one should be asking for advice, which more naturally focuses on future actions.
² It is for this reason that Kim Scott called her book Radical Candor and not Radical Honesty. Honesty carries a connotations that suggests knowing the truth. We’ll hear more from Scott later.
³ There are plenty of feedback templates that try to help you with this. I find it hard to use them while still sounding authentic, so I usually don’t. I think as long as you understand the thinking that went into creating these templates, you are ready to rumble. Being explicit, and pointing to the principles discussed in this essay is often more helpful than the template itself. “Hey, I might be wrong, but in case I’m not I thought you might find this observation helpful for the future.”
For completion’s sake, here are the templates that do make sense to me.
In The Effective Manager it goes: State the behavior with “When you…” and state the impact with “Here’s what happens…”
At McKinsey they say: “When you did [X], it made me feel [Y]. In the future, I would recommend that you do [Z]”
In Radical Candor, Scott recommends the Situation, Behavior, Impact mnemonic. To demonstrate:
Instead of yelling, “You asshole!” when somebody grabs your parking space, try saying, “I’ve been waiting for that spot here for five minutes [situation], and you just zipped in front of me and took it [behavior]. Now I’m going to be late [impact].”
⁴ Straight from Wikipedia:
This effect has been described as “the tendency to believe that what people do reflects who they are”, that is, to overattribute their behaviors (what they do or say) to their personality and underattribute them to the situation or context.
⁵ Character is hard to change, but there are other things that are even harder. Namely, everything outside of your circle of influence. Nothing is more frustrating than getting critiqued for something that you cannot control.
⁶ Bret Victor expounds the importance of immediate feedback whilst creating (creating: the yang to learning’s ying) in his excellent talk Inventing on Principle.
⁷ I can highly recommend reading Chapters 21 & 22 in Kahneman’s eye-opening Thinking, Fast and Slow. There he answers why one can trust some experts’ intuitions (like athletes or chess players) but not others’ (like clinicians or political scientists). What is the difference between these professions? It’s the quality and speed of feedback.
The job of clinicians and political scientists is harder, you say? “It is wrong to blame anyone for failing to forecast accurately in an unpredictable world. However, it seems fair to blame professionals for believing they can succeed in an impossible task.”, says Kahneman.
⁸ If you are only starting out with giving feedback, some people suggest to focus on giving praise much more often than criticism. I find keeping track of how much positive and how much negative feedback you dished out less and less necessary the more psychologically safe the interactions within your team are.
⁹ In Radical Candor, there is an interesting example of a manager congratulating an engineer on his excellent work on a successful feature in front of the whole company. But that engineer was not the only one that had contributed to said feature. By praising the engineer without being specific in his praise the manager accidentally created the impression that the engineer had stolen all the credit for himself, an impression the engineer later tried to correct by sending out a company-wide e-mail listing everyone that had worked on the feature with him.