Feedback that works: specific, descriptive, timely
You need to be more proactive. Great job on the presentation. Your communication could be clearer. Sound familiar? This kind of feedback is common, well-meaning, and almost completely useless, because the recipient never learns what, concretely, they should do differently.
In 2005, researchers Mark Cannon and Robert Witherspoon examined why so much feedback fails to change behavior, and what characterizes what they call actionable feedback. Their article has become a reference point for everyone working on the development of managers and employees.
What does the research say?
Cannon and Witherspoon show that feedback often goes wrong already in how it is worded. When we observe someone's behavior, we instantly draw conclusions about the person's traits and intentions, and it is those conclusions we then deliver: you are unclear, you seem disengaged. Such judgments are often wrong, they trigger defensiveness, and they are impossible to act on. Nobody knows how to stop being unclear.
Actionable feedback, according to the authors, has the opposite qualities. It is specific: it points to a particular situation and a particular behavior. It is descriptive rather than evaluative: it recounts what was actually said or done, not what that supposedly says about the person. And it connects the behavior to its consequences, so the recipient understands why it matters and what could be done instead.
This is the same logic behind the widely used Situation, Behavior, Impact model, abbreviated SBI: describe the situation, describe the observable behavior, describe the effect. The difference between you don't listen and in yesterday's status meeting you interrupted Sara twice before she could finish, and she stopped contributing after that, is the entire difference between an accusation and something a person can change.
What does this mean for you as a project manager?
Swap judgments for observations. Before giving feedback, ask yourself: what have I actually seen or heard? If your message contains words like always, never, attitude, or commitment, you have probably left the observations behind and started delivering conclusions. Step back to the concrete.
Give feedback close in time and in small doses. A short, specific remark the same day beats a long, saved-up review in the quarterly conversation. Stored examples both fade and grow into mountains, and the recipient can rarely remember the situation well enough to do anything with it.
Don't forget positive feedback, and make it specific too. Good meeting teaches nothing; when you opened by showing the decision question, we got focused right away tells the person exactly which behavior is worth repeating. Specific positive feedback is also the most powerful tool for reinforcing what already works, and it makes it easier for the recipient to accept corrective feedback the day it is needed.
How to practice this
This research also sets a standard for the feedback you receive when you practice. In Project-simulator, the debrief after every practiced conversation is built on exactly these principles: every assessment is tied to a verbatim quote from your own conversation and followed by a concrete suggestion for improvement, instead of general judgments. If you want to feel the difference between vague and actionable feedback, you can try it free for a week.
In Project-simulator you practice these conversations against an AI counterpart and get feedback grounded in research like this.
Try free for 7 days