Feedback: the modern model#

You hand in a piece of work. A week later it comes back with a number on the top—7 out of 10, or a B+—and maybe the word “good” in the margin. You file it away and move on. Be honest: what did you actually learn from that? Almost nothing. You know roughly where you stand, but not what to do next. That number is one of the weakest forms of feedback there is, and most of us have spent years mistaking it for the real thing.

Feedback is one of the most powerful influences on how well anyone learns—when it’s good. But a great deal of what gets called feedback isn’t, and some of it actively makes learning worse. So it’s worth being precise about what separates feedback that moves you forward from feedback that just tells you a score.

Feedback answers three questions#

The cleanest way to think about feedback comes from John Hattie and Helen Timperley, who pulled together a huge body of research and found that genuinely useful feedback answers three questions, in order:

Where am I going? What does success on this actually look like? (They call this feed-up.)
How am I doing? Where am I now, compared with that goal?
Where to next? What’s the specific move that closes the gap?

Notice that a grade only ever gestures at the middle question, and badly. “7 out of 10” tells you you’re somewhere short of the goal, but not what the goal really was or what to do about it. The third question—where to next?—is the one that does the heavy lifting, and it’s the one cheap feedback almost always skips. When you seek feedback, you are really hunting for an answer to that question.

The same researchers also found that feedback works at different levels, and the level matters enormously:

Task level — “This conclusion isn’t supported by your second example.” Useful and correctable.
Process level — “You’re checking each step against the rule; try working backwards from the answer instead.” This is the high-value zone, because it transfers to the next task.
Self-regulation level — feedback that helps you monitor and steer your own learning . The most powerful kind, because eventually you become your own source of it.
Self level — “You’re so clever.” This is the one to be wary of, and the next section explains why.

Aim it at the work, not the person#

Here is the finding that should change how you give and ask for feedback: praise directed at the person—“you’re a natural,” “you’re brilliant at this”—does essentially nothing for learning, and can quietly undermine it. A large review by Kluger and DeNisi found that more than a third of carefully studied feedback interventions actually reduced performance, and a common culprit was feedback that pulled attention onto the self rather than the task. When the message is about you, your ego steps in front of the work—defensive about criticism, complacent about praise, and either way no longer thinking about the task.

Feedback about the task and the process—what worked, what didn’t, and what to try next—keeps your attention exactly where the learning happens. So when you ask a teacher, coach, or colleague for feedback, steer them off “was it good?” and onto “where specifically did it break down, and what would you change?” And when you give feedback to someone else, talk about the work in front of you, not the kind of person they are.

This is also the quiet reason grades teach so little. In a now-classic experiment, students given only comments improved on the next task; students given a grade—or even a grade plus the same comments—did not. The number swallowed the comments: once a score is on the page, attention snaps to “what did I get?” and the useful information goes unread. If you have any say in it, read the comments before you see the mark.

Timing, and the slow handover#

Feedback has a shelf life. To be useful it has to arrive while you can still do something with it—while the task is live and the reasoning is fresh in your head. Feedback that turns up two weeks later, on work you’ve mentally closed the book on, has lost most of its force no matter how thoughtful it is. This is why building feedback into your practice beats saving it all for the end: check your answer against the worked solution right after you attempt the problem, not after you’ve done forty more.

But—and this is the kernel of truth in the old advice not to lean on feedback too hard—you cannot stay on a drip forever. Picture learning to fly. Early on, you want an instructor beside you, catching errors before they become dangerous. If that instructor keeps prompting you every single time you drift low or slow, though, you never learn to notice the drift yourself—and one day you’ll be flying alone. Good instruction fades the scaffolding on purpose: more correction at the start, steadily less as you take over the monitoring. The aim isn’t to be fed answers indefinitely; it’s to internalise the feedback until you can generate it yourself—which is exactly the self-regulation level, the most powerful kind there is.

Make feedback something you do, not something you wait for#

Most people treat feedback as weather—it happens to you, and you hope it’s favourable. Flip that. Feedback is a thing you go and get, deliberately, while it can still change what you do. A few habits make it work:

Decide the goal first. You can’t get a useful “how am I doing?” until you and your reviewer agree what “done well” looks like. Pin down the target before you ask.
Ask for the next move, not a verdict. “What’s the one thing I should change?” beats “Is this good?” every time.
Get comments before grades. Read the commentary, decide what you’ll change, then look at the number.
Build in your own loops. Self-test and check against a model answer; compare your work to a strong example; explain your reasoning aloud and listen for where it wobbles—feedback you can run alone, on demand. Doing this well depends on judging your own work honestly, which is its own skill: see metacognition .
Then act on it. Feedback you receive and don’t use isn’t feedback—it’s just information. The gap only closes when the next attempt differs from the last.

Takeaway#

Good feedback answers three questions—where am I going, how am I doing, where to next—and the last one matters most. Keep it on the task and the process, never on the person; a number on the top of the page is the weakest feedback there is, so chase the comments instead and read them before you see the grade. Get it while you can still act, build small feedback loops into your own practice rather than waiting for someone else’s, and treat the whole thing as a handover: the goal is to internalise it until you become your own best source of it.