People who keep saying ‘good job’ aren’t bad teachers — the real fix is more specific than kindness

You just said “good job” again and watched it slide right past. The learner nods, glances away, and makes the same mistake a minute later. So what actually pushes someone forward — the warm word, or something more specific? Decades of studies keep pointing at the specific thing, and the gap between the two is wider than most of us expect.

Before the sections open, the figures this page stands on — each one carrying its own source.

Why “good job” bounces off a learner

Praise feels good to give and good to hear. The trouble is that “good job” carries no information the learner can act on. It does not point at what worked, what failed, or what to change on the next try. So the learner ends up right where they started, plus a small warm feeling that fades fast.

Feedback that teaches does the opposite. It names the specific move, links that move to the result, and tells the learner what to repeat or fix. That is the difference researchers keep finding when they line up a bare grade or a compliment against detailed comments tied to the task itself.

Think about the last time someone told you “nice work” on something hard. It probably felt pleasant and forgettable. Now think about a time someone showed you the one line that broke your code, or the single step you skipped in a proof. That stuck, because you could do something with it. The warm word aimed at you. The useful comment aimed at the task.

What “The Power of Feedback Revisited” actually measured

In 2020, Wisniewski, Zierer and Hattie pooled hundreds of feedback studies covering tens of thousands of learners into an updated review called “The Power of Feedback Revisited.” The headline was not that feedback is magic. Averaged across everything, its effect came out solid but middling, and the results swung widely from one study to the next.

That swing is the interesting part. It means feedback is not one thing with one reliable payoff. What you actually say, and what it points at, decides whether it helps at all. High-information feedback — the elaborated kind that describes the task and how to improve on it — pulled far ahead of mere praise or a grade on the page. A number scrawled at the top of an essay tells you where you stand. It leaves out what to do next.

The review also found feedback does its clearest work on cognitive skills, like solving and reasoning, and on physical or motor skills, like a movement you are drilling. Its pull on motivation and behavior was noticeably weaker. Put plainly, feedback teaches the task far better than it lifts the mood around the task.

The contrast between the two kinds of feedback gets sharp once you set them side by side.

The same page reads differently depending on where you stand. Tell it where you stand.

Does guessing before a lesson help you learn it?

The same theme — specific beats general — turns up in a surprising place: guessing. A preregistered review looked at prequestions, where learners are asked to guess answers before they have studied anything. The idea sounds risky, since most of those early guesses are simply wrong.

The wrong guesses turned out not to matter much. Guessing gave a clear lift to the exact facts people had been quizzed on beforehand. It did almost nothing for the rest of the lesson sitting right alongside those facts. The act of guessing works like a spotlight, focusing attention narrowly on whatever was asked and leaving everything else in shadow.

That reads as a warning as much as a tip. Prequestions can sharpen recall of the pieces you target, so choose those pieces with care. Anything outside the spotlight barely benefits. The result mirrors the feedback finding almost exactly: a narrow, specific prompt drives the gain, and a broad one leaves most of the material untouched.

Can targeted cues during sleep strengthen memory?

Specificity even reaches into sleep. Researchers have tested targeted memory reactivation, which pairs something you are learning — a word, an image, a task — with a sound or a scent, then replays that cue quietly while you sleep. The brain appears to rehearse the linked memory in the background.

A meta-analysis gathering dozens of these experiments found a small but dependable memory gain from the replayed cues. The effect was strongest during deep, non-REM slow-wave sleep. It disappeared during REM sleep, and it did nothing when the cue was played to people who were awake.

Notice the pattern one more time. The benefit is tied to a specific cue, for specific material, delivered in a specific stage of sleep. A vague nudge does not do it. The gain is modest, which makes it a supplement rather than a shortcut. Precision is what it rewards.

A page should show what it grows from. Here is the tree — every leaf quoted, every receipt attached.

What feedback, guessing, and sleep cues have in common

Three different corners of learning research — classroom feedback, pre-lesson guessing, and memory cues during sleep — keep landing on the same point. Learning responds to specific, information-rich input aimed at a particular task or fact. It responds far less to broad, feel-good, or unfocused input.

That reframes what praise is for. A kind word can steady someone, build a relationship, and make hard practice feel safe. Those things carry real value. They are simply doing a different job than teaching the skill. When the goal is a genuine gain in what someone can do, the effective move is to hand them information they can act on.

This also explains why “try harder” and “you’ve got this” feel hollow after a while. They aim at effort and mood, which the feedback review pinned as exactly where feedback has the least grip. Point at the task instead, and you are working with the part of learning that responds the most.

What to say instead of “good job”

Turning this into words is the whole game. The shift is always from a verdict to a description: name the specific move, tie it to the result, and say what to do next. A few before-and-after lines show the change in action.

Where you would reach for “Great work!”, name the mechanism instead: “Your second paragraph backs the claim with a concrete example, and that is what made the argument land.”

In place of “You’re so smart,” point at the behavior that paid off: “You checked your answer against the question before moving on, and that step caught the mistake.”

Swap the empty “Nice job on the test” for something the learner can use: “You lost marks where the steps were skipped. Write each step out next time and those marks come back.”

For a physical skill, drop “Keep it up” in favor of the specific cue: “Your grip loosened on that last swing and the ball drifted off line, so hold the grip firm and repeat it.”

Who does this, and what can they do today? A parent can pick one specific thing from tonight’s homework — a step, a choice, a correction — and name it out loud rather than saying the work was good. A teacher can take the next stack of papers and replace one grade-only mark with a single sentence pointing at the exact step that worked or failed. A coach or trainer can film one repetition at the next session and point to the single movement that changed the result, since feedback does some of its clearest work on physical skills. A student studying alone can write one line about what specifically went right or wrong on the next practice problem before opening the answer key.

If you want a repeatable move for any of these moments, the sequence stays the same every time.

Common questions about feedback and learning

A few points come up again and again once people start trading praise for task-tied feedback.

Remember the note you left? It has been waiting for you.

Vague praise versus feedback tied to the task

• Information to act on — None, just a verdict — Names the specific move and its result
• What it aims at — Mood and effort — The skill or the answer itself
• What the learner does next — Unclear — Repeat what worked, fix what failed
• Where the research shows the strongest pull — Weakest on motivation — Strong on thinking and physical skills

Turning praise into feedback that teaches

• Name the specific move the learner made, whether it was the paragraph, the step, or the grip.
• Tie that move to the result it produced, good or bad.
• Say the one thing to repeat or change on the next attempt.
• Aim it at the task or the skill, never at effort or personality.
• Keep it short enough that the learner can act on it right away.

And if a nerve got touched just now, the help below is real and free, open at any hour.

Common questions about feedback and learning

• Does this mean I should never praise anyone? — No. A kind word can build trust and make hard practice feel safe, and that has real value. It just does a different job than teaching a skill. Use praise to steady someone, and use specific feedback to move the skill forward.
• Is more feedback always better? — Not automatically. The feedback review found results swinging widely from study to study. What mattered was the information content, since elaborated, task-focused comments beat a bare grade or a compliment whatever the amount.
• Why did guessing before a lesson only help some of the material? — Guessing focuses attention narrowly on whatever was asked. The facts people guessed about got a clear boost, while the rest of the lesson sitting right alongside barely moved. So point your prequestions at what matters most.
• Can I really learn in my sleep? — Only in a narrow way. Replaying a cue linked to something you already studied gave a small, dependable memory gain, and only during deep non-REM sleep. It supports studying rather than replacing it.

And before you close the tab, fold this page into one sentence of your own — the when and the how, decided now.

Keeping faith with every receipt above: here is how the page itself works, device by device.

This is general information about the mind, not therapy or a diagnosis. If things feel hard, please consult a professional. In a crisis, reach a free, confidential crisis hotline right away; findahelpline.com lists one for your country.