How to Give Effective Feedback: Hattie & Timperley

Effective feedback answers three questions: Where am I going? (feed up), How am I doing? (feed back) and Where to next? (feed forward). This model by John Hattie and Helen Timperley makes feedback concrete and useful for learning — the key is that it carries information about the task, the process and self-regulation, rather than stopping at praise or a grade.

What actually makes feedback good?

Good feedback is not a verdict but information that makes the gap between the current state and the learning goal both visible and bridgeable. In their widely cited review „The Power of Feedback“ (2007), education researchers John Hattie and Helen Timperley showed that feedback is among the strongest influences on school learning — but only when it carries the right information.

How large the effect is depends precisely on that. A large 2020 re-analysis put the average effect of feedback at a medium value (Cohen's d = 0.48). The decisive finding lies in the detail: feedback rich in information — about the task, the process and self-regulation — reached a strong effect (d = 0.99), whereas mere praise, reward or punishment did almost nothing (d = 0.24). For anyone who supports children and teenagers as they learn, that is the core message: what matters is not whether you give feedback, but which kind.

Which three questions does good feedback answer?

Hattie and Timperley frame feedback consistently from the learner's point of view. Effective feedback answers three questions:

• Feed up — „Where am I going?“ The goal and the criteria for success are clear.
• Feed back — „How am I doing?“ The current state is reported in relation to the goal.
• Feed forward — „Where to next?“ There is a concrete idea of how to proceed.

The three unfold their full power only together. A grade without a goal stays abstract; a goal without a sense of where you stand overwhelms; a position without a next step frustrates. Picture a student writing an argumentative essay — all three questions can be played through on her work.

Feed up: clarify the goal

Before you give any feedback, it must be clear what counts. „Write a convincing argument“ only becomes useful once learners know how to recognise a convincing argument — for instance by a clear thesis, by evidence and by a counter-argument that has been addressed. Good success criteria turn a vague assignment into an achievable goal.

Feed back: report where things stand, honestly

This is about the present: what is already working, what is not yet? Effective feed back is specific and descriptive rather than evaluative. „Your thesis in the first paragraph is clear, but the second piece of evidence doesn't support it yet“ says more than „good, but room for improvement“. It describes observable behaviour and the work itself — not the person.

Feed forward: show the next step

The most powerful part is often the look ahead. Feed forward translates the response into an action: „For the second paragraph, find evidence that directly supports your thesis, and cut the example that leads away from the topic.“ Now the student knows exactly what to do next — instead of merely learning that something is missing.

At which levels does feedback work?

Beyond the three questions, Hattie and Timperley distinguish four levels that feedback can target — with very different effects:

1. Task: Is the result right or wrong? Corrective feedback of this kind helps but says little about how to get there.
2. Process: Which strategy leads to the goal? Feedback on the working method helps learners avoid the same mistakes in future.
3. Self-regulation: How do learners steer, monitor and correct their own approach? Feedback at this level strengthens metacognition — the ability to think about one's own thinking — and makes learners more independent.
4. Self (the person): Praise such as „you're so clever“ targets the person, not the work.

The process and self-regulation levels are considered especially effective, because they enable learners to help themselves next time. Pure task feedback, by contrast, stays tied to the single case, and feedback at the personal level contributes least to learning.

Why is praise alone not good feedback?

Here comes the uncomfortable part: feedback is not automatically positive. One of the most comprehensive meta-analyses (Kluger and DeNisi, 1996, drawing on hundreds of studies) did find an improvement on average — but in more than a third of cases feedback actually made performance worse. Feedback that directs attention to the self rather than to the task is especially risky.

This is exactly where bare praise belongs. „Very good!“ or „you're a maths talent“ contains no information about what succeeded or how to continue — and it can even hold learners back: those praised for their ability later tend to avoid challenges so as not to endanger the label. That does not mean encouragement is wrong. It should accompany substantive feedback, not replace it. How closely feedback and drive are linked also shows up in the topic of how to motivate kids to learn: what motivates most is the experience of genuinely getting better through your own next step.

How do you give learning-focused feedback in practice?

A few simple rules follow from the model:

1. Goal first, feedback second. Clarify the success criteria before you evaluate.
2. Describe rather than judge. Name concretely what you see — about the task, not the person.
3. Always give a next step. Every response ends with something doable.
4. Keep it focused. Two or three precise pointers do more than twenty red marks.
5. Make feedback a dialogue. Ask „How do you explain that to yourself?“ so learners take ownership. This fits differentiated instruction, because the sensible next step looks different for every child.

For teachers this has a practical consequence: not every piece of work needs more marking time, but more targeted feedback. A single good feed-forward pointer per text often achieves more than blanket error correction — and it takes less time to write.

Feedback need not always come from outside, either. Testing yourself and checking the answers straight away produces a form of feedback that measurably strengthens learning. That is exactly what LearnCastAI's quizzes and flashcards, generated from your own material, let learners do, so they can keep checking where they stand.

Conclusion

Good feedback is not a verdict but a direction. Answering the three questions — where am I going, how am I doing, where to next — while aiming at the task, the process and self-regulation rather than the person turns a response into a real learning step. Praise warms, but it does not teach; the real power lies in the concrete next step. If you want to create learning material and matching self-tests from your own documents, LearnCastAI can help — but the pedagogical work of giving good feedback remains yours.