How to Give and Receive Peer Feedback on Writing

Useful feedback tells a writer what happened as you read – not whether you liked it.

Peer review is built into a great many courses, and it is usually done badly – not through unkindness but through vagueness. “This was really good, maybe check the commas” helps nobody. Learning to give specific, structured feedback makes you a markedly better editor of your own work, which is the real reason the exercise exists.

Read once before commenting

Read the whole piece through without a pen. Commenting as you go produces a list of local irritations – a clumsy sentence here, a missing comma there – while the significant issues are structural and only visible once you have seen the whole. After the first read, write down your overall impression in two sentences. That impression is the most valuable thing you can give the writer, and it disappears if you start marking commas first.

Comment on the big things first

Work in order of importance, not in the order things appear on the page. There is no point polishing sentences in a section that should be deleted.

• Argument: is there a clear thesis, and does the piece deliver it?
• Structure: does the order make sense, and does each section earn its place?
• Evidence: are claims supported, and is the support explained?
• Paragraphs: one idea each, opening with a claim?
• Sentences and mechanics: last, and only if there is time.

Describe your experience as a reader

The most useful feedback reports what happened while you read rather than delivering a verdict. “I lost the thread of the argument on page three” is more useful than “the argument is unclear”, because it locates the problem. “I expected the counterargument here and it did not come” tells the writer something specific about their reader’s expectations. This framing is also easier to hear, since it describes an effect rather than pronouncing a judgement.

Be specific and point to the text

Every comment should be attached to something concrete. Instead of “some parts are too wordy”, mark the sentence and say why. Instead of “needs more evidence”, identify the claim that is unsupported. Vague feedback leaves the writer guessing, and a writer who cannot locate the problem usually changes something else at random.

Say what works, and say why

Praise is not politeness; it is information. If a writer does not know which parts are working, they may well edit out the best thing in the piece. But keep it as specific as the criticism: “the opening example made the abstract point immediately concrete” tells them what to do again, while “great intro!” tells them nothing.

Ask questions instead of rewriting

Resist the urge to rewrite someone’s sentences into your own voice. It is their piece, and imposing your style teaches them nothing. Questions work better: “did you mean X or Y here?” or “what makes this the strongest of your three reasons?” A question invites the writer to solve the problem themselves, which is how the skill actually transfers.

Receiving feedback well

The instinct on receiving criticism is to explain what you meant. Suppress it, at least until the reviewer has finished. If a reader misunderstood, that is data – your explanation is not available to a marker reading the submitted version. Listen, take notes, and ask clarifying questions rather than defending choices.

Afterwards, look for patterns. One reader finding a section confusing may be a mismatch of expectations; three readers stumbling in the same place is a problem in the text. Weight repeated comments heavily, and treat single idiosyncratic reactions with more caution.

Decide what to act on

You are not obliged to accept every suggestion, and trying to satisfy every reviewer produces incoherent writing. Judge each comment against your purpose and the assignment brief: does acting on this serve the argument you are making? Where a reviewer identifies a real problem but proposes a solution you dislike, keep the problem and find your own fix – the diagnosis is usually more reliable than the prescription.

Agree what kind of feedback is wanted

Before exchanging drafts, ask the writer what stage the piece is at and what they want looked at. An early draft needs structural comment and nothing else; a near-final version may genuinely need line editing. Reviewing a rough draft for commas wastes both people’s time, and critiquing the structure of a piece due in two hours creates anxiety without a remedy. One question at the start – “what would be most useful?” – makes the whole exchange more valuable.

Reciprocate honestly

Peer review depends on both parties taking it seriously, and the temptation to trade easy praise is strong when you will see each other in class next week. Resist it. A reviewer who tells you the piece is fine when it is not has cost you the one chance to fix it before it is graded. The kindest thing you can do for someone’s work is to read it properly and say what you actually saw – delivered specifically and without hostility, that is a service rather than an attack.

Adapt to the stage of the draft

Feedback is only useful if it matches what can still be changed. An outline needs comment on argument and structure; an early draft needs the same plus development; a near-final draft needs clarity and mechanics. Line-editing an outline wastes both people’s time, and raising structural doubts two hours before a deadline creates anxiety without a remedy.

Ask what stage the piece is at before reading, and calibrate accordingly.

Use a structure so nothing is missed

Unstructured feedback tends to fixate on whatever the reviewer happened to notice first. A simple frame keeps it balanced: state the overall impression, identify what is working and why, raise the two or three most significant issues in order of importance, then note smaller points.

• Lead with your reading experience, not a verdict.
• Limit major points to three — more cannot be acted on.
• Anchor every comment to a specific passage.
• End with a question that helps the writer decide what to do next.

Give feedback that is actionable

The test of any comment is whether the writer knows what to do with it. “The argument is unclear” fails; “I could not tell whether you were arguing that the policy failed or that it was never properly implemented” succeeds, because it identifies the ambiguity precisely enough to resolve.

Handle disagreement between reviewers

Where two readers give contradictory advice, the writer must judge rather than average. Weight comments by how well they connect to the assignment brief and by whether the reader identified a problem or merely expressed a preference. Contradictory reactions to the same passage often mean the passage is genuinely ambiguous, even if neither proposed fix is right.

Use feedback to improve your own drafting

The largest benefit of peer review is not the specific comments but the pattern across them. If three reviewers over a term all note thin evidence, that is a drafting habit rather than a one-off. Keeping a short record of recurring feedback turns individual corrections into a genuine improvement in how you write, which is the point of the exercise.