How Do You Respond to Peer Reviewer Comments? A Point-by-Point Guide
How do you respond to peer reviewer comments? You answer every comment, one by one, in a separate response letter: quote each comment in full, say what you changed or why you did not change it, and give the exact page and line where the change appears in the revised manuscript. That letter is not admin. Editors read it as closely as the manuscript itself, and a careless one can sink a paper the reviewers were otherwise willing to accept. This guide covers the shape of a strong response letter, how to agree and disagree without damaging your case, what to do when reviewers contradict each other, and the checklist to run before you resubmit.
This is the fifth piece in a series for clinicians learning to publish. It follows the guides on publishing your first paper, what publishing costs, how long publishing takes, and publishing without a PhD. Those cover the submission sequence; this one covers the moment most first authors find hardest, the revision letter.
What a revision decision actually means
A request for revision is not a rejection, and it is not a promise of acceptance either. It means the editor sees a publishable paper inside your manuscript, provided the reviewers’ concerns are addressed. Nature’s guidance for its journals puts it plainly: authors invited to revise are expected to supply a revised manuscript that addresses the issues raised, a response to each reviewer replying to each point raised, and a cover letter with any additional information the editors requested, submitted through the link in the decision email rather than as a new manuscript.
The labels matter less than the comments. Wiley’s reviewer guidance notes that minor changes are typically assessed by the editor alone, while significant revisions usually go back to the original reviewers. But journals use “major” and “minor” inconsistently, and a “minor revision” with twenty fiddly points can take longer than a “major revision” with one clear fix. Read every comment before you plan your time, and treat the editor’s own comments as the priority list: they usually signal which reviewer points matter most for the decision.
Step one: read the decision letter properly
Before you touch the manuscript, extract four things from the decision letter. First, the deadline. Journals set revision deadlines, and if you need more time, ask the editorial office before the deadline passes rather than after; a late, unexplained resubmission reads as lack of enthusiasm. Second, the files the journal wants: typically a clean manuscript, a version with tracked changes or highlighted changes, and the separate response letter. Third, any comments from the editor, which may override or qualify reviewer points. Fourth, whether the resubmission goes back to the original reviewers or to the editor alone, which tells you how much detail your letter needs.
Then put the reviews aside for a day or two. Taylor and Francis advise authors to let the first reading settle before responding, so the reply is written objectively rather than defensively. That pause is not procrastination; it is the difference between a letter that argues with reviewers and one that answers them. A reviewer who misunderstood your methods has still told you something useful: a future reader may misunderstand them the same way, and clarifying the manuscript fixes both problems.
Step two: split every review into numbered points
Reviewers do not always write in neat numbered lists. Some write narrative reports where one paragraph contains three separate criticisms. Your job is to break each report into individual, numbered points yourself, so that nothing is missed and nothing is silently dropped. Springer Nature’s editorial guidance recommends copying the entire text of each review, without leaving anything out, because selective quoting makes reviewers suspect you skipped the hard parts.
Number the points under each reviewer separately, Reviewer 1 point 1, point 2, and so on, and keep the same numbering in your letter and in your own working notes. A simple table works well for planning: comment number, the reviewer’s comment, your response, and the page and line where you addressed it. Deal with the small points first if it helps momentum, typos, consistency, reference formatting, but do not mistake clearing the easy points for finishing the job. The difficult points are the ones the decision turns on.
The shape of a strong response letter
Every effective response letter has the same three parts. First, a short cover paragraph: thank the editor and reviewers, state that all comments have been addressed, and summarise the most important changes in two or three sentences. Do not oversell and do not list everything here; the editor reads this paragraph in seconds before moving to the detail.
Second, the point-by-point section, which is the bulk of the letter. Under each reviewer, quote each comment in full, then answer immediately below it. Make the reviewer’s words and your words visually distinct, by bolding, italics, colour or indentation, so an editor can scan the document without hunting for where your voice starts. For every change, give the exact location in the revised manuscript, by page and line number where the journal uses them, or by section and paragraph where it does not.
Third, the revised manuscript itself, with changes marked. Journals differ on the exact method, tracked changes or highlighted text, but the principle is the same: the reviewer should be able to see what changed without comparing two documents side by side. Nature Methods’ long-standing advice is that modifications should be underlined in the manuscript and accompanied by a point-by-point response, and that the best responses copy the reviewers’ report and insert an answer to each point, explaining what was done and where the change can be found. State in your letter which marking method you used.
When you agree with the reviewer
Most comments fall here, and the formula is simple: acknowledge, change, locate. “We agree with the reviewer. We have added the missing effect size and 95% confidence interval to Table 2 and now report it in the Results (page 6, lines 12 to 15).” Short, verifiable, done. Do not thank the reviewer separately in every single response; repeated thanks in every answer reads as padding, and editors notice. One genuine thanks at the top of the letter is enough.
Where a comment reveals a genuine gap, say what the change improves, not just that you made it. If the reviewer asked for a limitation you had not stated, add it and note where. If the reviewer spotted an error, correct it and say so plainly. Owning an error costs nothing and buys credibility for the points where you disagree.
When you disagree with the reviewer
Disagreement is legitimate and expected. What matters is how it is done. Never simply write “we disagree.” Explain your reasoning with evidence or a citation, and where possible make a small concession that shows you took the concern seriously even while holding your position, for example by adding a sentence justifying your choice in the Methods and noting the alternative as a limitation.
A usable shape is: “We acknowledge the reviewer’s point, but respectfully disagree for the following reasons.” Then give the reasons, briefly and with a reference where one exists. Nature Methods’ advice is blunt on this point: tackling the scientific substance comprehensively, usually by adding data, is far more effective than dismissing a concern, saying it is beyond the scope of the paper, or making a cosmetic change. If you cannot add the requested data, say why, and offer the nearest alternative you can defend.
Keep the disagreement about the science, never about the reviewer. Do not write that a reviewer failed to read the paper, misunderstood basic statistics, or asked for something absurd, even if you believe it. The editor, who chose that reviewer, reads your letter too. The sentence that sinks most letters is the one that argues tone rather than substance.
When two reviewers contradict each other
Contradictory reviews are common, and they are the editor’s problem to resolve, not yours to hide. Address both comments in full, explain the choice you made and why, and let the editor arbitrate. A useful formulation is: “Reviewer 1 suggested X and Reviewer 2 suggested the opposite. We have adopted X because [reason], and we have added a sentence in the Discussion acknowledging the alternative approach suggested by Reviewer 2.”
Do not silently follow one reviewer and ignore the other. The ignored reviewer will see the resubmission, or the editor will notice the gap, and either outcome is worse than an explicit, reasoned choice. If you genuinely cannot decide, say so in your confidential comments to the editor, if the journal provides that box, and ask for guidance. That is what the box is for.
When a reviewer asks for work you cannot do
Sometimes the requested analysis needs data you do not have, an experiment your ethics approval does not cover, or a sample size you cannot reach. Do the work if it is feasible; reviewers can tell the difference between a hard fix and an avoided one. If it is not feasible, explain tactfully why it cannot be performed, what you did instead, and what limitation you have added to the manuscript to cover the gap.
Three honest options cover most cases. One, the analysis is possible with existing data and you do it. Two, the analysis is not possible, so you add the point as a stated limitation and, where appropriate, as a direction for future research. Three, the request rests on a misunderstanding, in which case you clarify the manuscript text that caused the confusion and point the reviewer to the existing evidence. What you must not do is pretend the work was done, cite a source you have not read, or claim a limitation covers a flaw it does not.
A worked example: three exchanges
Consider a junior doctor revising a small cohort study after a major revision decision. Reviewer 1 writes: “The sample size justification is not provided.” A strong response: “We agree. We have added an a priori sample size calculation to the Methods (page 5, lines 8 to 14), based on the primary outcome and the effect size reported in the reference cohort study cited there.”
Reviewer 2 writes: “The authors should have used a random-effects model.” A strong response: “We respectfully disagree. Heterogeneity was low (I squared 12%), which supports the fixed-effect model under the Cochrane Handbook guidance cited in the Methods. We have added a sentence explicitly justifying this choice (page 8, lines 4 to 7) and noted the random-effects alternative as a limitation.”
Reviewer 2 also writes: “The authors should recruit a further 200 patients.” A strong response: “We agree this would strengthen the findings, and we have added this as a limitation and a direction for future research (Discussion, page 11, lines 20 to 24). Additional recruitment is not possible within this retrospective cohort, whose data collection closed before submission, and we have stated this explicitly so readers can judge the findings accordingly.” Each answer quotes, decides, evidences and locates. That is the whole genre.
Before you resubmit: the checklist
Run this list before you upload anything. Every reviewer comment has a numbered answer, including the editor’s own comments. Every claimed change is actually in the manuscript, at the page and line you cited. Changes are marked in the tracked or highlighted version, and a clean version is also supplied if the journal asks for one. Co-authors have read and agreed the response letter, not just the manuscript, because the letter commits all of them to its claims. The resubmission uses the journal’s revision link, not a fresh submission, so the paper keeps its history. And the deadline is met, or an extension has been agreed in writing.
Proofread the letter as carefully as the manuscript. A response letter with the wrong reviewer numbers, a page reference that points nowhere, or a promised change that is missing from the text tells the editor the revision was rushed, and that impression colours the whole resubmission. If your journal allows confidential comments to the editor, use them sparingly: for genuine conflicts between reviewers, for context the reviewers should not see, and for nothing else.
What happens after you resubmit
Minor revisions are usually assessed by the editor alone. Major revisions usually go back to the original reviewers, who check whether their points were addressed rather than re-reviewing the paper from scratch. Wiley’s guidance notes that reviewing a revised manuscript should generally be faster than the initial review, because the main task is confirming the requested changes were made, though new issues can still arise and another round of revisions is possible.
Editors will not send a resubmitted paper back to referees if it seems the authors have not made a serious attempt to address the criticisms. That is the real risk of a thin response letter: not a hostile reviewer, but an editor who decides the revision does not merit another round. Conversely, a thorough letter with a well-marked manuscript often returns a decision quickly, because you have done the editor’s comparison work for them.
If the decision is rejection after revision, an appeal is possible but narrow. Nature’s appeals policy states that decisions are reversed only if the editors are convinced the original decision was an error, for example where a reviewer made substantial errors of fact or there is significant evidence of bias, and only where reversing that reviewer’s opinion would have changed the original decision. An appeal that simply restates your disagreement with a reviewer’s judgement will not meet that bar. Most rejected papers are better served by revising further and submitting elsewhere, with the reviewers’ comments already addressed.
Frequently asked questions
Do I have to agree with every reviewer comment?
No. You must answer every comment, but you may respectfully disagree where you have evidence or a reasoned argument. What you cannot do is ignore a comment, or disagree without explaining why. Editors expect reasoned disagreement; it is part of scholarly exchange.
How long should a response letter be?
As long as the reviews require. A minor revision may need two pages; a major revision with three reviewers may need fifteen. Length is not a virtue, but completeness is: every comment answered, every change located. A short letter that skips hard points is worse than a long letter that answers them all.
Should I quote the reviewers’ comments in full?
Yes. Springer Nature’s editorial guidance recommends copying the entire text of each review without leaving anything out, so reviewers can see nothing was selectively skipped. Quote in full, then answer directly beneath, with your words visually distinct from theirs.
What if the reviewer is rude or factually wrong?
Answer the substance and ignore the tone. If a comment contains a factual error, correct it politely with evidence. If a review is abusive, raise it in your confidential comments to the editor rather than in the response letter itself. Never answer rudeness with rudeness; the editor reads both documents.
Can I ask for more time to revise?
Yes, by contacting the editorial office before the deadline passes and explaining briefly why more time is needed. Journals routinely grant extensions for data analysis, co-author availability or clinical duties. An agreed extension is normal; a silent late resubmission is not.
What if two reviewers tell me opposite things?
Answer both, explain the choice you made and why, and note the alternative in the manuscript where appropriate. Contradictory advice is the editor’s to resolve, and an explicit, reasoned choice helps the editor do that. Silently following one reviewer and ignoring the other is the one approach that reliably backfires.
Sources
Nature, Writing a response to the reviewers’ comments: https://www.nature.com/documents/Effective_Response_To_Reviewers-1.pdf . Nature Methods, A rough guide to publication: https://www.nature.com/articles/nmeth0907-675 . Nature Materials, Editorial process and peer review: https://www.nature.com/nmat/submission-guidelines/editorial-process . Nature Computational Science, How to respond to reviewers: https://link.springer.com/article/10.1038/s43588-025-00931-5 . Wiley, Reviewing revised manuscripts: https://authors.wiley.com/Reviewers/journal-reviewers/how-to-perform-a-peer-review/reviewing-revised-manuscripts.html . Noble et al., Improving the chances of manuscript acceptance: how to address peer reviewers’ comments: https://pmc.ncbi.nlm.nih.gov/articles/PMC3242570/ . Annesley, Critical tips on how to respond to peer reviewers: https://pmc.ncbi.nlm.nih.gov/articles/PMC8971667/ .
Related reading
How to publish your first research paper: a step-by-step guide for clinicians. Is it free to publish a research paper? APCs, waivers and no-fee journals. How long does it take to publish a research paper? Realistic timelines. Can I publish a paper without a PhD? Yes, here is how. Finding guidance as an international medical graduate: the mentor I never had.