How to respond to reviewer comments, with a response letter

This guide is for researchers at any stage handling their first or fiftieth revision. It is the pillar of our series on reviewing and publishing, which also covers how to turn a thesis into a journal article.

How to read a decision letter

Start by finding the decision, the deadline and what the editor wants most. The editor's own comments matter more than any single reviewer's.

A major revision is an editorial decision that a paper may be publishable after substantial changes, which will usually be checked by the reviewers again. It is not an acceptance, but it is a good result: the editor thinks the paper can get there.

Read the letter once for the outcome, then put it away for a day. Elsevier's advice to authors is to take a break before replying, but not to let the revision drag on [2]. On the second read, mark each comment by type.

What goes in a response to reviewers letter?

A response to reviewers has two parts: a short cover letter to the editor and a point-by-point response table. A response to reviewers letter (or rebuttal letter) is the document submitted with a revised manuscript that answers every reviewer and editor comment, states each change made and gives its location.

Noble's ten simple rules for writing a response to reviewers give a concise checklist [1]. The core of them:

• Start with an overview, then quote the full set of reviews.
• Be polite and respectful to all reviewers.
• Respond to every point.
• Begin each reply with a direct answer to the point.
• Do what the reviewer asks whenever possible.
• Be clear about exactly what changed.
• Make the response self-contained, so the editor does not need to open the manuscript to follow it.

How to respond to reviewer comments: step by step

1. Copy every comment into a table. Number them: R1.1, R1.2, R2.1. Split long comments that make several points.
2. Sort each comment into a type. Easy fix (typo, wording, a missing reference), clarification (the reviewer misread something), new work (analysis, data, experiment), and disagreement.
3. Discuss with co-authors. Agree who does each item and which disagreements you will hold.
4. Do the new work first. New analyses can change your results and your replies.
5. Revise the manuscript. Use tracked changes or highlighted text, as the journal asks.
6. Write each reply in three parts: the answer, the change, the location (page and line numbers in the revised version).
7. Write the cover letter summarizing the main changes in four to six bullets.
8. Read the response as the editor would. Elsevier's guidance suggests having someone outside the author team read both the revised paper and the response [2].
9. Submit before the deadline, or ask for an extension before it passes.

The response table: a template

The response table makes your revision easy to check. Use four columns.

Some journals prefer the same content as running text, with each comment in italics and the reply below it. Follow the journal's instructions; the content is the same.

How to disagree with a reviewer politely

Disagree with evidence, not adjectives. Reviewers are sometimes wrong, and editors expect you to say so when it matters, as long as you do it courteously [2].

1. Thank and restate. "We thank the reviewer for raising the question of confounding by income."
2. Explain with evidence. Cite a paper, show an extra analysis, or point to a design constraint.
3. Offer something. A sentence in the limitations, a sensitivity analysis in a supplement, or clearer wording.
4. Assume the misreading was your fault. If a reviewer misunderstood, the text was unclear: fix the text.

Phrases that work:

• "We have considered this carefully and respectfully disagree, for two reasons..."
• "We ran the suggested analysis (Supplementary Table S4). The results are unchanged, so we have kept the original model and report the new analysis as a sensitivity check."
• "This is outside the scope of the current study, which focuses on X. We now mention it as a direction for future work (p. 19)."

Avoid: "The reviewer is wrong", "As we clearly stated", "The reviewer failed to notice". Never argue with a reviewer about their tone. When two reviewers ask for opposite things, choose one, explain why in both replies, and tell the editor in the cover letter.

A sample response to reviewers letter

This sample is invented for illustration. Adapt it to your paper and journal.

Dear Dr Okafor,

Thank you for the opportunity to revise our manuscript, "Peer tutoring and math anxiety in 9th grade: a cluster-randomized trial" (MS-2026-0418), for the Journal of Example Education. We are grateful to you and the two reviewers for their careful reading.

We have addressed every comment. The main changes are:

• We now report the subgroup analysis by school type as exploratory and discuss its low power (Reviewer 1).
• We added six studies published since 2023 and expanded the discussion of how our effect size compares (Reviewers 1 and 2).
• We rewrote the analysis section to name the mixed-effects model and its random effects (Reviewer 2).
• We added a sensitivity analysis excluding the two schools with missing baseline data (Supplementary Table S4). Results are unchanged.

One request we could not meet: Reviewer 2 asked for 24-month follow-up data, which were not collected. We now state this as a limitation and as a priority for future work.

Our point-by-point responses follow, with page and line numbers referring to the revised manuscript with tracked changes. The word count is now 6,850, within the journal's limit.

Yours sincerely, Lena Marsh, on behalf of all authors

Then attach the response table.

How long do you have to respond to reviewer comments?

The decision letter or submission system sets the deadline; it varies by journal and decision type. Minor revisions usually get less time than major ones. If you need longer, email the editor before the deadline, say why (for example, a new analysis) and propose a date. A response that arrives long after the deadline may be handled as a new submission.

A realistic plan for a major revision, as an example:

Common mistakes in a response letter

Most failed revisions fail on process, not science. Watch for these:

• Skipping a comment. Even "We agree; corrected" counts as a reply. A silent omission looks like evasion [1].
• Vague replies. "Comment addressed" tells the editor nothing. Say what changed and where [2].
• Changing the manuscript without saying so. Every change, including ones no reviewer asked for, belongs in the response.
• Page numbers that point to the old version. Refer to the revised manuscript and check the numbers after the last edit.
• Answering in the letter but not in the paper. If a reviewer was confused, the next reader will be too. Put the clarification in the manuscript, then quote it in the reply.
• A defensive tone in the first reply. Editors read from the top. Start with the comments you accepted.

Can an AI reviewer help before you resubmit?

Yes, for checking the revised paper; no, for deciding the response. An AI reviewer is a tool that reads a manuscript and comments on it against review criteria, as a peer reviewer would. Use one as a final paper review before resubmission, to catch unclear passages and unsupported claims the real reviewers may raise again. Don't paste reviewer comments into a chatbot and submit its replies: you must stand behind every sentence, and journals may restrict uploading confidential review material to AI tools.

Check your revision in AI for Research

• Reviewer reviews a manuscript at a level you choose, scores each criterion out of 10 and quotes each weak passage with a suggested fix. Standard review costs 10 credits per 10,000 words; High quality, on Pro, costs 30. On the free plan, a preview of the first 3,000 words with the full scorecard and top 3 comments costs 3 credits. See AI peer review.
• Claim check labels up to 10 claims from a paragraph as supported, partly supported, not supported or no evidence, with the papers. Use it on new claims added in revision. 2 credits.
• Citation check (0 credits) confirms that each reference exists and flags retractions, corrections and wrong years, which matters when you add references in a hurry. See the citation checker.
• Find Papers (0 credits) to find the recent studies a reviewer asked you to cite, filtered by year and study type.
• Projects let co-authors edit or comment on the revision in one place.

Limits: Reviewer reviews your manuscript; it does not read the journal's review or write your replies. It cannot judge whether a reviewer's request is reasonable in your field. That call is yours and your co-authors'.

FAQ

How long do I have to respond to reviewer comments?

The deadline is set in the decision letter or the journal's submission system, and it varies by journal and by decision type. Minor revisions usually get a shorter window than major ones. If you need more time, ask the editor before the deadline and give a reason, such as a new analysis. Asking early gives the editor time to agree a new date; a missed deadline can mean the paper is treated as a new submission.

What does major revision mean?

A major revision decision means the editor will consider a revised version, but the paper needs substantial changes, such as new analyses, a reframed argument or a rewritten method section. It is not an acceptance: the revised paper usually goes back to the reviewers. Still, it is a good outcome. Address every point thoroughly and the paper has a real chance of being accepted.

How do I disagree with a reviewer politely?

Thank the reviewer, show you understood the point by restating it, then explain your reasoning with evidence: a citation, an extra analysis or a design constraint. Offer a partial change where you can, such as a sentence acknowledging the issue as a limit. Avoid words like "wrong" or "misunderstood". If the reviewer misread the paper, assume your text was unclear and clarify it.

What is a rebuttal letter?

A rebuttal letter, also called a response to reviewers or response letter, is the document you submit with a revised manuscript. It lists every reviewer and editor comment, your reply to each, the change you made and where to find it. Despite the name, most of it should agree and explain changes. Only a few replies should disagree, and those politely.

Can an AI reviewer help me respond to peer review?

It can help you check the revised manuscript before you resubmit: an AI reviewer can score the paper against review criteria and point to unclear passages, much as a peer reviewer would. It can also help find papers a reviewer asked you to cite. It cannot decide which comments to accept, run your new analyses or write replies you haven't checked against the paper.

Next steps

• Publishing from your thesis? Read how to turn a thesis into a journal article.
• Before you add new claims in revision, read how to check a claim against the research.
• Try a 3-credit preview of AI peer review on your revised manuscript.

Sources

1. Noble, W. S. (2017). Ten simple rules for writing a response to reviewers. PLOS Computational Biology, 13(10), e1005730. https://doi.org/10.1371/journal.pcbi.1005730
2. Tancock, C. (2018, August 23). "I submitted to journal X but all I got was this lousy revision." Elsevier Connect. https://www.elsevier.com/connect/i-submitted-to-journal-x-but-all-i-got-was-this-lousy-revision
3. AI for Research. AI peer review (Reviewer). https://aiforresearch.app/features/ai-peer-review
4. AI for Research. Pricing. https://aiforresearch.app/pricing