After weeks of waiting, the decision arrives: major revision. The editor's letter and two or three reviewers' reports are attached. Your response letter is now as important as the revised manuscript, because the editor often reads it first and decides how seriously to take your changes. This guide shows how to structure it, with a template and common mistakes.
Short answer: answer every comment, one by one, in the same order, quote the comment, say what you changed and where, and disagree politely with evidence when you must. Never skip a comment, even a small one.
The principles
- Be grateful and brief at the start. One short paragraph thanking the editor and reviewers is enough.
- Respond to every point. A skipped comment looks like an ignored comment.
- Make it easy to check. Say exactly where each change is (section, page or line number), and use tracked changes or highlights in the revised manuscript.
- Stay polite, even when you disagree. Explain your reasoning and add evidence or a reference. Offer a smaller change, such as a clarification or a limitation sentence, instead of a flat refusal.
- Do not invent. If you did not run the extra analysis, say what you did instead and why.
A structure that works
- A short cover paragraph to the editor with a summary of the main changes.
- Then, for each reviewer, a heading ("Reviewer 1"), followed by each comment numbered (R1.1, R1.2 ...).
- For each comment: the comment in italics or a quote block, then your response, then the change made and its location.
- A closing line that thanks everyone again.
Template for one comment
Comment R1.3: "The sample size is small and the claim about X is not supported."
Response: We thank the reviewer for this important point. We agree that the sample (n = 40) limits the strength of the claim. We have (a) added a power analysis in Methods, (b) softened the statement about X in the Abstract and Discussion, and (c) added this limitation to the Limitations paragraph.
Changes: Methods, page 6, lines 142 to 150; Discussion, page 11, lines 310 to 318.
When you disagree with a reviewer
- Thank them for raising the point, and show that you understood it.
- Explain your position in two or three sentences, with data or a reference.
- Say what you changed anyway, for example added a sentence that clarifies the issue so readers do not make the same objection.
Reviewers rarely expect you to accept every suggestion, but they do expect a reasoned answer.
Common mistakes
- Answering with "Done" or "Corrected" and no detail on what was done or where.
- Changing something in the manuscript but not mentioning it in the letter, or the reverse.
- Responding emotionally to a harsh comment. Wait a day and rewrite.
- Page and line numbers that no longer match the revised file.
- Treating all comments equally. Spend most effort on major concerns, and handle typos in one line.
Where AI can help, and where it must not
Many authors lose the most time on the mechanics: splitting the reports into numbered comments, keeping the order straight, and drafting a first version of each reply. AI can do this part faster. It can parse the reviewers' comments, group them by reviewer, propose a draft response to each one, and assemble a formatted letter. ProfAgent's Revision Assistant does this and exports the letter to Word.
AI must not decide what is true about your study. You should check every draft against your data and your manuscript, and never let a tool claim an analysis you did not run. Also check the journal's policy on AI use, and do not upload a manuscript to a tool that you are not allowed to use for confidential material.
A checklist before you submit
- Every comment has a numbered response.
- Every change in the letter exists in the manuscript, and the page or line numbers are right.
- The tone is polite throughout. Ask a colleague to read it.
- The tracked-changes or highlighted file and a clean version are both ready.
- You have read the editor's instructions on file format and deadline.
Try ProfAgent free
One AI for professors and researchers: reviews, revisions, grants, supervision, literature and recordings. No credit card required.
Start FreeFrequently asked questions
How long should a response to reviewers letter be?
As long as it needs to be to answer every comment clearly. Most letters run from a few pages to more than ten. Do not shorten answers by skipping points, but keep minor corrections to one line.
Should I copy each reviewer comment into the letter?
Yes. Quoting each comment, followed by your response and the change made, lets the editor and reviewers check your work without searching through their reports.
What if I disagree with a reviewer?
Thank them, explain your reasoning with evidence or a reference, and say what you changed to address the concern, such as a clarification or a limitation sentence. A polite, reasoned answer is accepted far more often than a flat refusal.
Can I use AI to write my response to reviewers?
AI can help with the mechanics, such as splitting comments, ordering them and drafting a first reply. You must check every statement against your data and manuscript, and follow the journal's policy on AI use.
Should I submit tracked changes?
Most journals ask for a marked-up version showing changes and a clean version. Follow the journal's instructions, and make sure the page and line numbers in your letter match the file you submit.