How to Respond to Reviewer Comments: Template and Real Strategy

They start writing too early.

When reviews arrive, most authors immediately want to defend the paper. That is understandable and usually counterproductive. The first 24 hours should not be used for drafting replies. They should be used for sorting the comments.

Build a simple table before you write the response letter.

This matters because not all comments deserve the same treatment. Some are manuscript problems. Some are reviewer preference. Some are hidden signs that the paper is still strategically weak.

The safest structure is still the best one.

Use this order

1. short thank-you paragraph to the editor and reviewers
2. one-sentence summary of the biggest revision themes
3. reviewer-by-reviewer responses
4. comment-by-comment numbered replies
5. exact page and line references to manuscript changes

Do not bury the actual work in a vague opening summary. The opening paragraph should stay short.

Template opening

Dear Editor,

We thank you and the reviewers for the careful and constructive evaluation of our manuscript. We have revised the paper substantially in response to the comments. The main changes include: (1) clarifying the primary outcome definition, (2) adding sensitivity analyses for the subgroup comparison, (3) revising Figures 2 and 3 for readability, and (4) narrowing the causal wording in the Discussion. Below, we respond point by point to each reviewer comment and indicate the corresponding manuscript changes with page and line references.

That is enough. Do not write a page of gratitude.

Every strong response usually contains four elements:

1. acknowledgment
2. direct answer
3. action taken
4. location of revision

Basic response template

Reviewer 1, Comment 3:
The sample exclusion criteria are not sufficiently clear.

Response:
We thank the reviewer for identifying this ambiguity. We have now clarified the exclusion criteria in the Methods section and specify that samples were excluded only before unblinding on the basis of predefined quality-control thresholds. This text has been added on page 9, lines 214-223.

This works because it does not force the editor to infer what changed.

Tone is not cosmetic. It changes how disagreement is received.

When you agree with the reviewer

Be brief and specific.

Bad:

We fully agree and have improved the manuscript accordingly.

Better:

We agree that the original wording was too broad. We have narrowed the claim in the Abstract and Discussion and now state that the association was observed in the validation cohort only. See page 3, lines 61-66, and page 14, lines 332-340.

When you partly agree

Do not pretend full agreement if the revision is narrower than the request.

Better:

We agree that the original analysis did not make the robustness of the finding sufficiently clear. While an additional experiment is not feasible within the current study design, we have added two sensitivity analyses and revised the claim to focus on association rather than mechanism. These changes appear on page 11, lines 247-269, and page 15, lines 355-362.

When you disagree

The key is to disagree with the request, not with the reviewer as a person.

Better:

We appreciate this suggestion. We respectfully did not add the proposed experiment because the requested intervention would test a different mechanistic question from the one addressed in the present study. Instead, we clarified the study boundary in the Discussion and added text explaining this limitation on page 16, lines 381-390.

This tone preserves professionalism while still drawing a line.

1. Comments based on misunderstanding

If the reviewer misunderstood something, do not write a smug reply. A misunderstanding is often evidence that the manuscript itself was not clear enough.

Wrong instinct:

The reviewer appears to have misunderstood our methods.

Better:

We appreciate that this point was not sufficiently clear in the original version. We have revised the Methods section to state explicitly that randomization occurred after eligibility screening and before treatment allocation. This clarification appears on page 8, lines 185-194.

That reply solves the paper, not just the rebuttal.

2. Requests for new experiments

This is where many response letters go wrong. Authors either promise too much or reject the request too abruptly.

Use a three-part test:

• Is the request central to the paper's main claim?
• Can it be addressed with existing data or analysis?
• If not, should the claim be narrowed?

If the answer is "the request is reasonable but impossible in revision time," say so directly and shrink the claim if needed.

3. Conflicting reviewer comments

This is where the editor matters most.

If Reviewer 1 asks for a broader discussion and Reviewer 2 says the discussion is already too long, do not try to satisfy both invisibly. Address the conflict explicitly.

Template:

We note that Reviewer 1 requested a broader contextual discussion, whereas Reviewer 2 recommended a more concise Discussion. To balance these requests, we added two short paragraphs on clinical relevance while tightening speculative language elsewhere. We hope this resolves both concerns while keeping the section proportionate.

When the conflict is more serious, state your reasoning and invite the editor to judge the tradeoff.

Use a traffic-light classification when you draft the response.

Green comments

Easy fixes. Clarifications, wording cleanups, figure-label issues, citation additions. Do these first to build momentum.

Yellow comments

Substantive but manageable. Extra analyses, narrower framing, reorganized results, stronger limitations language.

Red comments

These threaten the acceptance path. They usually concern:

• insufficient evidence for a central claim
• weak controls
• inappropriate statistics
• overreach
• mismatch between what the paper says and what it proves

If the revision still has unresolved red comments, the paper is probably not ready to go back without deeper work.

Editors and reviewers like navigation. Make the document skimmable.

Use:

• reviewer headings
• numbered comments
• bold labels for Comment and Response
• page and line references
• quotations or tracked-changes excerpts only when truly helpful

Do not use:

• walls of prose
• sarcastic emphasis
• vague phrases like "done as requested"
• emotional scene-setting

If the reviewer wrote twelve comments, your response should let the editor find all twelve in seconds.

Scenario 1: Clarification request

Comment:
The inclusion criteria for the validation cohort are not clear.

Response:
We thank the reviewer for pointing this out. We have revised the Methods section to define the validation cohort explicitly, including enrollment dates, exclusion criteria, and the minimum follow-up threshold. This text has been added on page 7, lines 161-176.

Scenario 2: Request for more cautious wording

Comment:
The conclusion appears too strong given that the analysis is observational.

Response:
We agree and have revised the wording throughout the Abstract, Results, and Discussion to avoid causal language. We now describe the findings as associations and added a sentence in the Discussion acknowledging the limits of causal inference. See page 2, lines 42-47; page 12, lines 271-275; and page 16, lines 388-392.

Scenario 3: Request you cannot fully satisfy

Comment:
The manuscript would be stronger with in vivo validation.

Response:
We appreciate this suggestion and agree that in vivo validation would strengthen the long-term program. However, this experiment is beyond the current study scope and cannot be completed within a proportionate revision timeline. To address the reviewer's core concern, we have added two orthogonal in vitro validation analyses and revised the Discussion to make the boundary of inference explicit. These changes appear on page 10, lines 221-239, and page 17, lines 401-410.

The point is not to win the argument. The point is to show mature judgment.

Nature editorial guidance explicitly supports discussion with the handling editor when reviewer requests are out of scope, too burdensome, or otherwise hard to resolve proportionately. Do not abuse that option, but use it when the revision path is genuinely ambiguous.

You should contact the editor when:

• reviewer requests conflict directly
• one reviewer is asking for a new project, not a revision
• a requested change would distort the paper's actual question
• you need guidance on whether narrowing the claim is acceptable

A short, calm note to the editor is far better than a defensive response letter that tries to solve an editorial judgment problem by brute force.

The most damaging response pattern is not disagreement. It is vagueness.

Editors lose confidence when authors say things like:

• "This has been addressed."
• "We revised the manuscript accordingly."
• "We believe the reviewer is mistaken."

Those responses create work for the editor and signal weak control of the revision.

Specificity builds trust. Vague reassurance destroys it.