How to write a point-by-point response to journal reviewers

You have revised the manuscript, but the response letter still says “Done” or “Corrected” beneath several reviewer comments. You know what changed. The editor and reviewers may have to search the whole paper to find out.

A useful point-by-point response makes that work easier. It identifies the concern, answers it directly, explains the action taken, and shows where the relevant change appears. Where you disagree, it gives the scientific reason rather than leaving the comment unanswered.

The response is part of the revision itself. It should accurately describe the paper you are resubmitting, including limitations that remain.

Read the decision letter before answering individual comments

Start with the editor’s decision and instructions. Identify the revision deadline, required files, and any priorities or conditions. A journal may require a clean manuscript, a marked version, a separate response document, or a particular method of showing changes.

Then read all reviewer comments before deciding what to revise. The same concern may appear in different wording, and a change requested by one reviewer may affect another reviewer’s point.

Keep the full comments in your working record. If a paragraph contains several requests, give each one an identifiable response without changing the reviewer’s meaning. William Stafford Noble’s guidance on response documents recommends making the full review and the author’s replies easy to distinguish and following each issue through explicitly.

Build a response record before polishing the letter

A simple working table can stop small requests from disappearing during a major revision. Give each point an identifier tied to the editor or reviewer, then record the decision, responsible author, manuscript location, and completion status.

Distinguish “needs analysis,” “needs author decision,” “text revised,” and “checked against manuscript.” A polished response is not evidence that an analysis has actually been completed.

Address consequential methodological and interpretive issues early. If they change the results or conclusions, later language edits may need to change too. Small corrections still deserve a response, but they should not distract from an unresolved challenge to the study’s reasoning.

Use the table as a working aid. The final response may be clearer as numbered comments and replies, depending on the journal’s instructions and the length of the material.

Make each reply answerable and traceable

For most comments, a useful reply contains:

A direct answer to the concern.
The action taken, or the reason for a different action.
The location in the revised manuscript.
A short excerpt of revised text where it helps the reviewer verify the change.

These are functions, not a requirement to write four sentences every time. A spelling correction can have a short response. A concern about an analytical assumption needs enough detail to assess the reasoning.

Avoid saying “We have clarified this” when the response does not explain what became clearer. Also distinguish material that was already present from text newly added. If an important detail was hard to find, moving or rephrasing it may still be a useful revision.

Worked example: three different responses to three different problems

All comments, manuscript details, and replies below are synthetic. They are invented examples, not client correspondence or a report of a real submission. The named sections are illustrative locations in the fictional manuscript.

Suppose a descriptive study uses employee questionnaires and organisational records. The reviewers raise the following points:

On a narrow screen, scroll the table sideways to read every column.

Reviewer comment	Author decision	Evidence of the response
“State the period covered by the administrative records.”	Accept and add the missing detail.	Data sources subsection now identifies the observation period.
“The term ‘performance’ is unclear. Does it refer to employee ratings or recorded completion times?”	Clarify the two measures and correct ambiguous labels.	Methods definitions, results-table labels, and discussion wording are checked together.
“Please conclude that training caused the observed improvement.”	Do not adopt the causal claim; explain the design limit.	Response gives the reason; discussion retains appropriately bounded wording.

Example reply · add a missing detail

The first reply could read:

We have added the observation period to the Data sources subsection: “The administrative records cover requests received during the first quarter of the study year.” This identifies the period used for record selection.

In a real response, use the actual dates where needed and give the exact page and line references required by the journal. Do not replace precise information with the generic wording of this illustration.

Example reply · clarify a measure

The second reply needs more than a word substitution:

The manuscript reports two different measures: employees’ self-rated task performance and recorded request-completion time. We agree that using “performance” for both obscured that distinction. We have clarified the definitions in Measures, revised the relevant table labels, and checked the corresponding discussion passages. The revised text refers to each measure by its specific name.

Example reply · explain a design limit

For the third comment, a reasoned response could say:

The observational comparison does not establish a causal effect of training. We have therefore retained association-based wording and clarified this limitation in the Discussion. The revised passage states that the observed pattern cannot determine whether training caused the difference.

The reply declines the requested conclusion while addressing the concern transparently. It does not argue that a causal effect is impossible; it explains that this study does not establish one.

Every statement about a change must match the actual revised manuscript. These examples illustrate response structure, not wording to paste into a letter without doing the work.

Disagree with the request, not with the person

You do not have to accept every suggested interpretation or addition. You do need to understand the underlying concern and explain your decision.

Taylor & Francis’s reviewer-response guidance recognises that authors may disagree or be unable to make a requested change, provided they explain why. A substantive answer may cite the design, available data, relevant literature, or the scope of the research question.

“This is beyond the scope” is usually more useful when followed by what the requested work would involve and why it is not supported by the present study. If a limitation remains, acknowledge it rather than suggesting that an explanation in the response letter makes it disappear.

Where reviewers make conflicting requests, identify the conflict and explain your proposed resolution. Seek the editor’s guidance when the choice would materially change the paper or the requirements are genuinely unclear. Do not silently satisfy one request and omit the other.

Handle requests for additional analysis as scientific decisions

A reviewer may request a sensitivity analysis, a different model, or further data. Before promising a result, assess whether the request is appropriate for the question and evidence available. Involve the authors responsible for the method and analysis.

If an additional analysis is conducted, report what it shows even when it does not support the preferred interpretation. Explain whether it changes the manuscript’s conclusions. Keep the methods, results, tables, supplementary material, and response consistent with one another.

If the request cannot be met because the necessary data were not collected, say so. Do not invent a variable, imply that an analysis was performed, or describe a limitation as resolved when it remains. A narrower conclusion may be necessary. The editor may still decide the available evidence is insufficient; a clear response cannot guarantee a favourable decision.

New analyses also require an accurate account of when and why they were undertaken. Do not present a reviewer-requested exploration as though it had been specified before seeing the data.

Keep the response separate from the manuscript’s explanation

The response letter is read alongside the review history. The eventual article usually is not. If a reply contains information necessary to understand or assess the study, consider whether that information also belongs in the manuscript or its permitted supplementary material.

For example, a detailed explanation of an exclusion rule should not exist only in a private reply to a reviewer if readers need it to understand the reported sample. Conversely, the manuscript does not need to reproduce the conversational history of how the reviewer raised the point.

Use labels or clear formatting to separate the reviewer’s comment, the author’s response, and quoted revised text. Do not depend on colour alone. A reader should still be able to navigate a printed or plain-text version.

Check locations after the revisions are settled

Page and line numbers can change during editing. Update them against the specific revised file you will submit and explain whether they refer to the clean or marked version.

Check each response by going directly to its stated location. Confirm that the change exists and that any quoted passage matches it. Then work in the other direction: inspect substantive manuscript changes and make sure the response does not overlook their implications for earlier replies.

Before resubmission, confirm that:

every editor and reviewer point has an identifiable answer;
the replies describe completed work accurately;
disagreements give reasons and acknowledge remaining limits;
added analyses and changed conclusions are reported consistently;
response excerpts and locations match the submitted files;
required clean, marked, supplementary, and anonymised files follow the journal’s instructions;
the authors have reviewed the scientific content of the revision and response.

Authorship includes responsibility for the work, as reflected in the ICMJE authorship recommendations. Follow the policy of your actual journal, including any requirements for acknowledging editorial assistance.

Our academic editing can help make your response and revised manuscript clearer and more consistent. Authors remain responsible for the scientific reasoning, analyses, substantive replies, and final submission.