Academic writing article coverAcademic writing article cover

Revised Manuscript Checklist: 9 Checks Before You Resubmit

A response that doesn't fully deal with the reviewers' comments can send your paper into another round of revise and resubmit, or worse, to rejection after revisions (Pallas, n.d.). So before you resubmit, run six checks on the response letter and the manuscript, then three on the files you upload.

The six: every change the letter claims is in the manuscript; every comment has a reply, including the editor's; each new reference leads to the paper you meant; in-text citations still match the reference list; page and line numbers in the letter fit the file you upload; and re-analyzed numbers are updated everywhere.

Is every change in your response letter actually in the manuscript?

Open the letter and the manuscript side by side and go to each location the letter gives. Revised text quoted in the letter should match the manuscript word for word. Paragraphs you edited again a day or two before submitting are often where the two drift apart: the manuscript has the new wording, and the letter still quotes the old.

Have you answered every reviewer and editor comment?

Paste each comment into the letter, number it, and reply under it; then look for any number without a reply. Christopher Pallas, writing for Taylor & Francis Author Services, asks authors to record the editor's or reviewers' words "without paraphrasing or shortening" (Pallas, n.d.). The editor's own requests often sit in the body of the decision letter, apart from the reviewer reports, so give them numbers too. If one comment holds several separate requests, split it so each gets its own answer.

You do not have to accept every suggestion. Pallas (n.d.) says you can respectfully disagree with a reviewer or the editor as long as you make a good argument why, and should note in the letter any revision you truly cannot make.

Did you verify every reference you added in this revision?

Count everything new, whether a reviewer suggested it, a supervisor or coauthor added it, or an AI assistant listed it. For each one, paste the DOI into doi.org and confirm it opens this exact paper, with the same journal, volume, issue and pages. Without a DOI, check the journal's website or a database record (how to find a DOI). Then reread the abstract to confirm the paper supports your point. A title search alone is not enough.

In one revision, a supervisor added four references. One had the right authors, title and year, and a title search found the paper. But the journal name, volume, pages and DOI were all wrong: the DOI had no record at doi.org, and the article had in fact been published in a different journal. Nothing about the entry looked wrong on the page.

If that entry goes back unchanged, a reviewer who opens the DOI gets an error page instead of the paper. And when Editorial Manager builds the PDF, a journal article that can't be matched to an online record is marked "Not validated" in a report made for the editor (what that label means).

AI-suggested references go wrong in the same way. In our test of ChatGPT references, every paper was real, but 4 to 9 of the 50 could not be used as written. The usual faults were a missing DOI, a missing author, a wrong article number, or a DOI that led to the preprint.

A title search would have passed the supervisor's entry, but a field-by-field comparison would not. In PerfectRefs, paste your reference list; you do not need to upload the manuscript. Style check compares authors, title, journal, year, volume/issue/pages and DOI with the database record, shows the record's version with a source link, and lets you accept a fix with one click. It uses premium credits (2 per reference), and new accounts get 10. Existence check looks each reference up in Crossref, OpenAlex and other scholarly databases and comes with a free daily allowance.

Do your in-text citations still match the reference list?

Author-date styles, such as APA 7

APA's rule runs in both directions: each work cited must appear in the reference list, and each work in the reference list must be cited in the text, or in a table, figure, footnote or appendix (American Psychological Association, 2022). Check both ways. Go through each parenthetical or narrative citation and find its entry, with the same author spelling and year. Then go down the list and confirm each entry is cited somewhere: a deleted sentence often leaves an orphan entry. For the citation forms, see APA 7 in-text citations.

Numbered styles, such as Vancouver or IEEE

The ICMJE recommendations for medical journals say references should be numbered consecutively in the order in which they are first mentioned in the text (International Committee of Medical Journal Editors, 2026). A reference added halfway through the paper shifts every number after it. In the text, each number should appear for the first time in order, 1, 2, 3, with none skipped; a later citation of the same source reuses its number. In the list, IEEE's guide says there must be only one reference with each number (IEEE, 2025). If a reference manager inserted your citations, use its refresh or update command once the text is final and before you make the clean copy.

Do the page and line numbers in your letter match the file you upload?

Fill in page and line numbers last, after every change is made, and say which version they refer to. A tracked-changes copy still shows deleted text, so its page and line numbers differ from the clean copy. Spot-check three to five of them in the exact file you will upload.

Did re-analyzed numbers change everywhere they appear?

List every value that changed and search the full text for each old value, such as the old sample size. Search for both forms of a value, with and without a leading zero (0.32 and .32). Check the abstract, table notes and figure captions separately.

When reviewers ask for changes to the analysis, Pallas (n.d.) advises providing new and updated tables and figures, and explaining in the response letter whether the new analysis led to a different interpretation of the data.

Three checks before you click submit

  • Your files are really in the system. Editorial Manager is used by Elsevier and other publishers. Elsevier's guide to submitting a revision says to untick any file from the previous version that you are replacing or removing, and that there should not be two versions of the same file. Building the PDF does not submit anything: the revision goes to the journal only when you click "Approve Revision" and then "OK" (Elsevier, n.d.).
  • The tracked and clean copies are the same document. Make the clean copy from your final tracked copy by accepting all changes, deleting all comments and saving it under a new name. Do not edit the two copies separately. Whether the journal wants both is in the decision letter, which Elsevier says should give any journal-specific instructions, such as a second copy of the revised manuscript with tracked changes (Elsevier, n.d.).
  • Double-anonymized journals carry no names in the files. Elsevier asks authors to remove identifying information, including author names, from file names, and to make sure document properties are anonymized as well (Elsevier, 2026). Tracked changes and comments also carry your word-processor user name. In Word for Windows, Inspect Document (File > Info > Check for Issues) can remove personal information; on the tracked copy, leave "Comments, Revisions and Versions" alone so the changes survive, then check that your name is gone.

FAQ

What should I check before resubmitting a revised manuscript?

Six things in the letter and manuscript: claimed changes are made, every comment is answered, new references lead to the right papers, citations match the list, page and line numbers fit the uploaded file, and changed numbers are updated. Then three in the files: the right uploads, matching tracked and clean copies, and approval in the submission system.

Do I have to make every change the reviewers ask for?

No. You can respectfully disagree when you can make a good argument why, and revisions you truly cannot make should be noted in the response letter (Pallas, n.d.).

How do I check that a new reference is real and correct?

Paste the DOI into doi.org and confirm it opens the same paper, journal, volume and pages. A title search can find a real paper even when the other details are wrong, so compare every field. PerfectRefs Style check does that comparison against database records for a whole list.

Why does Editorial Manager still show my revision as not submitted?

Building the PDF is not the last step. Click "Approve Revision" and confirm; the revision then appears under "Revision Being Processed" (Elsevier, n.d.).

Do I need to upload both a tracked-changes version and a clean version?

It depends on the journal. Check the decision letter: Elsevier gives a second tracked-changes copy as one example of a journal-specific instruction (Elsevier, n.d.).

References

  1. American Psychological Association. (2022, July). Basic principles of citation. APA Style. https://apastyle.apa.org/style-grammar-guidelines/citations/basic-principles
  2. Elsevier. (n.d.). Submitting a revision in Editorial Manager [Video transcript]. https://supportcontent.elsevier.com/Support%20Hub/Journals/Editorial%20Manager%20Videos%20Transcript/Submitting%20a%20revision%20in%20Editorial%20Manager.pdf
  3. Elsevier. (2026, August 28). What are the requirements for double-anonymized peer review? https://www.elsevier.support/publishing/answer/what-are-the-requirements-for-doubleanonymized-peer-review
  4. IEEE. (2025). IEEE reference style guide for authors (Version 3.28.2025). https://journals.ieeeauthorcenter.ieee.org/wp-content/uploads/sites/7/IEEE_Reference_Guide.pdf
  5. International Committee of Medical Journal Editors. (2026). Recommendations for the conduct, reporting, editing, and publication of scholarly work in medical journals. https://www.icmje.org/icmje-recommendations.pdf
  6. Pallas, C. L. (n.d.). How to respond to reviewer comments. Taylor & Francis Author Services. https://authorservices.taylorandfrancis.com/blog/peer-review/how-to-respond-to-reviewer-comments/

Original article by PerfectRefs. To republish it, credit PerfectRefs and link to this page. Do not edit it or use it commercially.