# Clean Copy vs. Tracked Changes: What to Submit for Resubmission A clean copy is your revised manuscript with every edit accepted and all markup hidden, while a version with tracked changes displays every insertion, deletion, and comment for the editor to inspect. The rule of thumb is simple: if the journal will accept two files, send both, with the tracked-changes file as the marked-up version and the clean copy as the reading and production version. If the journal allows only one file, follow the decision letter exactly, and when it is silent, send the clean copy and attach a short note offering the marked-up version on request. Most confusion at revision time comes from treating this as an either-or choice. It usually is not. The real question the editor is asking is whether they can verify what changed without re-reading the entire paper. This article explains what each version is, when to use clean copy, when tracked changes are required, how to build both files without errors, and how they fit together with your reviewer response letter. ## The Clean Copy vs. Tracked Changes Dilemma, Stated Precisely The 'clean copy' vs. 'tracked changes' dilemma appears at almost every revision decision point in peer review. The editor returns a decision letter. It asks you to revise. Somewhere near the end, one of three sentences appears: 1. "Please upload a marked-up copy showing all changes." 2. "Please upload a clean copy of your revised manuscript." 3. "Please upload both a clean copy and a version with tracked changes." The first sentence means the editor or reviewers want to audit your revisions. The second means the journal already knows what to look for, or the reviewers will re-read the whole paper anyway. The third is the most common modern practice at journals that run full re-review, because it serves two audiences at once. The difference between clean copy and tracked changes for resubmission is not cosmetic. Each file answers a different question. The tracked-changes file answers "what did the authors do in response to our critique?" The clean copy answers "what does the paper now say?" A resubmission package that answers both questions, in the format the journal asked for, moves through editorial screening fastest. ## Side-by-Side Comparison The table below is the core of the decision. Keep it open while you prepare your files. | Aspect | Clean copy | Tracked changes | |--|--|--| | What it is | The revised manuscript with all edits accepted and all markup hidden or removed | The revised manuscript with insertions, deletions, and comments visible as markup | | What it shows | The final text as it would be published | A line-by-line record of what changed since the previous version | | Who reads it first | Production staff, copyeditors, and reviewers reading for content | Editors and reviewers verifying that requested revisions were made | | Typical file name | Revised manuscript (clean) | Revised manuscript (marked-up or tracked changes) | | Risk if submitted when not asked | Usually harmless, but wastes reviewer time at journals expecting markup | A serious error if clean copy was requested, because it obscures the revisions | | Risk if omitted when asked | Delayed processing or a second upload request | Delayed processing, and reviewers may assume revisions were not made | | Best paired with | The reviewer response letter and the tracked-changes file | The reviewer response letter and the clean copy | | LaTeX equivalents | A fresh compile of the revised source, optionally with the latexdiff-generated marked version | A latexdiff comparison PDF compiled from the original and revised sources | | Machine readability | Fully clean, no leftover revision metadata | Contains visible markup and sometimes hidden comments, which must be reviewed before production | ## What Each Version Is, on Its Own Terms ### What a clean copy is A clean copy is the manuscript as you want it to be read. Every revision has been made, every change has been accepted, and the document displays no revision marks, no comment balloons, and no highlighted text. If someone opened the file and printed it, the printed pages would look like the final article. Clean copy has two distinct jobs. First, it is the reading version for anyone who wants to evaluate the science without the visual noise of markup. Second, it is the production source. Once your paper is accepted, the journal's copyediting and typesetting workflow starts from a clean file. If a marked-up file reaches production by mistake, leftover revision metadata can surface during typesetting as unwanted strikethrough text or orphaned comment balloons. This is when to use clean copy in its purest sense: as the single authoritative text of the revised paper. Everything else, the marked-up file and the response letter, is commentary on that text. ### What tracked changes are Tracked changes is a word-processing feature, available in Microsoft Word and comparable tools, that records every edit as visible markup. Inserted text appears underlined and usually in a color. Deleted text appears with strikethrough rather than disappearing. Margin comments attach notes to specific passages. Each edit carries the name of the person who made it and a timestamp. The value of tracked changes at resubmission is auditability. A reviewer who asked for a power analysis to be added can search the marked-up file, find the new statistics paragraph, and confirm the addition without cross-referencing the response letter against a silent text. When revisions are extensive, this verification step is faster with markup than without it, which is why editors at journals that send papers back to the same reviewers often require it. Tracked changes also carry a privacy obligation. The markup stores author names and edit history. Before uploading, confirm that all edits are attributed to you and your coauthors rather than to anyone whose name should not appear, and confirm that no internal comments from your drafting process remain in the file. ## What Journals Actually Ask For at Resubmission ![Comparison of four decision letter patterns showing what each asks for and what to upload in response.](/article-images/illustrations/clean-copy-vs-tracked-changes-for-resubmission-letter-patterns.webp) Decision letters fall into a small number of recognizable patterns. Learning the patterns tells you what to prepare before you touch the manuscript file. **Pattern one: the letter explicitly requests both versions.** Prepare two files, one clean and one marked-up, and upload them under the correct designations in the submission system. Many systems have separate file-type dropdowns labeled clean and marked-up. Choosing the wrong dropdown is a common cause of administrative hold-ups. **Pattern two: the letter requests only the marked-up version.** Some editors rely on markup plus the response letter, and never need a separate clean file until acceptance. Submit exactly what was asked. The production team will request a clean, editable file later, and producing it then takes minutes if your tracked changes were managed properly throughout. **Pattern three: the letter requests only a clean copy.** Submit the clean file. Do not attach a marked-up version unless the submission instructions elsewhere in the journal's author guidelines call for one. An unrequested markup file adds clutter and can confuse a reviewer who wanted to read the paper straight through. **Pattern four: the letter is silent on the question.** Check the journal's instructions for authors under revision or resubmission. If those are also silent, the safe default is a clean copy uploaded as the manuscript, with a brief note in the response letter stating that a version with tracked changes is available on request. Some authors instead upload the marked-up version as a supplementary file. That practice is generally accepted but adds a file the editor may not want. The consistent theme across all four patterns is that the decision letter and the journal's author instructions outrank any general rule. When the two conflict, ask the editorial office. A one-line email resolves the ambiguity, and editors answer these questions routinely. ## How to Decide: The Criteria That Matter ![Decision path from the decision letter wording to the file to upload, ending in sending both files or the clean copy alone.](/article-images/illustrations/clean-copy-vs-tracked-changes-for-resubmission-which-file-to-send.webp) If you are weighing clean copy or tracked changes for resubmission without an explicit instruction, walk through five criteria in order. **Criterion one: will the same reviewers see the paper again?** If the editor plans to send the revision back to the original reviewers for re-review, tracked changes are high value. Those reviewers remember their critiques and want to verify each one was addressed. Markup lets them verify quickly, which raises the odds of a fast, favorable second pass. **Criterion two: how extensive were the revisions?** A revision that changes three sentences reads fine as clean copy. A revision that reorders the results section, adds a new figure, and rewrites half the discussion benefits from markup, because the extent of change is itself informative. Reviewers calibrate their scrutiny to the size of the edit. **Criterion three: what did the reviewers actually ask for?** When reviewers request specific textual changes, tracked changes prove the changes were made exactly and in the right place. When reviewers request new experiments or analyses, the burden of proof lives in the reviewer response letter and the new data, not in the markup. In that situation the clean copy matters more, because reviewers are evaluating new content that did not exist in the original. **Criterion four: does the journal need a production-ready file soon?** If the paper is a minor revision on a fast track and acceptance is near, producing the clean copy carefully from the start saves a later step. **Criterion five: what format is the manuscript in?** Word documents handle tracked changes natively. LaTeX does not have an equivalent feature built in, so authors in LaTeX typically generate a marked-up comparison with the latexdiff utility, which takes two source files and produces a third that highlights the differences. If you work in LaTeX and cannot generate a usable diff, a clean recompiled PDF plus a detailed response letter with line references is the practical substitute, and many journals accept exactly that. Scoring these five criteria takes about two minutes and resolves nearly every real case. In practice, criterion one dominates. Re-review means markup. No re-review means clean copy usually suffices. ## Worked Scenarios ### Scenario one: major revision with re-review A paper receives a major revision decision with four reviewer reports. Two reviewers want new analyses, one wants the introduction reframed, and one wants the limitations expanded. The letter asks for a point-by-point response and "a marked-up copy of the manuscript." The right package contains three files. The tracked-changes file, with every edit recorded and attributed. The clean copy, for the editor to read the final state. The reviewer response letter, quoting each reviewer comment in full and answering it beneath, with page and line numbers pointing into the revised manuscript. All three files must agree with each other. If the response letter says the new analysis appears on page 12, the clean copy and the tracked-changes file must both show it on page 12. ### Scenario two: minor revision, no re-review A paper receives a minor revision decision. The editor states the reviewers will not see the manuscript again and asks for a clean revised copy with changes highlighted. The phrase "with changes highlighted" means the journal wants the edits visible, which is a tracked-changes request in substance. In this situation the marked-up file is the required deliverable, and the clean copy is the companion. This pattern trips up many authors, because the word "clean" in the letter does not mean no markup. ### Scenario three: the journal accepts one file only A submission portal allows a single manuscript file. The instructions request both a clean and a marked-up version. The practical solutions, in order of preference, are to upload the tracked-changes document as the main manuscript and note in the response letter that a clean copy is available, or to upload the clean manuscript and embed the marked-up version as a second PDF within the supplementary files. Whichever route you choose, state it explicitly in the response letter so the editor knows both versions exist and where to find them. ### Scenario four: a coauthor objects to showing the edits A coauthor worries that tracked changes will expose drafting shortcuts or make the team look careless. The rebuttal is practical, not rhetorical. Markup is not a judgment on the writing. It is a service to the reviewers that shortens their verification work, and editors interpret extensive markup as evidence of thorough revision, not as evidence of a weak original. If a coauthor remains uncomfortable, the compromise is to anonymize the markup by turning off name attribution in the tracking settings before saving, so the file shows what changed without showing who changed it. ## How to Prepare Both Files Correctly ### Building the tracked-changes file Turn on track changes before you make the first edit to the revised manuscript. Making revisions with tracking active from the first keystroke guarantees a complete record. If you revise across multiple sessions, keep tracking on in every session, and check at the start of each session that the feature is still enabled, because it resets more often than authors expect. Manage the markup as you go. Accept or reject your own draft-level experiments rather than leaving junk markup in the file. Every change that appears in the final tracked-changes file should correspond to a real revision you can defend in the response letter. Before saving the final version, run through the document one last time. Check that all edits are attributed correctly, that no comments from internal discussions remain, and that the markup does not extend into the reference list formatting in ways that will look broken to a copyeditor. ### Building the clean copy from the marked-up file The only correct way to produce a clean copy is to accept all changes in the tracked-changes document. Do not retype the manuscript from scratch, and do not manually delete markup characters by hand. Accepting all changes is a single command in Word, found under the Review tab, and it leaves a document that is textually identical to the marked-up file. After accepting all changes, take two verification steps. First, confirm no markup remains by checking that the display shows the document without strikethrough text or margin balloons. Second, confirm that hidden comments were deleted, because accepting changes does not remove comments. Then save under a new file name that clearly distinguishes the clean version from the marked-up version. A subtle failure mode deserves its own warning. If you accepted changes in some places and rejected them in others while drafting, the two files can drift apart without your noticing. After producing the clean copy, spot-check three or four passages against the tracked-changes file. The clean copy must contain every accepted edit and none of the rejected ones. ### Keeping the two files synchronized File synchronization problems arise when authors make a late edit in one file and forget the other. The fix is procedural, not technical. Treat the tracked-changes file as the master document throughout revision. Make every edit there, with tracking on. Generate the clean copy only at the end, as a derived file. If a late change arrives from a coauthor after the clean copy exists, make the edit in the master file first, then regenerate the clean copy by accepting all changes again. Never edit the clean copy directly, because those edits will not appear in the markup and the two files will silently disagree. Line numbers deserve a final check here. If your manuscript displays continuous line numbers, and reviewers will cite them, confirm that the clean copy and the marked-up file use the same line-numbering settings. Markup itself does not change line numbers, but layout shifts caused by inline markup can make the marked-up file paginate slightly differently from the clean copy. If a reviewer quotes a line number from the markup, that reference must land on the same sentence in the clean copy. When pagination drifts, the fix is to regenerate the marked-up PDF from the final accepted document rather than patching individual lines. ## The Reviewer Response Letter and How the Files Fit Together The reviewer response letter, sometimes called the point-by-point response or rebuttal letter, is the document that organizes the entire resubmission. Learning how to respond to peer reviewers well is largely a matter of structure, and the structure is settled practice. Open with a short summary thanking the editor and reviewers and stating in one or two sentences the major changes made. Then list every reviewer comment verbatim, usually in italics or as a block quote, numbered by reviewer. Beneath each comment, write your response. Where a change was made, describe it and cite the exact location in the revised manuscript by page, line, section, or paragraph. Where new text was added, quote it in the letter so the reviewer sees the wording without hunting. Where you disagree, respond respectfully and give the reasoning and any supporting evidence. Disagreement with a reviewer is acceptable and sometimes necessary, but it must be argued, not asserted. The response letter is also where you declare the file situation. One sentence near the top does the job. For example: "Attached are a clean copy of the revised manuscript and a version with tracked changes highlighting all edits relative to the original submission." Three habits distinguish strong response letters. First, complete coverage. Every single comment gets a response, including the ones you declined to act on. Second, traceable references. A reviewer should be able to read the letter with both manuscript files open and reach the relevant spot in seconds. Third, matching tone. The letter addresses reviewers by their numbered comments, never by name unless the journal uses open review, and never disputes a critique in a register sharper than the critique itself. The response letter and the two manuscript files form one package. The letter tells reviewers what changed and why. The tracked-changes file shows where. The clean copy shows the result. Any inconsistency among the three, even a small one such as a response letter citing a paragraph that the clean copy places elsewhere, costs credibility and review time. ## Common Mistakes and Limitations These are the errors that most often slow down a resubmission, listed with their fixes. **Uploading the wrong file to the wrong slot.** Submission systems with separate clean and marked-up designations receive swapped files constantly. Check the file names and open each uploaded file before final submission. **Editing the clean copy by hand.** Deleting strikethrough text manually instead of accepting changes leaves fragments, duplicate sentences, and formatting residue. Always accept all changes through the review feature, never by hand. **Leaving internal comments in the file.** Comments to coauthors such as "check this citation" have appeared in published supplementary materials. Use the document inspector or the comment-deletion command to strip all comments before uploading. **Letting edits go untracked.** A revision made with tracking off leaves no record, so the marked-up file understates the actual changes and reviewers may conclude that requested edits were ignored. Keep tracking on from the first revision session to the last. **Producing a marked-up PDF that flattened the markup.** Exporting a Word file to PDF can flatten tracked changes into final text if the print settings exclude markup. Before uploading a PDF, open it and confirm the insertions and deletions are visibly present. **Sending markup that attributes edits to the wrong person.** Files edited by multiple coauthors carry per-person attribution. Confirm that the attribution matches the author list, or turn off attribution if the journal prefers anonymous markup. **Ignoring the decision letter's exact wording.** A letter that says "changes highlighted" wants markup. A letter that says "clean revised manuscript" wants no markup. Match the letter, and ask the editorial office if the wording is genuinely ambiguous. **Assembling the response letter late.** Authors who write the response letter after finalizing the manuscript files discover inconsistencies too late to fix them cleanly. Draft the response letter alongside the revisions, updating the page and line references as the text moves. On limitations. The practices described here reflect the standard conventions of journal peer review and the documented behavior of common word-processing tools, but journals differ in their house requirements, and some editorial offices maintain specific file-format rules that override general custom. Individual manuscripts and individual journals can call for arrangements this article does not anticipate, so the decision letter and the journal's author instructions are always the controlling documents for a specific submission. When the instructions are unclear, the editorial office staff are the correct authority to consult, and asking them a direct question carries no penalty. ## Frequently Asked Questions ### Should I submit clean copy or tracked changes for resubmission? Submit both if the journal accepts two files, with the tracked-changes file labeled as the marked-up version and the clean copy labeled as the revised manuscript. If the journal allows one file, follow the decision letter, and if it is silent, send the clean copy and note that a marked-up version is available. ### What is the difference between clean copy and tracked changes? A clean copy shows the revised manuscript with no visible markup, reading like the final article. A tracked-changes file displays every insertion, deletion, and comment so editors and reviewers can see exactly what changed between the original submission and the revision. ### How do I create a clean copy from a file with tracked changes? Open the tracked-changes document, use the accept-all-changes command under the Review tab, delete any remaining comments, and save the result under a new file name. Never build a clean copy by manually deleting markup text, because that leaves formatting residue and errors. ### Can I show tracked changes in a PDF instead of a Word file? Yes, many journals accept a PDF with visible markup, and LaTeX authors commonly provide a latexdiff comparison PDF. Verify after export that the markup survived the PDF conversion, because some export settings flatten tracked changes into final text. ### What should I do if the journal asks for only one file but I have both? Upload the version the letter or portal requests as the primary manuscript file, add the other version as a supplementary file where the system permits, and state clearly in the reviewer response letter that both versions exist and where each one is located. ### Do tracked changes replace the reviewer response letter? No. The response letter explains why each change was made and where to find it, quoting each reviewer comment and answering it. Tracked changes only show what changed, so the two documents serve complementary purposes and journals that request one almost always expect the other.