How to remove all annotations from a pdf before submission
- Step 1Action every comment first — Incorporate all supervisor and reviewer feedback into your manuscript before cleaning. Removal is irreversible on the output and all-or-nothing — once the annotations are gone you have lost the notes, so make sure you've captured them.
- Step 2Re-export a fresh PDF from your source — Ideally export a clean PDF from LaTeX or Word after applying changes, so annotations were never added to the final. Use this tool when you only have an annotated PDF and no clean re-export.
- Step 3Keep the annotated copy — Archive the supervisor-annotated PDF separately. You may need to refer back to a specific comment during the viva or revisions.
- Step 4Open the tool and drop the PDF — Go to the PDF Annotation Remover and drag your manuscript in. It accepts one PDF and runs automatically on drop — there is no settings step.
- Step 5Verify the cleaned manuscript — Reopen the output and page through it: no margin comments, highlights, or sticky notes should remain. Check that figures, equations, and page numbering are intact.
- Step 6Scrub metadata, then submit — For blind review, follow with the Metadata Scrubber to clear your name and producer fields, then upload the file (
thesis.annotation-remover.pdf) to the portal.
Submission-cleaning checklist
What this tool handles and what you still need a sibling tool for when preparing an academic submission.
| Concern | Handled here? | Tool to use |
|---|---|---|
| Supervisor margin comments / sticky notes | Yes | This tool (deletes /Annots) |
| Highlights and strikethroughs | Yes | This tool (same pass) |
| Author name / institution in metadata (blind review) | No | Metadata Scrubber |
| Personal data in the body text (e.g. participant names) | No | PII Redactor |
| File over the portal's size limit | No | Lossy / Lossless compress |
| Locking a fillable cover sheet's values | No (fields are cleared) | PDF Flatten first |
Free vs Pro limits
Enforced before processing; both size and page count are checked on free. Theses routinely exceed both.
| Tier | Max file size | Max pages | Files per job |
|---|---|---|---|
| Free | 2 MB | 50 pages | 1 |
| Pro | 50 MB | 500 pages | 5 |
| Pro + Media | 500 MB | 2,000 pages | 50 |
Cookbook
Real submission scenarios for a thesis or journal paper and exactly what the single-pass cleaner does.
A thesis chapter returned with margin comments
Your supervisor's comments are FreeText and Text annotations in the margins. After you've applied the changes, the comments must not appear in the submitted file.
Before: chapter3-reviewed.pdf (38 pages) ~60 Text/FreeText comments across margins Highlights on 12 pages After drop (auto-run): /Annots deleted on every page → chapter3-reviewed.annotation-remover.pdf body text, figures, equations unchanged
Preparing for blind peer review
A journal requires an anonymised manuscript. Removing annotations is step one; scrubbing your identity from the metadata is step two. Annotation removal alone does not anonymise the file.
Step 1 annotation remover → comments/highlights gone
Step 2 metadata scrubber → Author, Title, Producer cleared
Result: no reviewer-visible comments AND
no author name in document propertiesA thesis over the free 2 MB / 50-page limit
A full thesis with figures easily exceeds both free-tier limits. Compress first or upgrade — the tool blocks before processing if either limit is exceeded.
thesis-full.pdf → 4.7 MB, 180 pages Free tier: blocked (over 2 MB AND over 50 pages) Option A: pdf-compress-lossy → under 2 MB, then re-check pages Option B: upgrade to Pro (50 MB / 500 pages)
Strikethrough suggestion that must be a real deletion
Your supervisor struck through a paragraph. Removing the StrikeOut annotation only removes the line — the paragraph stays. You must delete it in your source and re-export before cleaning.
Before: paragraph with StrikeOut over it
Wrong: run annotation remover only
→ line gone, paragraph STILL THERE
Right: delete paragraph in LaTeX/Word, re-export,
then annotation remover for any leftoversConfirming the file the portal will accept
The result panel reports counts only. Page through the output to confirm it is examiner-clean before uploading.
Result panel: Input pages: 142 Input size: 9.2 MB (Pro) Output size: 9.0 MB Processing: Browser Open → no comments, no highlights, ToC intact Upload thesis.annotation-remover.pdf to portal
Edge cases and what actually happens
Annotations removed before feedback was applied
IrreversibleThe tool deletes all annotations in one pass with no undo on the output. If you clean the file before incorporating the supervisor's notes, those notes are lost from that file. Always action feedback first and keep the annotated original archived.
Free-tier thesis over 2 MB or 50 pages
BlockedBoth limits are enforced before processing, and full theses with figures routinely exceed them. The run is blocked with a message naming the limit. Compress with lossy compression, or upgrade to Pro (50 MB / 500 pages) or Pro + Media (500 MB / 2,000 pages).
Removing comments does not anonymise the file
Incomplete for blind reviewYour name can still sit in the document metadata (Author, Title, Producer) after annotations are gone. For genuine blind review, run the Metadata Scrubber as a second step. Annotation removal handles the visible comments, not the hidden identity fields.
Strikethrough text remains in the document
By designA StrikeOut annotation marks text for deletion but does not delete it. Removing the annotation removes the line through the words; the words stay. Make the actual deletion in your manuscript source and re-export before submitting.
Interactive cover-sheet fields are cleared
ExpectedSome submission templates include a fillable cover sheet (candidate name, ID). Because form fields are annotations, the tool clears them. If you need the typed values to remain, run PDF Flatten first to bake them into the page, then remove annotations.
Encrypted manuscript from the supervisor
Often worksThe tool loads documents with encryption ignored and processes many protected files fine. If a strongly encrypted PDF fails to open, remove the password first with PDF Unlock, then clean it.
Comments embedded as real page text
PreservedIf a note was typed into the body rather than added as an annotation (e.g. an inline [TODO: rewrite] left in the text), it is part of the content stream and survives. Proofread the cleaned file; the tool only removes the /Annots layer.
Already-clean re-export from LaTeX/Word
By designIf you exported a fresh PDF after applying changes, it may have no annotations at all. Running the tool on it is harmless — pages with no /Annots array are skipped and you get a valid, re-saved file.
Participant names in the body (ethics)
Wrong toolRemoving a comment that mentions a participant does not redact participant data in the manuscript body. For research ethics, detect and black out personal data with the PII Redactor before submission.
Corrupt PDF
ErrorA truncated or invalid PDF fails to load and the run errors. Repair it with PDF Repair first, then clean and submit.
Frequently asked questions
Should I remove annotations before or after applying my supervisor's feedback?
Always apply the feedback first. The tool deletes all annotations in one pass with no undo on the output, so if you clean before incorporating the notes, you lose them from that file. Action every comment in your manuscript, archive the annotated copy, then clean the version you submit.
Will the submission portal detect that annotations were removed?
No. Removed annotations leave no trace — the page /Annots array is deleted and the file re-saved as a normal PDF. There is no 'annotations were here' marker. The portal sees a clean document, exactly as if it had never been annotated.
Does this anonymise my thesis for blind review?
Partially. It removes the visible comments and highlights, but your name can still be in the document metadata (Author, Title, Producer). For true anonymisation, run the Metadata Scrubber afterward to clear those fields. Annotation removal handles what an examiner sees on the page; metadata scrubbing handles what they see in document properties.
Can I keep some comments — for example, ones I still need to address?
No. The tool is all-or-nothing: it deletes the entire annotation layer on every page in one pass. There is no way to keep selected comments. Keep the annotated original as a working copy and only run the cleaner on the version you intend to submit.
My supervisor struck through a paragraph — will removal delete it?
No. A strikethrough is a StrikeOut annotation drawn over the text; it doesn't remove the text. Cleaning removes the line but leaves the paragraph. If the strikethrough meant 'delete this', make the deletion in your LaTeX or Word source and re-export the PDF before cleaning.
Is my unpublished manuscript uploaded anywhere?
No. The tool runs entirely in your browser using pdf-lib. Your thesis or paper never leaves your device — the result panel states Local browser processing · 0 bytes uploaded. This matters for unpublished work; only an anonymous usage counter is recorded when you are signed in.
Does it work on PDFs annotated in Acrobat, Preview, or Mendeley?
Yes. Comments and highlights from any standard PDF tool — Acrobat, macOS Preview, Microsoft Edge, Mendeley, Foxit — are stored as conformant annotations in the page /Annots array. The tool deletes that array regardless of which app added the annotations.
My thesis is too big for the free tier — what now?
Free tier allows 2 MB and 50 pages, both enforced before processing, and a full thesis usually exceeds both. Compress with lossy compression to get under 2 MB, or upgrade to Pro (50 MB / 500 pages) or Pro + Media (500 MB / 2,000 pages) to clean the whole document in one go.
Will my figures, equations, and page numbers survive?
Yes. The tool only deletes the annotation overlay; it never touches the page content stream where figures, typeset equations, tables, and page numbers live. The cleaned PDF is visually identical to the input minus the comments and highlights. Always page through it once to confirm.
Do I have to choose anything before it runs?
No. This is a one-click tool with no options panel. The moment you drop the manuscript in, it processes automatically and shows you the result — useful when you're cleaning a file under a submission deadline and don't want to fiddle with settings.
What about participant or interviewee names in the body text?
Those are page content, not annotations, so this tool does not touch them. Removing a comment that mentions a name does not redact the name where it appears in the manuscript body. For ethics compliance, use the PII Redactor to detect and black out personal data before submission.
Can I trust a browser tool for something this important?
The operation is deliberately simple and deterministic: it deletes the /Annots array per page using pdf-lib, the same well-tested library many PDF tools rely on, and re-saves the file. There is no AI guessing and no network round trip. Verify the result by reopening the output and confirming the comment panel is empty before you submit.
Privacy first
All PDF processing runs locally in your browser using PDF-lib and pdf.js. No file is ever uploaded — only metadata counters are saved for signed-in dashboard stats.