Redact a PDF, and actually remove the text
Hide account numbers, ID numbers, names or anything else in a PDF before you share it. PDF Desk deletes the text under each box from the saved file, so it cannot be copied, searched or recovered by removing the box. It all happens in your browser; the document is never uploaded.
Free. No account. Files are opened inside your browser and never uploaded.
Why a black box is not enough
Many tools, and most "draw a rectangle" workarounds, only paint over the text. The words are still in the file underneath: anyone can select them, copy them, or find them with search. When PDF Desk saves, it rewrites the page and removes every character whose centre falls inside a box, along with any links or comments there. Everything else on the page stays exactly where it was.
Two ways to redact
| Tool | Use it for |
|---|---|
| White-out | Drag a box over anything: a line, a signature, a photo, part of a scan. White by default, and you can pick any colour, including black. |
| Redact (find every occurrence) | Type a word or number and cover it on every page at once, with match case and whole word options. Or pick a ready-made pattern: PAN, Aadhaar, card number (masked ones too), account number (any run of 9 to 18 digits), IFSC code, GSTIN, email address or phone number. |
How to redact a PDF
- Click Open a PDF to redact above, then the big button on the page that opens, and choose the file. The Redact panel opens by itself; close it if you would rather draw boxes with White-out.
- Search for text or choose a pattern, check the list of matches, and apply. Each match gets a cover you can see, move or undo.
- Click Download. This is the step that deletes the text underneath.
- Open the downloaded file and try to select or search for what you removed. It should not be there.
Pictures and scans
Part of a picture cannot be deleted the way text can. When a box sits over an image, such as a scanned page or a photo of an ID card, that page is saved as a flat image with the box burned in, and PDF Desk tells you which pages it did this to. The rest of the document keeps its selectable text.
Find-every-occurrence works on text only. A scanned page has no text to search (there is no OCR), so on scans use White-out and draw the boxes yourself.
What gets tested
Every release runs automated checks that open statement layouts, find card and account numbers, save the file and read it back to confirm they are gone, plus files built to trick redaction tools. Even so, no tool can promise every trace is gone from every possible PDF, so always open the result and check before sharing it.
Questions
Can I undo a redaction?
Yes, until you download. Covers are ordinary boxes while you work: undo, move or delete them. The text is only removed in the file you download; your original file is not changed.
Is the document uploaded anywhere?
No. The file is opened and rewritten by code running in your browser. Anonymous usage counts are sent, as the privacy page describes, but never the file, its name or its contents.
Does it catch every copy of a number?
Find-every-occurrence covers every match of what you searched for, on every page with text. It cannot know about details written differently (a number with spaces in one place and without in another), so skim the document before you download.
Free. No account. Files are opened inside your browser and never uploaded.