Redact a PDF
What is redacted really disappears from the file.
Your files never leave your deviceDrop the PDF to redact
or click to browse your files
How to redact a PDF, step by step
Redacting is not drawing a black rectangle over a passage: it is removing that passage from the file. Here is the full procedure in VellumPDF, where everything is computed on your own device.
Drop in the PDF you want to redact
Drag the document onto the drop area, or click to browse your files. The first page appears straight away. A password-protected file is refused with a plain message: unlock it first with the dedicated tool.
Draw the areas to mask on the preview
Click and drag directly on the displayed page to draw a rectangle over each passage to remove. A counter shows how many areas sit on the current page and how many in total. “Undo last”, “Clear all” and the cross on each area fix a mis-drawn box.
Let patterns propose the repetitive areas
The “Redact by pattern” panel searches the whole document for AHV numbers, IBANs, e-mail addresses, Swiss phone numbers, dates, amounts in francs or euros, a free text, or a regular expression of your own. Matches are proposed on the preview, never applied on their own.
Check page by page before applying
Move through the pages with the arrows and read every area again, proposed or hand-drawn. One area too many is removed with a click, and “Remove proposed zones” cancels the whole automatic detection at once.
Click “Redact and download”
VellumPDF rebuilds the document without the masked content, then downloads it. Copying and pasting over the redacted area returns nothing, and your reader’s search no longer finds the word — the test a black rectangle fails immediately.
The trade-off is stated openly: every redacted page is rebuilt as an image, so its text is no longer selectable; the other pages keep theirs unless you tick “Turn every page into an image”. If you need a searchable file, run the result through character recognition. And on a scan with no text layer, pattern detection finds nothing — run OCR first, or draw the areas by hand.
Frequently asked questions
Can redacted text in VellumPDF be recovered by copy-paste?
No. Unlike a black rectangle drawn on top of text — whose content remains copyable — VellumPDF genuinely removes the redacted passage from the file it produces. You can verify yourself that copying the masked area no longer returns anything.
How do I redact a passage in a PDF?
Open the document, draw a rectangle over each area to hide directly on the page preview, then confirm: the covered content is deleted before the final PDF is generated.
Does pattern-based redaction detect everything automatically?
No. Proposed zones — IBAN, Swiss AVS number, e-mail address, phone, dates, amounts, free text or a regular expression — are suggestions to review on the preview: nothing is redacted until you click “Redact and download”. You can remove a single proposed zone or all of them at once, and add your own by hand. A scanned document with no text layer must go through the OCR tool first.
Is the file sent to a server for redaction?
No, everything happens on your own device in the browser; the document is never uploaded. The tool is free and does not work on a file that is already password-protected — you need to unlock it first.
Why does the redacted document have no selectable text left?
Because VellumPDF’s redaction does not lay a rectangle on top: each page that carries an area is rendered as an image, the areas are painted black on that image, and the page is replaced by the image. The original text of those pages therefore no longer exists in the file produced. The trade-off is deliberate: all the text on a redacted page becomes an image; the other pages keep their text unless you tick “Turn every page into an image”. If you need a searchable document, run the result through text recognition.
Can a scanned document be redacted?
Yes, and it is in fact the simplest case, since the method already works on the page image. Automatic pattern detection, however, needs a text layer: on a raw scan it finds nothing. Run the document through text recognition first, or draw the areas by hand.
How do I check that the redaction worked?
Open the file produced and search for the removed term with your reader’s search function. If it returns nothing, the text is no longer in the file. That is exactly the test a black rectangle fails: the word is still findable underneath.