Guide

How to redact a PDF permanently

Redacting a PDF means removing information from the file, not hiding it on screen. Here is the method, the traps, and how to check the result.

Updated

In short

  • Work on a copy, never on the original.
  • Find every value to remove: names, addresses, identifiers, and what identifies indirectly.
  • Remove the content itself: a shape drawn over the text leaves it in the file.
  • Clean the metadata, annotations, form fields and attachments.
  • Check the copy before sending it: select, search, copy and paste.

What redacting a PDF means

A PDF is not a picture of a page. It holds text objects with their positions, images, annotations, form fields, layers, attachments and metadata. What you see is only the rendering of all of that.

Redacting means removing the information from every place it lives in the file, so that nobody can recover it by selecting the text, copying it, searching it, extracting it with a tool, or reading the document properties. Hiding it on screen is not enough.

1. Start from a copy

Keep the original as it is: you may need it again, and a redaction mistake on the only copy cannot be undone. Name the redacted copy so that nobody mistakes one for the other.

2. Find what has to go

List what the recipient must not see, then look for every occurrence. A name appears in the header, the body, the signature, a table and sometimes a footnote; an identifier can be written with or without spaces.

  • Direct identifiers: names, postal and e-mail addresses, phone numbers.
  • Numbers: IBAN, social security number (NIR), company numbers (SIREN, SIRET), payment cards, passport and identity card numbers, licence plates, client or file references.
  • Indirect identifiers: a job title, a date, a place or a rare event that points to one person once the name is gone.
  • What is not text: a signature, a stamp, a photo, handwriting. These need a zone drawn by hand.

3. Review each finding

Automatic detection saves time but does not decide. Some values must stay (the name of the organisation sending the document, a public figure), others are missed (a name in an unusual place, a badly scanned page). Read the document once with the findings in front of you, accept or reject each one, and add what is missing.

4. Remove the content, then the hidden layers

The removal has to affect the file's content, not its appearance. The safest method rebuilds each page as an image with the masked areas painted in, so that no text object survives underneath. Then remove what the page does not show:

  • metadata (author, organisation, title, subject, keywords, editing software, history);
  • annotations and comments, which keep their own copy of the text;
  • form fields and their values;
  • attachments, layers and bookmarks.

5. Check the copy

Open the redacted copy in another reader. Try to select the text under a masked area, search for a masked value, copy a whole page into a text editor, and look at the document properties. On a scanned page, remember that a text layer added by OCR also holds the words.

A check by hand finds the obvious mistakes, not all of them. A tool that rereads the exported file on its own, independently of the export, is a better guarantee that nothing is left.

Redacting a PDF with Obscelis

Obscelis follows these steps on your computer, without sending the document anywhere. It finds names, addresses, e-mail addresses, phone numbers, IBAN and BIC, NIR, SIREN, SIRET, payment cards and other identifiers, validates those with a check key, and reads scanned pages with OCR. You accept or reject each finding and draw zones over what is left.

The export writes a new file next to the original. Its pages are rebuilt as images with the accepted areas painted black, metadata and hidden content are removed, then the copy is reread independently: if a masked value can still be found in it, the export is refused.

Questions

Can I redact a PDF with a free reader?

Most free readers can draw a black rectangle or a highlight, which does not remove the text underneath. You need a tool with a real redaction function that removes the content, then cleans the hidden layers and metadata.

Does a redacted PDF stay searchable?

Not when its pages are rebuilt as images, which is the safest method: the masked text is gone, and so is the text layer. If the recipient needs to search the copy, OCR can add a new text layer afterwards, which only contains what is still visible.

Is redacting the same as anonymising?

Not quite. Redacting removes chosen information from a document. A document is anonymous only if nobody can identify the people in it, even by combining what remains with other information. Redaction is the means; whether the result is anonymous is an assessment you make.

Can I redact a scanned PDF?

Yes, provided the text is read first: OCR turns the page image into words with their positions, which can then be found and masked. Check the result carefully, since OCR can misread a badly scanned page.

Try Obscelis for 30 days

Download it for macOS, Windows or Linux: the trial needs no account and no card. A question, or a demonstration for your team? Write to us.

Download ObscelisWrite to us

latelierdudev.contact@gmail.com

Opens your e-mail app. We only use your address to reply.