Redact a PDF
so it stays redacted.
Permanently black out emails, phone numbers, card and IBAN numbers — or any exact words you list. The hidden text is deleted from the file, not just covered with a box, so it can't be copied, searched, or recovered.
Redact a PDF →Available on Pro (€9.99/mo), Business (€49.99/mo), and Enterprise. EU-hosted, files deleted after 1 hour.
How it works
What it can and can't redact
✅ Selectable text in the PDF
If you can highlight the text with your cursor, we can redact it. Detectors match patterns (emails, phone numbers, SSNs, cards, IBANs) and your word list matches exact strings — both remove the underlying glyphs from the content stream.
✅ Multiple detectors and terms at once
Combine any detectors with any number of literal terms in a single pass. Every match across every page is redacted together.
❌ Text inside scanned images
If your PDF is a scan (a photo of a page), the text is pixels, not characters — there's nothing for the detectors to match. Run OCR first to add a real text layer, then redact.
❌ Information you didn't ask us to remove
Because it's rule-based, we only redact what your detectors and word list match. A phone number written in an unusual format, or a name spelled a different way, won't be caught unless you list it. Always review the result.
Does redaction also remove hidden metadata?
Yes. Every redaction also clears the document's own metadata — author, title, subject, keywords and the XMP packet — because a spotless page is no use if File › Properties still names who wrote it. For EXIF/GPS metadata in images, audio and video, use the metadata strip tool.
Is there a file size limit?
Pro users get 500 MB per file. Business users get 5 GB. Redaction is fast — even a long contract processes in well under a minute.
Can I redact PDFs through the API?
Yes. Send the file to /api/v1/convert with op=redact and a redact object listing your patterns and terms. See the API docs for the full request format.
Who needs the text actually gone
A black rectangle drawn over a name in a PDF viewer is a picture of a redaction. The text is still in the file — select it, copy it, or run any extraction tool, and it comes straight back. Every case below has been lost that way at least once.
Law firms and notaires
Disclosure bundles go to the other side. Third-party names, addresses and account numbers have to come out while the argument stays readable — and once the file has left your office, you do not get to correct it.
Accountants, payroll and finance
Bank statements and payroll runs get shared with auditors, lenders and clients. IBANs and salary lines are precisely what must not travel with them.
HR and internal investigations
A grievance file or a restructuring list has to be read by managers who need the case but not the identities. Same document, different audience, no second version to maintain.
Hospitals, clinics and healthcare administration
Procurement contracts, tender responses, supplier agreements, incident reports and staff files all leave the building carrying names, salaries and bank details. Redaction removes the text itself — and clears the Author field along with it, the one that quietly names whoever prepared the document.
Journalists, NGOs and public bodies
Source protection and freedom-of-information responses have the same requirement: publish the document, not the people in it. The metadata matters as much as the page — more than one leaked file has been traced back to its author by the Author field.
What we do about it
Matches are removed from the PDF content stream, not covered over. Document metadata is cleared on every run. Files are processed on EU servers, never shared with third parties, deleted within an hour, and never used to train AI. There is no human review step.