Redact PDF
Permanently black-out sensitive information.
or click to browse — supports PDF files up to 100MB
How to use
- 1 Drop or click to upload your file
- 2 Adjust options if shown
- 3 Click Run Tool
- 4 Download your result instantly
- ✓ Files up to 1GB
- ✓ Unlimited jobs/hour
- ✓ Batch processing
- ✓ Priority support
Files are processed securely and permanently deleted within 1 hour. We never store, read, or share your documents.
Why this works
Permanently remove sensitive information from a PDF \u2014 names, account numbers, SSNs, addresses, anything you can\u2019t share. Redaction is destructive: the text is gone, not hidden.
Redaction is the right tool when a document needs to be shared but contains information that legally, contractually, or ethically can\u2019t go to the recipient. Common cases: bank statements going to a mortgage lender where account numbers should be masked; medical records going to a third party where patient identifiers need removal; legal discovery materials where privileged information must be stripped before production; FOIA-released documents where personal info gets redacted before public release.
The critical thing about redaction \u2014 and the reason this tool exists rather than asking you to draw black boxes manually \u2014 is that redaction must be destructive. Drawing a black rectangle over text in a PDF editor and saving the file does *not* redact: the original text remains in the PDF\u2019s underlying data layer, and any recipient with a copy-and-paste capability or a basic forensic tool can recover it. Redact removes the actual text and image content, then renders the redaction marks (typically black bars) over the now-empty areas.
Two modes. Words mode (the default) takes a comma-separated list of words or phrases to redact \u2014 the tool finds every occurrence across the document and removes them. Case-insensitive matching by default. Use this when you have a known list of items to strip (account number 1234-5678-9012, the name \u201cJohn Smith\u201d, the email john@example.com). Region mode lets you draw rectangles on page previews and redacts whatever falls inside them \u2014 useful when the sensitive content is positional (always in the top-right of the page) or when you need to redact images, not text.
After redaction, run the output through a final visual review. The destructive layer means recovery is impossible \u2014 but it also means mistakes are permanent. If you redacted the wrong word, you can\u2019t un-redact it; you\u2019ll need to start from a fresh copy of the original.
For compliance contexts (HIPAA, GDPR Article 6, attorney-client privilege), redaction alone may not be sufficient \u2014 you may also need to scrub PDF metadata (author, comments, revision history) that can leak information. Use the Flatten tool after Redact for additional cleanup, and check Document Properties in Adobe Reader to confirm metadata is clean.
How it works
-
1Upload your PDFDrop the document containing sensitive information into the upload box.
-
2Pick redaction modeWords mode for a comma-separated list of items to strip throughout. Region mode for visual rectangle selection on page previews.
-
3List the items (Words mode)Type the words or phrases separated by commas \u2014 e.g. `John Smith, 555-12-3456, account 1234567890`. Case-insensitive by default.
-
4Run, review, distributeThe output has the redacted regions blacked out and the underlying content destroyed. Review the result before sharing \u2014 redaction is permanent.
Real-world uses
Lawyers
Discovery production: strip privileged information before producing documents to opposing counsel.
Mortgage applicants
Redact account numbers on bank statements before sending to lenders.
Healthcare admin
Remove patient identifiers from records before sharing with researchers or auditors.
Journalists
Redact source-protective information from documents before publishing them with a story.
Common questions
Is the redacted content really gone?
Yes. The underlying text and image data is removed from the PDF, not just covered. A recipient cannot copy-paste, search, or use forensic tools to recover redacted content. This is the entire point of using a redaction tool versus drawing black boxes manually.
Can I un-redact if I made a mistake?
No. The destructive nature of redaction means once the output is generated, the redacted content is gone. If you redacted the wrong word, restart from a fresh copy of the original. Keep your original safe until you\u2019ve verified the redaction is correct.
Will redaction find every occurrence of a word?
Words mode finds every occurrence of the literal text strings you provide, case-insensitively, across every page. It doesn\u2019t match variants (so \u201cJ. Smith\u201d won\u2019t match if you listed \u201cJohn Smith\u201d) \u2014 list every variant explicitly. For images of text or scanned documents, run OCR first so the text exists; then redact.
Can I redact images, not just text?
Yes \u2014 use Region mode and draw rectangles over the image areas to redact. The image content under the rectangle is replaced with the redaction colour.
Does Redact remove PDF metadata too?
Not automatically \u2014 metadata (author, title, creation date, edit history) is a separate layer. For full sanitisation, run the redacted output through Flatten next, which re-renders the document and clears most metadata as a side-effect.
Is this safe for HIPAA or attorney-client work?
The redaction itself is technically sound, but compliance contexts often require additional steps: metadata cleansing, validated workflows, audit trails. We make no warranty of HIPAA / attorney-client / GDPR compliance for your specific use case \u2014 consult your compliance team. Redact gives you the destructive content removal; the workflow around it is your responsibility.