How to redact a PDF in four steps
- Open your PDF. Drag it onto the box above or tap Choose file. It opens instantly because nothing is uploaded.
- Mark what to remove. Drag across anything you want gone. To remove every mention of a name or account number, type it into Search & redact and press Redact all. Or tap Find sensitive data and approve the suggestions one by one.
- Check your boxes. Scroll through every page. Click a box to move or resize it, or press Delete to remove it. Undo and redo work with Ctrl+Z and Ctrl+Shift+Z.
- Apply and verify. Press Apply redactions & verify. When you see the green Verified badge, download the file. If verification finds a problem, the tool tells you exactly what failed so you don’t share an unsafe file.
Why “drawing a black box” isn’t enough
Some of the most embarrassing data leaks have come from PDFs where someone drew black rectangles over sensitive text. The rectangles looked convincing, but the original text was still in the file — anyone could select it, copy it or run it through a text extractor. Court filings, government reports and contracts have all been exposed this way.
Proper redaction means removing the information, not hiding it. This tool does that by rendering each page you redact to a high-resolution image (200 or 300 DPI), burning the black boxes into the pixels, and building a new PDF from those images. The hidden text layer, fonts and drawing instructions for those pages are thrown away. Pages without redactions are copied across exactly as they were, so the rest of the document stays sharp and searchable.
Metadata is another quiet leak. A PDF’s properties often contain the author’s name, the company, the original file name or editing software. The exported file carries none of that.
What to redact
Most people redact one of three things: personal data (names, addresses, dates of birth, ID numbers), financial data (card and bank numbers, salaries, account balances) and privileged information (legal advice, trade secrets, other people’s details in a document you have to disclose). Think about indirect identifiers too: a job title plus a small town can identify someone just as clearly as a name.
Don’t forget content that isn’t body text: headers and footers, page numbers that reveal document length, signatures, photos, stamps, comments and form fields. The flattening step removes comments and form fields on redacted pages automatically.
Limits to be aware of
- Redacted pages become images, so they are larger than text pages and can’t be searched. That’s the trade-off that makes the redaction permanent.
- Auto-detection is a helper, not a guarantee. It finds common patterns, but names, handwritten notes and unusual ID formats need a human eye. Always review every page.
- OCR accuracy depends on scan quality. Faint or skewed scans may miss words — draw boxes by hand where needed.
- Very large files are limited by your device’s memory. Up to a few hundred pages is fine on most laptops.
Who uses this tool
Lawyers and paralegals preparing disclosure bundles, HR teams sharing investigation reports, journalists protecting sources, landlords and property managers sending documents to third parties, and anyone sending a bank statement or ID copy who wants to hide the parts that aren’t needed. Because your file never leaves your device, it’s suitable for confidential material that you shouldn’t upload to an online PDF service.
Frequently asked questions
Is this a real redaction or just a black box drawn on top?
It is a real redaction. Every page that has a box is re-drawn as a flat image with the box burned in, so the text underneath no longer exists in the file. A box drawn in a normal PDF editor only covers the text, which can still be copied, searched or extracted.
Is my PDF uploaded anywhere?
No. The PDF is opened, redacted and saved entirely inside your browser using pdf.js and pdf-lib. Nothing is sent to our servers, and you can disconnect from the internet after the page loads to prove it.
What does the Verified badge check?
After exporting, the tool re-opens the new file and checks that no extractable text sits under any box, that the words you redacted don't appear anywhere in the document, that the pixels under each box are solid black, that the page count is unchanged and that the metadata has been removed.
Why can't I select text on the redacted pages any more?
Because those pages are now images. That is what makes the redaction permanent. Pages you didn't touch are copied unchanged, so their text stays selectable and searchable.
Can it find personal data automatically?
Yes. 'Find sensitive data' looks for email addresses, phone numbers, payment card numbers (checked with the Luhn algorithm), IBANs, US Social Security numbers, UK National Insurance numbers, Australian Tax File Numbers, Canadian SINs, Indian Aadhaar and PAN numbers, IP addresses and dates. You approve or reject each match.
Does it work with scanned PDFs?
Yes. If a page has no selectable text, run OCR (optical character recognition). The text is read on your device with Tesseract so search and auto-detect work on scans too. You can also draw boxes by hand on any page.
Does it remove metadata?
Yes. The exported file has no title, author, subject, keywords, creation software, XMP metadata, bookmarks, attachments or JavaScript from the original.
Can I redact on my phone?
Yes. Draw boxes with your finger in Draw mode; switch to Select mode to scroll. Very large PDFs (hundreds of pages or more than 50 MB) work best on a computer.
What if my PDF is password-protected?
If it needs a password to open, unlock it first in your PDF reader (Print → Save as PDF creates an unlocked copy). If it only has editing restrictions, the tool can still redact it, but every page will be flattened.