Quick answer: Drawing a black rectangle over text does not remove it — anyone can select and copy the text underneath. Use Find & Redact to genuinely delete the text, then Remove Metadata to clear hidden document data.
The black-box mistake that keeps making headlines
Courts, governments and companies have repeatedly leaked names, prices and classified details by "redacting" PDFs with drawn black rectangles. The text is still in the file — selecting the area and copying it (or just converting the PDF to text) reveals everything.
Real redaction means deleting the content, not covering it.
How to redact a PDF properly
1. Open the redaction tool
Go to Find & Redact. Fittingly for a redaction task, your file is processed entirely in your browser — the sensitive document never leaves your device.
2. Find what needs to go
Search for the terms to remove — names, account numbers, emails. The tool finds every occurrence across all pages, so nothing slips through on page 47.
3. Apply redaction
The matched text is removed from the file's content and covered with a marker. After saving, the text no longer exists in the document — there is nothing to copy, extract or un-hide.
4. Verify
Reopen the result and try selecting/searching for the redacted terms. You'll find nothing — that's the difference from a black box.
Redaction alone isn't enough — clear the hidden layers
A PDF carries more than its visible pages:
| Hidden data | Risk | Tool |
|---|---|---|
| Metadata (author, company, software, edit dates) | Reveals people and organizations | Remove Metadata |
| Embedded attachments | Whole files hidden inside the PDF | Edit Attachments |
| JavaScript and active content | Tracking or malicious behavior | Sanitize PDF |
| Form field history | Previously entered values | Flatten PDF |
| Annotations and comments | Internal notes left in review | Remove Annotations |
A thorough pre-publication pass: Find & Redact → Remove Metadata → Sanitize → Flatten. You can chain these in one go with Workflows.
When to redact
- Legal discovery — produce documents without privileged content
- FOIA / public-records releases — personal data must be removed, not masked
- Sharing contracts — hide pricing from one party, terms from another
- Screenshots-to-PDF reports — strip customer data before circulating
- Selling a business document template — remove your client specifics first
Frequently asked questions
Is highlighting text in black the same as redacting? No — that's exactly the mistake this guide is about. Highlighting changes appearance; the text remains fully extractable.
Can redacted text be recovered? Not from a properly redacted file — the characters are deleted from the content stream, not hidden.
What about scanned PDFs? A scan is an image, so text-search redaction needs OCR first to locate the words.
Why does in-browser processing matter here especially? Uploading a document to a redaction service means sending the unredacted version — with all its secrets — to someone's server. Local processing avoids that entirely.
Conclusion
Redaction is one of the few PDF tasks where doing it wrong is worse than not doing it at all, because it creates false confidence. Delete the content, clear the metadata, and verify.
Related guides: