How to Redact a PDF
Redaction means deleting the text from the file, not painting a black rectangle over it. A rectangle leaves the words underneath: anyone can select, copy or extract them. Real redaction removes the text, the image pixels and the drawings under each box, strips hidden data, and then re-opens the finished file to prove nothing can be pulled out. Checked 17 November 2026: a sample with a Social Security number, email, phone, card number, project codename and salary showed 0 sensitive strings after redaction (pdftotext, pikepdf and pdf.js all agreed), while unmarked sentences stayed searchable.
At a glance
- A black box is not redaction. If the text is still selectable or extractable, it is not redacted.
- Real redaction deletes text, image pixels and vector drawings under each box, then strips metadata, attachments, scripts, comments, form data and bookmarks, and rewrites the whole file.
- Always verify after export: search the finished file for every redacted word. If anything is still findable, do not share the file.
Steps
- Open the PDF in the redaction tool — the file stays on your device.
- Drag a box over each sensitive passage, or search for a word and mark all matches.
- Click Redact & check. The tool removes the text, pixels and drawings under each box, strips metadata, attachments, scripts and comments, and rewrites the file.
- Read the verification report. If any redacted word is still findable, the download stays blocked — fix the marks and try again.
- Download the finished PDF and run the three checks above before sharing it.
What real redaction removes
Three things have to go: the visible text under each box, the image pixels under each box (a scanned page is one big image, so the words are pixels), and the hidden data — document info, XMP metadata, embedded files, JavaScript, comments, form values and bookmarks. A file that keeps any of these can still leak. In our check, the redacted sample kept its unmarked sentences fully searchable, and the 7,855 non-white pixels inside one marked image box came out 100% uniform fill, with the pixels outside the box 100% unchanged.
How to redact with the free tool
Open the PDF in the redaction tool, drag a box over what to remove (or search for a word and mark every match), then apply. The tool removes the content, strips the hidden data, rewrites the file, and re-opens it with a second, independent PDF reader to check every redacted word. If anything can still be pulled out, the download is blocked and the report says what failed.
How to verify the result yourself
Three quick checks: (1) open the finished file and try to select or copy the redacted words — nothing should be selectable; (2) search the file for each redacted word or number; (3) run pdftotext finished.pdf - | grep -i "<word>" — it should print nothing. Also check the file properties for leftover author, title or metadata, and confirm there are no attachments.
Frequently asked questions
Is painting a black box over text enough to redact a PDF?
No. The text stays in the file under the box and can be selected, copied or extracted. Real redaction deletes the text from the file.
Is it free to redact a PDF?
Yes, with the tool linked above: free, no account, no watermark, and the file never leaves your device.
Does redaction work on scanned PDFs?
Yes for the pixels: the marked area is erased from the image. Text search cannot find words on a scan in the first place, because a scan has no text layer — that is also why search-based redaction cannot find words on scans.
What else has to be removed besides the words?
Document info and XMP metadata, embedded files, JavaScript, comments, form data and bookmarks. The tool strips all of them and rewrites the file; a file that keeps an attachment can leak through it.
Related guides: How to check a redacted PDF · How to redact a PDF without Acrobat · How to fill a PDF form · All guides · Tool: PDF Probe