How to Redact a PDF (Black Out Sensitive Text Properly)

Quick quiz: you open a PDF, draw black rectangles over the sensitive parts in an image editor, save it. Is the text hidden? No. Anyone can select the text underneath, copy it, and read every word. This mistake has burned lawyers, journalists, and governments.

Real redaction means the hidden text is gone — covered permanently so it can't be selected, copied, or recovered. Here's how to do it right, in about a minute.

Here's the uncomfortable part most tutorials skip: a lot of 'redaction tools' just paint black rectangles on top of your text. The words are still there underneath, sitting in the file, waiting for someone to select and copy them. That has caused real scandals — hidden names in court filings recovered by journalists, personal data leaking from 'redacted' government reports.

So the golden rule of this guide: after you redact, verify like an attacker. Try to select the blacked-out text. Try to copy it. Zoom in to 400%. If you can still get at the words, they're not hidden — they're just covered. We'll walk through exactly how to check.

Step 1: List what needs hiding

Before you start, write down every term to redact: names, phone numbers, addresses, ID numbers, account numbers. Be thorough — miss one occurrence and the whole exercise is pointless. Search the document for each term first so you know what you're dealing with.

Step 2: Redact every occurrence

Open the Redact PDF tool, upload your file, and type each term on its own line. The tool finds every match across all pages and covers each one with a solid black box baked into the file.

Don't redact one page at a time or eyeball it. Let the tool catch all occurrences — humans miss things, especially in long documents.

Step 3: Verify before sharing

This step is non-negotiable. Open the redacted PDF and try to select the text under the black boxes. Try copying it. Try searching for the hidden terms with Ctrl+F. If anything leaks through, redo it.

For truly sensitive documents — legal filings, medical records — have a second person check too. Paranoia is a feature here, not a bug.

What about scanned PDFs?

Scanned PDFs are just images of text — there's no text layer to search, so redaction tools can't find matches. Run the scan through an OCR tool first to add a text layer, then redact. Or physically cover the sensitive areas on the scan before sharing.

The metadata trap most people forget

Even if the visible text is perfectly covered, PDFs carry invisible baggage: the author's name, the software used, creation dates, sometimes even edit history. If you're redacting a document for privacy, that metadata can give you away. Run the file through a metadata editor afterwards and strip out anything identifying — it takes twenty seconds.

Work on a copy, always

Never redact your only copy of a document. Make a duplicate first, redact the duplicate, and keep the original somewhere safe and private. Redaction is a one-way street — once text is covered and the file is saved, there's no "undo" that brings it back.

Common questions

Keep reading