How to redact sensitive information in a PDF
Remove confidential text permanently, rather than covering it with a black rectangle.
The most consequential mistake in document handling is drawing a black rectangle over sensitive text and calling it redacted. The rectangle is a graphic sitting on top. The text is still there, still selectable, still recoverable by anyone who copies the page into a text editor. Organisations have leaked names, addresses and case details exactly this way, in documents they believed were safe.
Real redaction deletes the content and then draws the box. That is what this tool does: the text objects are removed from the document, so copying from the redacted area yields nothing at all.
The steps
- Upload the document.
- Enter the exact text to remove, such as a name or account number.
- Choose whether capitalisation must match, and limit the pages if you want.
- Run the tool and download the redacted file.
Find everything before you remove anything
Redaction acts on the exact text you enter. Names appear in more forms than you expect — with and without a middle initial, in a header, inside a reference number. Search the document first and note every variation, then redact each one.
Case sensitivity matters here. Turning it off catches a name that appears capitalised in a heading and lower case in body text.
It cannot be undone
The content is gone from the output file, which is the guarantee you want and also the risk. Keep your own copy of the original before you redact. You cannot recover a redacted passage from the downloaded document, and neither can we.
What redaction does not cover
Text in a scanned image is a picture, not text, so a search will not find it and redaction cannot remove it. Document metadata — author, title, producer — is separate from page content. If your document is a scan, or if the sensitive detail is an image rather than words, check the result yourself before sending it.
