Why Black-Box PDF Redaction Fails

Every year, court filings, government reports and company documents are published with "redacted" text that anyone can copy and paste. It is almost never a sophisticated attack. Someone drew a black box over the words, the PDF looked right, and it was sent.

Want to skip the reading? The tool this guide covers is free and needs no sign up.

PDF Redaction Checker

Why a black box is not a redaction

A PDF page is built in layers. Text is placed on the page, and anything drawn afterwards is painted on top. When you draw a black rectangle over a sentence, the sentence is still there, underneath. It is hidden from your eyes, but not from text selection, search, copy and paste, or any program that extracts text.

We tested this ourselves. We made a PDF with the line "CONFIDENTIAL SETTLEMENT AMOUNT 9,50,000" and drew a black box over it. On screen it looked properly redacted. Our redaction checker still pulled the full line, amount included, straight out of the file.

Five ways redactions fail

A black box is the best-known mistake, but there are others that are just as easy to make:

1Shapes or highlights drawn over text. The text stays in the file.
2Redaction marks that were never applied. Acrobat lets you mark areas for redaction and then requires a separate "Apply" step. Skip it and nothing is removed.
3An earlier version left inside the file. Many programs save edits by adding them to the end of the PDF. If the redaction was a later save, the unredacted version is still in there.
4Leftovers in metadata, comments or attachments. A title like "Settlement with R. Sharma", a reviewer comment quoting the original, or the source spreadsheet attached to the PDF.
5White text, or text behind a white box. It does not show, but it is fully searchable.

How to redact properly

Use a tool that deletes the content under each box, not one that draws over it. In Acrobat Pro that is the Redact tool, followed by Apply. On WorkSathi, Redact PDF rebuilds each page you mark so the words beneath no longer exist in the file.

Then make sure there is only one version of the file. Save it as a new file rather than saving over the old one, or run it through Flatten PDF, which writes a fresh copy.

Redacting a scanned page needs extra care. If the scan has an OCR text layer, the words exist twice: once in the image and once as hidden text. Both have to go.

Check before you send

The simplest manual test is to press Ctrl+A on each redacted page, copy, and paste into a plain text editor. If any redacted words appear, the redaction failed.

That works for one or two pages. For longer files, our PDF Redaction Checker renders every page, compares the text the file contains with what is actually visible, and lists anything hidden under boxes. It also checks for unapplied marks, earlier versions, white text and ID numbers you may have missed.

Frequently asked questions

Does flattening a PDF remove text under black boxes?

Not necessarily. Flattening merges annotations and form fields into the page, so a black box drawn as an annotation becomes part of the page, but the text underneath can still be in the content. Only a true redaction, which removes or rasterises the covered text, makes it unrecoverable.

Is converting the PDF to an image safe?

Converting every page to an image does remove the hidden text, because only the pixels remain. But the result has no selectable text at all, which makes it inaccessible and unsearchable. Redacting only the marked areas is the better approach.

Can metadata reveal redacted information?

Yes. The title, subject, keywords and edit history are separate from the pages and are not touched by redaction. Check them with a metadata checker and clear anything sensitive.