Does drawing a black box over text in a PDF hide it?

No. A PDF stores text as text and the rectangle as a separate drawing instruction. The text remains underneath and can be recovered by selecting it and copying, or by searching the document. This is how numerous public redaction failures happened.

How do I properly redact a PDF?

Use a tool that removes the underlying text objects rather than covering them, then flatten the document so annotations and form values are not retained as separate data, and clear the document metadata. Verify afterwards by selecting all text and searching for the redacted term.

Is converting a PDF to images a reliable way to redact?

Yes, and it is the bluntest reliable method. Rasterising each page discards the text layer entirely, so there is nothing left to extract. The trade-off is that the result is no longer searchable or selectable and the file is usually larger.

How can I check whether my redaction worked?

Open the file fresh, select all the text including over the blacked-out areas, and paste it into a text editor. Then search the document for the exact redacted term. Extracting the full text and searching that catches occurrences on pages you did not check by eye.

Why Blacking Out Text in a PDF Does Not Redact It

Published 22 August 2026 · 7 min read

This is the most consequential misunderstanding in everyday PDF use, and it has embarrassed governments, law firms and newspapers. Drawing a black box over a name in a PDF does not remove the name. It draws a black box. The name is still there, underneath, intact — and anyone can retrieve it by selecting the text and pasting it elsewhere.

A PDF is not a picture of a document. It is a set of instructions: draw this text here, then draw this rectangle there. Adding a rectangle adds an instruction. It does not delete the earlier one.

How the mistake happens

Every one of these feels like redacting and none of them is:

Drawing a filled black rectangle with an annotation or drawing tool.
Highlighting text in black with a markup tool.
Setting the text colour to white so it "disappears" against the page.
Covering a region with an opaque image.
Placing a black box in Word, then exporting to PDF — sometimes safe, often not, depending on how the export handles layering.

In each case the original text remains a text object in the file. Select over the blacked-out region, copy, paste into a text editor, and it reappears. You do not need special software; a PDF reader and a keyboard are enough.

It is worse than just selectable text

Even if you remove the visible text correctly, a PDF carries other copies of the same information in places people forget to look:

Document metadata — author, title, keywords, the software used, and sometimes the full file path on the original machine, which can leak a name or an internal project code.
Annotation contents — comment text is stored separately from the page and survives flattening the visual layer if done wrong.
Form field values — a filled field holds its value as data, not as page content.
Embedded thumbnails and previews — occasionally generated before the redaction was applied.
Earlier revisions — PDFs support incremental saving, where edits are appended and the previous version remains in the file. A "removed" page can still be recoverable.

How to redact properly

Real redaction removes the underlying content rather than covering it. There are two reliable approaches.

Remove the text, then flatten

Use a tool that deletes the text objects in the selected region rather than drawing over them, then flatten the document so annotations and form data are baked into the page and no longer exist as separate retrievable objects.

Redact PDF — select regions and remove the content beneath them.
Find and redact — search for a name, account number or term and remove every occurrence, which is far safer than spotting them by eye across fifty pages.
Flatten PDF — merge annotations and form fields into the page so nothing remains as separate data.
Edit PDF metadata — clear author, title and producer fields.

Or destroy the text layer entirely

The blunt, extremely reliable method: cover the sensitive areas, convert every page to an image, then rebuild a PDF from those images. Rasterising discards the text layer completely — there is no text left to recover, because there is no text at all.

The cost is that the result is not searchable or selectable, and the file is usually larger. For a document going to an adversarial party, that is often a trade worth making.

How to check your work

Never assume. Verify, every time, before the document leaves your hands:

Open the redacted file in a fresh reader.
Select all the text on the page — including over the black areas — copy it, and paste it into a plain text editor. Read what appears.
Use the reader's search function to look for the exact term you redacted. If it is found, you have not redacted it.
Check the document properties for author and title metadata.
Extract the text from the whole document and search that output. This catches occurrences on pages you did not think to check.

That last step is the one that catches real mistakes — the second mention of the name, forty pages later, that nobody remembered was there.

One more reason to do it locally

Redaction is applied to precisely the documents you least want to upload: legal filings, medical records, HR files, anything under disclosure. Uploading a document to remove its secrets, before the secrets have been removed, sends the unredacted version to a third party. Doing the work locally avoids that ordering problem entirely.

Try it yourself

Redact PDF — runs in your browser, nothing is uploaded.

Frequently asked questions

Does drawing a black box over text in a PDF hide it?
No. A PDF stores text as text and the rectangle as a separate drawing instruction. The text remains underneath and can be recovered by selecting it and copying, or by searching the document. This is how numerous public redaction failures happened.
How do I properly redact a PDF?
Use a tool that removes the underlying text objects rather than covering them, then flatten the document so annotations and form values are not retained as separate data, and clear the document metadata. Verify afterwards by selecting all text and searching for the redacted term.
Is converting a PDF to images a reliable way to redact?
Yes, and it is the bluntest reliable method. Rasterising each page discards the text layer entirely, so there is nothing left to extract. The trade-off is that the result is no longer searchable or selectable and the file is usually larger.
How can I check whether my redaction worked?
Open the file fresh, select all the text including over the blacked-out areas, and paste it into a text editor. Then search the document for the exact redacted term. Extracting the full text and searching that catches occurrences on pages you did not check by eye.