Black boxes don't redact a PDF. Here's what does.

Court filings, corporate disclosures and government releases have all leaked this way. Somebody drew a black box over a name, sent the PDF, and the name was still sitting underneath — selectable, copyable, indexed.

A PDF is not a picture of a document. It's a set of instructions: draw this text, in this font, at this position. A black rectangle is one more instruction, drawn on top. The text instruction underneath is untouched.

Why the box fails

Select the area and copy it. Or open the file in any tool that extracts text. Or just hit Ctrl+A, Ctrl+C and paste into a text editor. The hidden words come straight out, because nothing ever removed them.

• Copy and paste reveals it in seconds, with no special software.
• Text extraction pulls the whole page, boxes included.
• Search engines index the text if the file is ever published.
• Document metadata may separately hold the author, original filename and revision history.

If you can select the text under the box, so can everyone you send the file to.

What real redaction does

Proper redaction deletes the underlying content and then covers the area, so there is nothing left to recover:

1. 1The text or image objects inside the marked area are removed from the page.
2. 2A filled shape is drawn over the region so the gap is visibly marked.
3. 3The page is written out without the removed objects.

The Redact PDF tool does this in the browser, so the document you're trying to keep private is never sent anywhere in the process — which would be a strange first step for a privacy operation.

Check your work before you send

Redaction is one of the few document tasks where you should actively try to break your own output:

• Open the redacted file and try to select the covered area. Nothing should come out.
• Copy the whole page into a plain text editor and read what appears.
• Search the document for a name or number you removed.
• Check the pages you weren't thinking about — headers, footers, appendices and email chains repeat information.

Converting the result with PDF to Text is the fastest complete check: it extracts everything a machine can read, so if a redacted name shows up there, the file isn't clean.

The parts people forget

• Metadata — author, company, original filename, and the software used.
• Attachments embedded inside the PDF.
• Bookmarks and comments, which often quote the text they point at.
• Scanned pages that have been OCR'd, where an invisible text layer sits under the image.

That last one catches people out. The page looks like an image, so a box over it seems safe — but the OCR layer underneath is real text. Redact it as text, not as a picture.

When you only need to remove a page

If an entire page is sensitive, don't redact it — take it out with Remove Pages, or pull only the pages you want to share using Extract Pages. A page that isn't there can't leak.

Common questions

Can redacted text be recovered?

Not if the content was genuinely deleted. It can be trivially recovered if someone only drew a shape over it — the text objects are still in the file and copy straight out.

Does a black box work if I flatten the PDF afterwards?

Flattening to an image does remove the text layer, but it also destroys every other page's selectable text, balloons the file size, and leaves metadata untouched. Real redaction is a better answer than a lossy workaround.

How do I check a PDF is properly redacted?

Extract the document to plain text and search it for what you removed. If nothing comes back, the content is gone; if it appears, the file is still leaking.