All articles
Document security

PDF Redaction Done Right: Why a Black Box Is Not Enough

The NamahaPDF Team 6 min read

Some of the most embarrassing data leaks in history were not hacks. They were documents that had been “redacted” by drawing a black box over the secret part — leaving the actual text sitting underneath, ready to be copied out by anyone who clicked and dragged. Redaction is one of the easiest security tasks to get wrong, and the failure is invisible until it is too late.

Why a black box fails

A PDF is not a picture of a page; it is a set of instructions. The text you read is stored separately from the graphics drawn over it. When you place a black rectangle on top of a name, you have added a shape — you have not removed the name. The original characters are still in the file's content stream, which means:

  • The hidden text can be selected, copied, and pasted straight out.
  • If the box was added as an annotation, it can often simply be deleted or dragged aside.
  • Search tools and screen readers can still read the “covered” words.

The same trap applies to images: lowering a layer's opacity or covering a photo does not delete the underlying pixels.

The leaks you do not see coming

Even a careful visual redaction can miss data that never appears on the page:

  • Metadata — author, title, and the software used, stored in the document properties.
  • Revision history — PDFs can be saved “incrementally,” appending changes while keeping older versions recoverable inside the same file.
  • Comments, form fields, and attachments that carry text outside the main page content.
  • Hidden layers and embedded thumbnails that may preview the original, un-redacted page.

What real redaction does

Proper redaction does not hide content — it destroys it. A correct tool will:

  1. Remove the underlying content — the actual text runs and image data beneath the redaction area, not just paint over them.
  2. Flatten the result so the redaction is part of the page, not a movable or deletable object.
  3. Strip metadata and other off-page content that could reveal what was removed.
  4. Re-save the file cleanly, without preserving an incremental history of the original.

A checklist before you share

Run through this every time a redacted document is about to leave your hands:

  1. Search the document for the names or numbers you removed — you should find nothing.
  2. Select and copy the redacted regions; confirm nothing pastes out.
  3. Check the document properties for author, title, and history.
  4. Look for comments, form fields, and attachments.
  5. Confirm the file was re-saved fresh, not appended to the original.
Redaction is a deletion problem disguised as a drawing problem. If the data still exists in the file, it is not redacted — it is hidden.

The takeaway

If a tool lets you “redact” by drawing a rectangle and nothing else, be suspicious. True redaction removes the content from the file and cleans up everything that points back to it. When the stakes are real, verify with the copy-paste test, and use a tool built to redact permanently rather than one that merely covers the page.

Keep reading