How to redact a document so it can’t be undone
A black box over text in a PDF often leaves the text underneath, ready to copy. Here is why redactions fail and how to make sure hidden information is really gone.
Redaction failures make the news regularly, including in court filings and government reports and nearly all of them happen the same way: someone drew a black box over text and assumed the text was gone.
Why black boxes fail
A PDF isn’t a picture of a page. It is a set of instructions: draw this text here, this image there, this rectangle on top. Drawing a black rectangle in most PDF editors adds one more instruction. The text underneath is still in the file, in full and anyone can select it, copy it into another document or simply delete the rectangle.
Images have a version of the same problem. Some editors keep layers and some cameras and programs keep a small thumbnail of the original picture inside the file, which a redacted version can still carry.
What proper redaction does
Proper redaction destroys the information rather than covering it. For an image, that means changing the pixels themselves and saving a new, flat image with nothing hidden in it. For a PDF, it means either a dedicated redaction feature that deletes the underlying text or turning the page into an image so there is no text left to find.
The redaction tool does the first: it paints solid blocks or pixelates or blurs, directly into the pixels and saves a new image without metadata or thumbnails. Solid blocks are the safe choice for text; blur and pixelation are for faces in photos.
Redacting a PDF step by step
- Turn the pages you need to redact into images with PDF to PNG.
- Redact each image, covering everything sensitive with solid blocks.
- Rebuild the PDF from the redacted images with image to PDF.
The result has no text layer, so nothing can be copied from under the boxes. The trade is that the document is no longer searchable, which for a redacted copy is usually what you want anyway.
The other places information hides
Photos carry the place they were taken; remove it as described in how to remove location data from a photo. Documents carry author names and comments. Screenshots show more than you think, such as tab titles, notification previews and the time. Look at the whole image at full size before you send it.
Check before you send
Try to select the text under each box. Search the file for a word you hid. Open it in a different program. Doing it on your own device matters too: uploading a document to a website to redact it hands the unredacted original to that website first.
Common questions
Is drawing a black rectangle in a PDF enough?
Usually not. In most PDF editors the rectangle is a new shape laid over the page and the text beneath it is still in the file. Anyone can select it, copy it or delete the rectangle.
Is blurring or pixelating safe?
Light blur and fine pixelation can sometimes be reversed, especially for text in a known font, by trying candidate words until the blurred result matches. Solid blocks are the only method with nothing to reverse. Use blur for faces in photos, not for account numbers.
What else can leak from a redacted document?
Metadata, such as the author name and editing history in a document or GPS coordinates in a photo. Embedded thumbnails, which some cameras and editors store as a small copy of the original image. And text layers in scanned PDFs that have been through OCR.
How do I check that a redaction worked?
Try to select and copy the text under the box, search the document for the hidden words and open the file in a different program. If you redacted an image, check it at full size, since a small preview can hide a partial redaction.