It's one of the most consequential mistakes in document handling, and it's been made by law firms, government agencies, and Fortune 500 companies — repeatedly, publicly, and expensively.
Someone needs to hide a name, an address, or a salary in a PDF. They open a PDF editor, draw a black rectangle over it, save, and send. It looks perfect. The text is invisible.
The text is still there.
What a black box actually does
A PDF is a stack of objects, not a picture. When you draw a black rectangle over some text, you have added one more object on top — a filled shape. You haven't touched the text underneath. It's still in the file, still a text object, still fully intact. You've just put something opaque in front of it.
Which means anyone can retrieve it, with no special tools and no skill:
- Select the text and copy it. Drag across the black box, hit Ctrl+C, paste into a text editor. The hidden text appears.
- Search the document. Ctrl+F for the name you "hid". It matches.
- Convert the PDF to Word or text. Every word comes through, rectangle or not.
- Move or delete the rectangle in a PDF editor. It's just an object; it can be selected and removed.
The same is true of a white box (invisible-on-white, still copyable), a highlighter drawn over text, and covering it with an image. All of them are decoration. None of them remove anything.
The mistakes that look similar but aren't fixes
- Cropping the page. Cropping hides the area outside the crop box — it doesn't delete it. The content is still in the file and can be revealed by resetting the crop.
- Flattening after drawing a box. This one feels like it should work, and it does destroy the live text layer — but the result is a picture of a page with a black box on it. That's better than nothing, and it's still not the right tool: you've rasterised your entire document (losing searchable text everywhere) to solve a problem redaction solves properly.
- Deleting the page. If the sensitive content is on a page you don't need, deleting the page does genuinely remove it. But it doesn't help when you need to keep the page and hide part of it.
What actually works: redaction
Redaction is a fundamentally different operation. It doesn't cover the content — it deletes the underlying text and image data and then draws the box. There is nothing left underneath to copy, search, or extract.
Open Blackpdf's Redact PDF tool and drop your file in.
Mark what needs to go. Three ways:
- Manual marking — drag a box over any area.
- Find & Redact — search for a specific word, name, or number and redact every occurrence at once.
- Auto Detect — scan the document for emails, phone numbers, SSNs, credit card numbers, dates, and IP addresses.
Pick a redaction colour (black is conventional; white can be used to make content simply vanish).
Apply the redactions and download. The data is gone from the output file.
Common questions
How do I check whether my redaction actually worked?
Two quick tests on the output file. First, try to select and copy the area you redacted — you should get nothing. Second, search (Ctrl+F) for the text you removed — it shouldn't be found. If either test surfaces the content, it wasn't redacted, it was covered.
I already sent a document with black boxes. What now?
Assume the content is exposed. Anyone who received the file can extract it in seconds. Re-issue a properly redacted version and, depending on what was disclosed, treat it as a data incident rather than a formatting error.
Does redaction remove the information from the document's metadata too?
Redaction removes the marked content from the page. Metadata — author name, title, editing history — lives separately, so review it as well if the document is sensitive. Our redaction guide covers this.
What about a scanned document — is a black box safe there?
A scan is an image, so there's no text layer to copy — which makes a black box appear safer. But if the scan has been OCR'd, there is a hidden text layer, and it's fully copyable. Don't rely on "it's just an image"; redact properly.
Why do professional organisations keep getting this wrong?
Because the failure is invisible. The document looks correct on screen, so there's no feedback telling you it's broken — right up until someone copies the text out. That's exactly why it's worth testing your output every time.
Wrap-up
A black rectangle is a drawing, not a deletion. The text under it stays in the file and can be copied, searched, or extracted by anyone who receives the document — no expertise required.
To actually remove sensitive content, use Redact PDF, which deletes the underlying data rather than covering it. Then verify: select the area and try to copy it, and search for the text you removed. If both come up empty, you're safe. For the full walkthrough of all three redaction methods, see our redaction guide.