How to redact a PDF properly
Black boxes drawn over text hide nothing. Here is why that keeps leaking documents, and how to remove the words for good.
· 6 minute read
The mistake almost everyone makes
You open the PDF, draw a black rectangle over the name or the account number, and save. On screen it looks finished. The problem is that a PDF keeps its text and its drawings as separate objects. The rectangle sits on top; the words are still underneath, exactly as they were.
Anyone can get them back. Select the area and copy it. Open the file in a different reader. Run any text extractor over it. The black box moves or the text comes out regardless, because nothing was ever deleted.
This is not a rare mistake
Court filings, government reports and company documents have all been published with "redactions" that could be copied straight out. The people doing it were not careless — the software simply let them believe the job was done.
What real redaction means
Redaction is finished only when the words are no longer in the file. There are two honest ways to get there.
- Delete the text objects themselves, and save a file that never had them.
- Flatten the page to an image with the boxes painted into the pixels, so there is no text layer left to recover.
The second is what SlimFile.app does. A page you mark is rendered to an image with your black boxes burned in, and that image replaces the page. There is nothing underneath to select, copy or extract.
How to do it, step by step
- Open Redact PDF and add your file. Password-protected PDFs are fine — you will be asked for the password once.
- Click Redact to see the pages, then drag a box over everything that has to disappear. Mark as many areas and pages as you need.
- Apply, and download the finished PDF.
- Check it: try to select the blacked-out area and copy it. Nothing should come out.
The cost, stated honestly
A flattened page becomes a picture, so it is no longer selectable or searchable, and the file may get bigger. That is the real price of the text being gone. Pages you do not mark are copied across untouched and keep their proper text.
If the result needs to be searchable again, run it through OCR afterwards. The recovered text will be only what is still visible on the page.
When to be extra careful
- Anything going to a court, a regulator or a journalist. Assume it will be examined properly.
- Scanned documents, where the text may sit in an OCR layer you cannot see.
- Files where partial words are visible around the edge of a box — widen the box rather than trusting it.
For the most sensitive material, the safest approach is to remove the page altogether rather than redact part of it.