The failure everyone keeps repeating
Draw a black rectangle over a paragraph in an ordinary PDF editor and the text is still in the file — selectable straight through the box with Ctrl+A, extractable by any parser, visible to anyone who copies the page into a text editor. This exact mistake has exposed redacted names and figures in legal filings and official releases, repeatedly, because the documents looked redacted. The lesson is simple: redaction that only covers content isn’t redaction. The content itself has to be removed from the file. If a redaction can be undone with select-all, it was never a redaction — just a decoration.
Step by step: burn the boxes in
- Open the Redact PDF tool on SafeFileConvert and drop the document onto the dropzone. Rendered previews of the pages appear.
- Click and drag boxes over everything private — the account number, the salary line, the name. Boxes are drawn per page, so information that repeats needs a box wherever it appears; the previews make that a quick scroll-and-drag pass.
- Apply the redaction. Every page carrying a box is re-rendered to pixels, the boxes are painted onto those pixels, and the page is rebuilt as an image — the covered text is never carried into the output file at all. Pages without boxes are copied untouched.
- Download the redacted copy and check it: try selecting where the boxes are. There’s nothing there to select.
The trade-off is the proof
Redacted pages become images, so text on those specific pages is no longer selectable or searchable — that’s the price of certainty, and for documents that need redaction it’s the right one. Untouched pages keep their original quality and text layer. If selectability across the whole document matters more than guaranteed removal, redaction is the wrong operation to reach for in the first place. Extract the redacted copy’s text or print it if you want further proof: the removed content isn’t there to find.
Two finishing moves
First, remember metadata: a PDF’s hidden fields can carry the author’s name and edit dates, and the Remove Metadata from PDF tool on this site cleans those separately, also locally. Second, appreciate what didn’t happen here — uploading a document to a stranger’s server in order to remove its sensitive parts would be absurd, and this tool never does: the previews are rendered and the boxes burned in by code on your own machine, with no copy of the document, redacted or not, transmitted anywhere.