The mistake this tool exists to prevent
Open a PDF in almost any editor, draw a black rectangle over a name, and save. It looks redacted. It is not. The rectangle is a graphic painted above the page, and the text is still sitting underneath it, unchanged — selectable, copyable, and returned in full by any text extractor.
This is not a theoretical failure. Governments, law firms, hospitals and newspapers have all published documents redacted exactly this way and had the hidden text pulled out within hours. It keeps happening because the wrong method looks identical to the right one.
How this works instead
Every page is rendered to an image at the resolution you choose. The boxes you drew are painted onto those pixels. Then a new PDF is assembled from the images.
After that, the redacted content is not concealed — it does not exist in the file. There are no text objects to recover, no hidden layer, no earlier revision. Everything you did not cover remains perfectly visible; what you covered is black pixels and nothing else.
Because the document is rebuilt from scratch, the original's metadata does not come with it either. Author, title, producing software and revision history are all left behind — details that have given away as much as visible text has.
What it costs
No text layer. The output cannot be searched or selected. This is unavoidable and it is the point: text that can still be searched is text that is still there. A tool offering redaction with searchable output has not removed anything.
A larger file. Text is compact; images are not. Pick 150 DPI for something to be read on screen, 300 DPI for print.
If you need both a clean redaction and searchable text, redact here and run the result through OCR afterwards — which will read back only what is still visible, as it should.
Redacting, cropping and flattening are three different things
Cropping sets a box telling a viewer to show less of the page. Everything outside it is still in the file and returns the moment someone changes the box.
Flattening bakes form field values into the page. Useful for finalising a form, irrelevant to hiding anything.
Redacting, as done here, removes the content. Only this one is safe to use before publishing a document.
Common uses
- Removing account numbers from a statement before sharing it
- Taking names or addresses out of a document before publishing
- Preparing a contract for disclosure with commercial terms removed
- Sending an ID or record with sensitive fields taken out
Privacy
The document never leaves your device. Uploading a file to have its secrets removed is a contradiction, and this tool does not ask you to.