If you've ever redacted a PDF by opening it in a basic editor, drawing a black rectangle over a sensitive line, and saving it — you may not have actually removed anything. This is one of the most common, well-documented redaction failures out there, and it has caused real embarrassment for governments, law firms, and companies who assumed a black box meant the text underneath was gone. It usually wasn't.
The Black Box Problem
A PDF is not a flat image by default — it's a layered document. The visible page is one layer, but the actual text is stored underneath as selectable, searchable, extractable data, independent of whatever graphics are drawn on top of it. When you draw a black rectangle over a Social Security number or an account number using a basic PDF editor or annotation tool, you're typically just adding a new graphic object on top of the existing page. The rectangle covers the text visually, but the text itself is still sitting there in the document's data.
Anyone who opens that "redacted" PDF can often select the text under the black box with a normal text-selection drag, or copy the whole page's text with Ctrl+A / Cmd+A and paste it into a text editor — and the "hidden" content comes right out, box and all. This isn't a theoretical risk; it's a well-known failure mode that has led to real data leaks when redacted court filings, government documents, and legal disclosures were shared before anyone checked whether the underlying text was actually gone.
Why This Keeps Happening
It happens because drawing a box is the intuitive, visually-obvious way to "hide" something, and most general-purpose PDF editors don't make the distinction between visual redaction and actual redaction clear. The box looks right on screen. It prints looking right. It's only when someone selects the text, copies it, or runs the file through an automated text-extraction tool that the mistake becomes obvious — often after the document has already been shared.
What Real Redaction Requires
True redaction means the underlying text data no longer exists in the document at all — not hidden, not covered, gone. The only reliable way to guarantee that is to flatten the page: convert it from a text-plus-graphics layer into a single rasterized image, with the redaction box baked permanently into the pixels before the text layer is discarded. Once a page is an image, there's no text layer left to select, copy, or extract — the sensitive content was destroyed at the same time it was flattened, not just covered afterward.
How to Redact a PDF Properly
- Open our free PDF Redactor tool.
- Upload your PDF — it's processed entirely in your browser, nothing is sent to a server.
- Draw a box over each area you want to redact — an SSN, an account number, a name, a signature.
- Once you're happy with the placement, run the redaction. Each page is rasterized into a flattened image with your redaction boxes permanently baked in.
- Download the redacted PDF and verify it (see the FAQ below for how).
The Tradeoff You're Accepting
Because true redaction flattens each page into an image, the output PDF is no longer searchable or selectable text — you can't highlight a sentence, run Ctrl+F to search it, or copy a paragraph out of it anymore. This is expected, not a bug: it's the direct consequence of actually removing the underlying data rather than just hiding it. For a document you're redacting specifically because it contains something sensitive, this is a reasonable and necessary trade. If you also need a searchable, non-redacted version for your own records, keep your original file separately — the redaction process is one-directional, the same way compression or any other destructive edit is.
FAQ
Can I test whether my redaction actually worked? Yes — after downloading the redacted PDF, open it in a PDF viewer and try to select or copy text from the area you redacted. If the redaction worked, there will be nothing to select there, because that whole page is now part of a flattened image with no text layer underneath. If you can still highlight text, something went wrong.
Why can't I search the redacted PDF anymore? Because true redaction removes the underlying text layer entirely, not just the visible characters in the redacted area — the whole page becomes an image. This is the tradeoff for guaranteeing the sensitive text is actually gone, not just hidden.
Is a black box added in Preview, Adobe's basic markup tools, or PowerPoint safe to use for redaction? No, generally not, unless the tool specifically states it flattens or rasterizes the page. Most basic annotation tools only draw a graphic on top of the existing text layer, which can still be selected and extracted from underneath.
Is my file uploaded anywhere during redaction? No. Everything happens locally in your browser — your PDF is never sent to a server, which matters given that the whole point is keeping the content private.