A law firm redacts a witness’s name before filing a document. A government agency blacks out a source’s identity before releasing a report. A company covers a salary figure before sharing a contract externally. In every one of these real, documented cases, someone opened the “redacted” PDF, selected the black box with their cursor, hit copy, and pasted the supposedly hidden text right out from underneath it. The box was never the redaction — it was a sticky note. The words were still there the whole time.
Why the obvious way to redact is the wrong way
The instinctive approach to hiding something in a PDF is to draw a black rectangle over it, the same way you’d cover a line with a marker on paper. On screen it works — the text disappears from view. But a PDF page isn’t ink on paper; it’s a stack of layers, and a shape drawn “on top” in a PDF editor is usually just another layer sitting above the original text, not a replacement for it. The characters are still there in the file’s structure, still selectable, still present in a text search, still recoverable by anyone who copies the page contents or opens the file in a tool that ignores drawing order. Zoom, extract-text, or even just triple-click-and-copy, and the “hidden” text comes right back.
This isn’t a hypothetical. It’s a well-known category of security incident precisely because the mistake is so easy to make — the box looks like it did the job, so nobody checks. The only way to be sure something is actually gone is to remove it from the underlying content, not paint over it.
What actually has to happen for redaction to be real
Real redaction has one non-negotiable requirement: the pixels or characters you’re hiding cannot exist anywhere in the output file once you’re done — not in the text layer, not in an embedded image, not in a layer you drew over. The two ways to get there are to strip the sensitive content out of the document’s structure entirely, or to flatten the page into a single image with the sensitive area painted over before that flattening happens, so there’s no live text layer left to search under any box.
How Hide Text does it
Hide Text takes the second approach. You upload a PDF, click and drag a box over each part of a page you want gone — a name, an account number, a signature, a photo, a paragraph — and when you run it, here’s what actually happens to any page you marked: the tool renders that whole page to a canvas at high resolution, paints your marked areas solid black directly onto that canvas, and only then converts the canvas into a flattened image that replaces the page. The blackout is baked into the pixels before the page becomes an image, not layered on top of a page that still has live text. There is no text underneath left to select, search, or extract — because there’s no text layer on that page at all anymore, redacted or not.
Pages you didn’t mark are left completely alone — copied through untouched, keeping their original text, quality, and file size. Only pages with at least one box on them get converted, so a 40-page contract with one redacted clause doesn’t turn your whole document into a stack of images.
Everything happens in your browser. The PDF is never uploaded anywhere, which matters — a file you’re specifically trying to keep something out of is not a file you want to hand to a server in order to redact it.
The honest tradeoff
This is destructive by design, and worth being deliberate about: once you redact a page, that page’s text stops being selectable, searchable, and copyable — not just the part you blacked out, but the whole page, because the whole page is now one image. That’s the price of the guarantee. If you need the rest of a page’s text to stay searchable after removing one sensitive line, this isn’t the right tool for that page — you’d need to redact only the specific line some other way, accepting that any solution that keeps the rest of the text alive also has to touch the document’s text layer directly, which is a fundamentally different (and fussier) operation than a click-and-drag box tool. For most redaction — hiding a specific detail before sharing a document more broadly — losing that one page’s searchability is a reasonable trade for actually knowing the information is gone.
Also worth knowing before you start: there’s no built-in undo once you download. Keep your original file if there’s any chance you’ll need an unredacted copy later — the tool only ever works from what you upload, and it doesn’t keep a copy for you.
Redacting a name or number that shows up more than once
Hide Text only redacts what you draw a box over — it doesn’t search the file for you. If the thing you’re hiding appears on more than one page (a client’s name repeated through a 20-page report, an account number on both the cover sheet and page 14), catching every occurrence is on you, one page at a time.
The reliable way to do that: before you open Hide Text, open the same PDF in any viewer with a search box — your browser’s built-in PDF viewer, Acrobat Reader, whatever you’ve already got — and search (Ctrl+F on Windows and Linux, Cmd+F on a Mac) for the exact text you’re about to redact. Note every page number it turns up. Then work through Hide Text page by page and box the match on each one. Skip that step and it’s easy to redact the name on the cover page and miss the same name sitting in a table on page 11 — the tool did exactly what you told it to; it just was never told about page 11.
Redacting a whole page, or removing it instead
Sometimes what needs to disappear isn’t one line, it’s the entire page — an exhibit covered by an NDA, a cover sheet with someone’s full contact details, an internal notes page that shouldn’t have made it into the file you’re sharing. Hide Text doesn’t have a separate “redact this whole page” button, but you don’t need one: draw a single box that stretches corner to corner, and the page gets rendered to an image with that box painted over the whole thing, exactly the same as it would for a box the size of one line. A page turns into a flattened image the moment it has any box on it at all, so covering all of it instead of part of it doesn’t cost you anything extra.
If you’d rather the page vanish from the file entirely instead of turning into a solid black rectangle, that’s a different job — Delete PDF Pages removes a page outright, which is the better choice when a page shouldn’t be part of the document at all, rather than merely unreadable within it.
What redacting does to metadata and other hidden data
A black box that covers text on the page doesn’t touch anything that isn’t on the page. Document metadata (Title, Author, the software that created the file), bookmarks, and file-level attachments all live outside the page content itself — covering a paragraph doesn’t automatically remove any of that.
Hide Text ends up clearing this out anyway, but as a side effect of how it’s built rather than because it specifically hunts for hidden data: because the output file is assembled fresh, page by page — every page, redacted or not, is copied into a brand-new document rather than the original being edited in place — none of that document-level baggage makes the trip. A source PDF’s Author field, its bookmarks, any file attachments: none of it survives, whether or not you redacted anything on the page it was attached to. That’s worth knowing either way. If you were counting on a document’s bookmarks or attached files surviving a run through Hide Text, they won’t. And if you want to check what metadata a file carries, or clear fields on a document you’re not otherwise redacting, Edit PDF Metadata lets you view and overwrite the Title, Author, and similar fields directly.
Where this actually gets used
- Sharing a contract or invoice externally — blacking out a salary figure, bank account number, or internal pricing term before it leaves the company.
- Responding to a records request or FOIA-style disclosure — removing a name, address, or other identifying detail before a document goes public.
- Sending a form or ID scan to a third party — covering a Social Security number, passport number, or date of birth while still sharing the rest of the page.
- Cleaning up a document for a case study or public example — hiding a client’s name or figures in a real file you want to show without exposing who it belongs to.
If you just need to cover something temporarily while presenting a document on screen — a redaction that isn’t meant to survive the file being shared — a simple annotation tool is fine. But the moment that PDF is going to leave your hands, Hide Text is built for the version of “redacted” that actually means gone.
One more thing worth doing before you actually send a redacted file out: redaction only controls what’s visible inside the document, not who can open it in the first place. If the file is heading somewhere sensitive — over email, into a shared drive, to an opposing party in a legal matter — running it through Password Protect a PDF afterward means the parts you didn’t black out are also only reachable by whoever you actually gave the password to.