Watermark PDF

Add a text watermark across a PDF (tiled over every page or once in the middle) with the angle, size, colour and opacity you choose. DRAFT, CONFIDENTIAL, COPY and SAMPLE are one click; anything else is a text box.

A watermark answers a question the document cannot answer for itself: what status is this copy. A contract without one looks identical whether it is the final signed version or the third draft, and a document that circulates without saying which it is eventually gets treated as the wrong one. That is the entire job, and it is worth doing well.

What a watermark is not is protection. It does not encrypt anything, it does not stop the file being opened, copied, printed or forwarded, and because it is drawn into the page like any other content it can be taken off again by anyone with a PDF editor. It sits next to “protect PDF” in every tool menu on the web and does something entirely different.

The file is opened, marked and saved in this browser. That matters here more than on most tools, because the documents people watermark are by definition the ones not yet meant to be public: drafts, confidential copies, and work sent to a client for review.

A watermark is a status label that travels with the page. Paper documents solved this with a rubber stamp, and the reason the digital version copies the look (diagonal, large, faint) is that the convention is already understood. Nobody has to be told what a page reading DRAFT across the middle means, which is most of the value.

The status it carries is usually one of three things. That the document is not final, so nothing in it should be relied on or quoted. That it is confidential, so the reader knows the expectation before they forward it. Or whose copy it is, which is why review copies sometimes carry a client's name, not because the name stops a leak, but because it makes an anonymous one harder.

None of those are enforcement. This is the distinction that matters most and the one tool menus obscure by listing “watermark” beside “protect”. A watermark is a message to a reader who is looking at the page. It has no effect at all on someone who is not reading it in good faith: they can select the text under it, print it, screenshot it, or open the file in an editor and delete the mark. If the requirement is that the wrong person cannot open the document, that is encryption, and it is a different thing entirely.

It is also removable in the ordinary case, not just the adversarial one. Because the mark is drawn into the page content rather than held as a separate removable layer, taking it off means editing the page, which any competent PDF editor will do. That cuts both ways: it is why nobody should treat a watermark as tamper-evidence, and it is why you should keep the unmarked original rather than expecting to recover it later.

Two settings decide whether it works in practice, and both are usually got wrong in the same direction. Opacity that looks right on a screen frequently vanishes on paper: under about 10% often disappears through a printer or a photocopier entirely, which defeats the purpose on exactly the copies most likely to go astray. And a single small mark in a corner is trivially cropped out, whereas one tiled across the page cannot be removed without removing content with it.

The last thing worth knowing is what a watermark does to the text underneath. Drawn on top at a low opacity it does not damage anything: the text stays selectable, searchable and accessible to a screen reader, because nothing has been rasterised or covered over destructively. A page image with a watermark burned into it loses all of that, which is one reason to mark the PDF rather than export it to an image and back.