Add Watermark to Image

Place it by dragging, not by guessing

The watermarked image renders as soon as you add a file and re-renders every time you change a setting, so you are looking at the actual result rather than a description of it. Then drag the watermark to wherever it should sit.

That is more useful than it sounds, because the right place depends entirely on the photograph. A corner works on a landscape with empty sky. On a portrait or a product shot, the corner is exactly where a crop will remove it — so the watermark needs to be somewhere it cannot be cut away without ruining the image, which is a judgement you can only make while looking at the picture.

Nine presets are there for speed, with bottom-right as the default, and the opacity, size, colour and rotation all update the preview live. Rotation starts at −30°, which is the angle that reads as a watermark rather than as a caption.

Tiled, and why it is the only version that really works

A single watermark in a corner is a courtesy. Anyone who wants your image without it crops the corner off, and it takes about two seconds.

Tiled repeats the watermark in a grid across the entire picture. It cannot be cropped away, and removing it means editing over your subject rather than trimming an edge. It is uglier — that is the trade — so it suits proofs, previews, drafts sent to clients, and anything posted publicly that you genuinely do not want reused. Keep the opacity low, around 20–30%, and tiling stays readable as an image while still being unusable as a clean copy.

For a portfolio piece you want people to admire, one watermark at 40–50% opacity placed over a low-detail part of the subject is the usual compromise.

Text or your own logo

Text is the quick option — a name, a handle, a © line, in one of five fonts at any size from 8 to 200 pixels, in any colour. White at around 50% opacity is the standard because it reads on both light and dark areas.

The image option takes your own logo or signature file instead, scaled from 5% to 80% of the photo's width. That is the one to use if you have brand marks, because a logo is recognised instantly where a name has to be read.

A watermark is a deterrent, not protection

Worth saying plainly, as with documents. A watermark makes casual reuse awkward and traceable, and it advertises who made the image. It does not prevent anyone determined from removing it — that is what editing software is for.

So watermark to discourage and to attribute, and if an image genuinely must not circulate, the answer is not to publish it at full resolution. A small, watermarked preview and a full-resolution file sent only to the buyer is how photographers have always handled this, and it still works.

Several photos, and what the output is

Add multiple files and the same watermark settings are applied to all of them, returned as a ZIP. Since watermarking is almost always something you do to a batch of images rather than to one, that is the normal case.

The output is a JPEG, which is right for photographs but means transparency is not preserved — a PNG with a transparent background will come back with that background filled in. This is a tool for photos, not for graphics that need to stay transparent.

Related tools

Add Watermark to PDF does the same job on PDF documents, where the options are text-only but include multiple independent watermarks. Resize Image is worth running first when you are publishing previews, and Compress Image reduces the finished files for upload.