Give a logo breathing room, round its corners, put it in a circle, drop it into a square — with a solid, transparent or blurred-photo background.
This is the finishing pass for an image: give it breathing room, round the corners, put it in a circle, add a border or a soft shadow, or drop it into a square or social ratio with a solid, transparent or blurred background.
A trimmed logo usually needs padding put back. Trimming makes the file exactly the shape of the artwork, which is right for the file but wrong for how it gets used — placed flush against a heading or an edge it looks cramped. Adding an even percentage margin back gives it the space a designer would have left, and because the margin is a percentage it is identical across a whole set of logos of different shapes.
Fitting into a fixed shape solves the other common problem. Social platforms, marketplaces and app stores want a specific ratio, and if you hand them the wrong one they crop it themselves, usually through the middle of the subject. Choosing a ratio here fits the entire image inside it and fills the leftover space with your background — nothing is cut off. The blurred-photo background is the trick used for portrait photos in a square feed: the gap is filled with an enlarged, blurred copy of the image itself, so the result reads as intentional rather than as a mistake with white bars.
Rounded corners and the circle crop are what turn a square export into an avatar or an app-style tile. Both cut real transparency into the image, so they only survive in a format that supports it — PNG or WebP. Save a rounded image as JPEG and the corners come back as solid blocks, which is why the format note warns you.
Everything is composed on a canvas in this page, so nothing is uploaded and nothing is watermarked — and the preview sits on a chequerboard so you can tell real transparency from white at a glance.
Yes. As well as JPEG, PNG, WebP, GIF, BMP, ICO and AVIF, this tool reads HEIC and HEIF from a phone, TIFF, Photoshop PSD and PSB, TGA, Netpbm, Apple ICNS, and camera RAW from every major brand. Anything the browser cannot open by itself is decoded on your own device before the tool touches it.
The whole finished canvas — background, padding and all. So a logo padded onto a white square comes out as a rounded white card, which is what an app icon or an avatar tile needs. On a photo with no padding the canvas is the photo, so it rounds the photo.
Because it was saved as JPEG, which cannot store transparency — the rounded-off corners have to become some colour, and white is what the encoder picks. Save as PNG or WebP instead and the corners stay genuinely transparent.
No. The whole image is scaled to fit inside the new shape and the leftover space is filled with your chosen background. Nothing is cut off, which is the opposite of what a platform does when it crops to fit on your behalf.
It fills the empty space with an enlarged, heavily blurred copy of the image itself. It is how portrait photos are made to work in a square feed without white bars down the sides, and it keeps the colours of the post consistent with the photo.
Around 5 to 10 percent of the shorter side is a reasonable default, which is close to the clear-space rule most brand guidelines specify. Because it is a percentage, the same setting gives a consistent result across a folder of differently shaped logos.
The shadow is drawn inside the canvas, so it needs padding to land in. Increase the padding until there is room for it — a shadow with a softness of 30 generally needs at least 8 percent padding.
Yes. The border is drawn around the outside edge of the finished canvas, so with padding at zero it sits directly against the artwork. Add padding first and the border moves out to frame the margin as well, which is the matted look.
Yes. Drop them all in, set the padding, shape and background once, and every image is rendered with the same settings. Percent-based sizing means a mixed batch still comes out looking consistent, and the whole set downloads as a ZIP.