How does image compression save space?
JPEG and WebP are lossy formats: they divide a photo into small blocks and store each block's fine detail more or less precisely depending on the quality setting. At high quality the encoder keeps almost everything; lower down it rounds away texture you're unlikely to notice, such as noise in a blue sky or grain in a shadow, and the file shrinks dramatically.
Phone cameras save JPEGs at very high quality, often 3–8 MB for a 12-megapixel photo. Re-encoding at around 70% typically cuts that by 60–85% with no difference you'd see on a phone or laptop screen. The width and height don't change; only the way the pixels are stored does.
What quality should I choose?
The slider runs from 30% to 95% and starts at 70%, a good balance for most photos:
- 80–95%: for photos you might print or crop later, or with fine detail such as text in a screenshot or a product's fabric.
- 60–75%: the sweet spot for websites, email, chat and social media. Most people can't tell it from the original.
- 30–50%: for thumbnails, previews, or squeezing many photos under an upload limit. Expect visible softening and blockiness in gradients.
What happens to each format
JPEGs stay JPEG and WebP stays WebP, re-encoded at your chosen quality. PNGs are saved as WebP, which keeps transparency (so logos and cut-out product shots keep see-through backgrounds) and is usually far smaller than the PNG. Other formats, including HEIC photos from iPhones, become JPEG.
If a file is already well compressed and the new version wouldn't be smaller, you get the original back, marked as already optimized, so compressing never makes a file worse or bigger. Each result shows the size before and after and how much you saved.
Compress or resize?
Compression keeps every pixel and stores them more efficiently. Resizing removes pixels. A 4000×3000 photo shown in a 1200-pixel-wide blog column carries ten times more pixels than needed, so for web use, resizing first (with the Image Resizer) and then compressing gives the smallest files of all. If a form demands a strict limit like under 100 KB, combine both: resize to the size it will be shown at, then lower the quality until it fits.
Batches, privacy and metadata
Add as many images as you like and they're compressed together. Download them one by one, or all at once as a ZIP. Everything happens in your browser, using the same image engine that displays pictures on screen, so your photos are never uploaded.
The compressed files carry no EXIF metadata: GPS location, date and camera details are left out, which is a privacy bonus when posting photos online. Phone photos are turned upright first, so nothing comes out sideways.
Hitting a specific size limit
Upload forms often set a hard limit, such as 20–50 KB for an ID photo or 200 KB to 2 MB for a document scan. There's no target-size box here, but two moves get you there reliably. First resize: a form photo doesn't need 4000 pixels, and 600–1000 pixels on the longer side is usually plenty. Then compress, lowering the quality until the shown size is under the limit. Shrinking the dimensions does far more than lowering quality alone, and keeps the image looking clean.
For scanned documents with text, stay around 60–75% quality and reduce the pixel size instead of pushing the quality very low, so small print stays readable.
A note on WebP output
PNGs are compressed to WebP because it keeps transparency at a fraction of the size. Every modern browser, WordPress (since version 5.8), Google Docs and most social networks accept WebP. A few older apps and strict upload forms only list JPG and PNG. If one rejects the WebP, convert the original with PNG to JPG instead (transparent areas become white), then compress that JPG.