How JPG compression works
JPG (identical to JPEG — same format, two spellings of one file extension) is a lossy format: it saves space by discarding image detail your eye is least likely to notice. The encoder splits the image into 8×8 pixel blocks, converts them to frequency data, and rounds away the subtle high-frequency information. The lower the quality setting, the more aggressively it rounds.
That is why re-exporting a JPG at quality 80–85 typically cuts the file size by 60–80% while looking identical at normal viewing size. Cameras and phones deliberately save at very high quality (92–100), which is why photos straight off a device are 3–12 MB when 300–800 KB would look the same on screen.
What quality setting should you use?
| Quality | Typical savings | Best for |
|---|---|---|
| 90–100 | 10–40% | Print, archiving, further editing |
| 80–85 | 60–80% | Web, email, portfolios — the sweet spot |
| 65–75 | 75–88% | Thumbnails, large galleries, slow connections |
| Below 60 | 85%+ | Only when a hard size limit forces it — artifacts become visible |
Our default is 82. Combined with resizing to 1920 px (plenty for any screen), a 4 MB phone photo usually lands between 250 and 500 KB.
When JPG is the right choice — and when it isn’t
JPG is ideal for photographs and complex images: gradients, skin tones, landscapes. It is the wrong choice for screenshots with text, logos, or anything needing transparency — JPG has no alpha channel and smears sharp edges. For those, use PNG or better, WebP.
If the images are going on a website, consider exporting to WebP instead: it is 25–35% smaller than JPG at the same visual quality and supported by every modern browser. See WebP vs JPG for a full comparison.