"Make this image smaller" means two completely different things, and using the wrong one is why people end up compressing a photo into mush and still find it is too big to upload.
Two different levers
Resizing changes the dimensions — how many pixels wide and tall the image is. A 4000 × 3000 photo resized to 1000 × 750 has a quarter as many pixels in each direction and one sixteenth of the total.
Compressing keeps the dimensions and stores the same pixels more efficiently, discarding detail the eye is unlikely to notice. The image still fills the same space on screen; the file is just smaller.
Why resizing wins
File size tracks the total pixel count, and pixel count grows with the square of the dimensions. Halving the width and height leaves a quarter of the pixels — typically around a quarter of the file size, with no visible quality loss at all provided you did not need those dimensions.
And most of the time, you did not need them. A photo from your phone is about 4000 pixels wide. If you put it in a blog post, it is usually shown at around 800 pixels wide — five times smaller. The browser still downloads the whole 4000-pixel photo, then shrinks it on screen. Almost all of that download was wasted.
Resizing to 800 pixels yourself gets rid of that waste before anyone downloads it. The picture looks exactly the same to your visitor, because it was being displayed at 800 pixels either way — the file is just far smaller and loads far faster.
Target sizes that work
| Destination | Width | Rough file size |
|---|---|---|
| Email attachment | 1200–2000 px | under 500 KB |
| Image in a blog post | 1200–1600 px | 150–400 KB |
| Full-width web banner | 1920–2400 px | 300–800 KB |
| Thumbnail | 300–400 px | under 50 KB |
| Profile picture | 400–800 px square | under 100 KB |
A note on modern screens: phones and laptops often pack roughly twice as many physical pixels into the same space. If you want an image to stay crisp on those, supply it at about double the width it will be displayed at — 1600 px for an 800 px column. Beyond that, you are shipping data nobody sees.
Choosing a compression level
For JPG, quality is a number from 1 to 100. The useful range is narrower than people expect:
- Below 60 – blocky squares and smeared halos become visible
- 75–85 – the sweet spot for photographs; large savings, no visible difference
- Above 90 – the file grows quickly for almost no perceptible gain
Always check the result at full size rather than trusting the number. Photos with large areas of smooth gradient — skies especially — show compression artefacts sooner than busy, detailed scenes.
Format matters too
If your image is a screenshot, a diagram, or anything containing readable text, JPG is the wrong format — its compression makes assumptions that fall apart at hard edges, so lettering comes out fuzzy. Use PNG, which keeps text crisp and often produces a smaller file for flat graphics anyway.
For photographs going on a website you control, WebP is roughly 25–35% smaller than the equivalent JPG at the same visual quality. For photographs you are sending to another person, stick with JPG for compatibility.
A workflow that works every time
- Keep the original. Everything below is one-way; archive the full-size file first.
- Resize to the dimensions the destination actually needs. This is the big win.
- Pick the right format — JPG or WebP for photos, PNG for anything with text or transparency.
- Compress at around 80% quality.
- Look at the result at full size before you send or publish it.
Common questions
My image is still too big after compressing. What now?
You almost certainly skipped the resize step. Compression alone cannot rescue a 4000 px photo that needs to be under 200 KB — push it that far and it will look terrible. Reduce the dimensions first and the compression barely has to work.
Does resizing lose quality?
Making an image smaller discards pixels you were not displaying, so at the size it will actually be viewed there is no visible loss. Making one bigger is the lossy direction — the software has to invent detail that was never captured.
Why is my PNG so much bigger than a JPG?
Because it is probably a photograph. PNG is lossless, and lossless compression cannot do much with millions of subtly different colours. PNG wins on screenshots and flat graphics; JPG and WebP win on photographs.
Can I compress the same image twice?
You can, but do not. JPG damage accumulates — each save discards a little more, working partly from the artefacts the last pass introduced. Go back to your original and compress it once at the right setting.