← Back to Blog

May 28, 2026

Large image files cause real problems: slow-loading web pages, attachments that bounce back, storage that fills up, and uploads that crawl. The good news is that most images can be made dramatically smaller with little or no visible loss in quality. Here is how image compression actually works and how to do it well.

Lossy vs lossless compression

There are two kinds of compression, and understanding the difference is the key to keeping quality.

Lossless compression shrinks a file without throwing away any image data. The image is reconstructed perfectly, pixel for pixel. PNG uses lossless compression. The downside is that there is a limit to how much you can shrink a file this way.

Lossy compression achieves much smaller files by discarding data that the human eye is unlikely to notice. JPG and lossy WebP work this way. Used carefully, the quality loss is invisible. Used too aggressively, you get blocky artifacts and fuzziness.

Choose the right format first

The single biggest factor in file size is choosing the right format for the content. A photograph saved as PNG might be 5MB, but the same photo as a high-quality JPG could be under 1MB with no visible difference. Meanwhile a screenshot saved as JPG looks terrible, but as PNG it stays crisp and reasonably small.

As a rule: photos should be JPG or WebP, and screenshots, logos, and graphics should be PNG or lossless WebP. WebP almost always produces the smallest file at a given quality, so if your destination supports it, it is the best choice.

Resize before you compress

People often overlook the simplest trick: if an image is larger than it needs to be, shrink its dimensions. A photo straight from a phone camera might be 4000 pixels wide, but if it will only ever be displayed at 800 pixels wide on a web page, you are storing four times more detail than anyone will ever see. Resizing to the actual display size can cut file size by 80 percent or more on its own.

Tune the quality setting

When you export a JPG or WebP, most tools let you choose a quality level from 0 to 100. The sweet spot for photos is usually between 75 and 85. At that range, the file is much smaller than maximum quality, but the difference is invisible to the naked eye. Going above 90 adds a lot of file size for almost no visible benefit. Going below 60 starts to show artifacts.

Practical steps to compress an image

1. Decide where the image will be used and how large it needs to display.
2. Resize the dimensions to match that display size, with a little headroom for high-resolution screens.
3. Choose the right format: JPG or WebP for photos, PNG or WebP for graphics.
4. Export at a quality of about 80 for photos.
5. Check the result. If it looks good, you are done. If the file is still too big, drop the quality slightly or reduce dimensions a bit more.

A note on re-compressing

Avoid compressing an already-compressed JPG over and over. Each time a lossy image is re-saved, more data is discarded and quality degrades — a problem known as generation loss. Always compress from the highest-quality original you have, not from a copy that has already been squeezed.

Hosting your compressed images

Once your image is optimized, you need somewhere to host it. IMGBS hosts your uploads at full quality with no additional forced compression — what you upload is exactly what gets served. That means the careful optimization you did is preserved, and you get a fast direct link to share or embed. IMGBS supports JPG, PNG, WebP, and GIF, so you can upload in whichever format gives you the best size-to-quality balance.