How to Shrink Images Without Visible Quality Loss

File size comes from three things: how many pixels the image has, how those pixels are encoded and how much extra data rides along. Of the three, pixel count is usually the biggest lever and the one most people skip.

A modern phone photo is around 4032 × 3024 pixels, about 12.2 million. If it will appear on a web page at 800 pixels wide, nobody will ever see most of those pixels. A 1600 × 1200 version (enough for sharp display on a high-density screen) has under 2 million pixels, roughly one sixth as many, before you touch any compression settings.

Decide where the image will be shown and size it for that. A few rules of thumb:

• Web content column: 800 to 1200 pixels wide is typically enough. For sharp display on high-density screens, use about twice the displayed width.
• Full-width hero image: about 1600 to 2000 pixels wide.
• Social posts and thumbnails: match the platform's recommended size, which is often smaller than people expect.
• Email: 1000 to 1600 pixels on the long side is plenty for viewing.

Our image resizer keeps the aspect ratio so nothing is stretched. Reduce the dimensions only; do not enlarge a small image, because upscaling makes it blurry without adding detail. Use the cropper if parts of the frame are not needed, since fewer pixels also means a smaller file.

Format choice affects size as much as quality settings do. Photos belong in JPG or WebP; flat graphics and screenshots often do better as PNG. A screenshot saved as JPG may look fuzzy around text and may not even be smaller. Our JPG vs PNG vs WebP guide covers the choice in detail. For photos on the web, WebP is commonly around a quarter to a third smaller than JPG at comparable quality.

For JPG and lossy WebP, the quality slider controls how much detail is discarded. The relationship is not linear: dropping from 100 to 85 shrinks the file a lot while changing the look very little, but below about 60 artifacts become more visible, especially around edges and in smooth skies.

A practical approach:

1. Start around 80 to 85 for photographs.
2. Compare the result with the original at 100% zoom, looking at edges, text and gradients, not just the whole picture.
3. Lower the quality in steps until you first notice a difference, then go back up one step.

The image compressor shows before and after sizes as you adjust the slider, so you can see the saving for each setting. The right value depends on the image: busy textures hide artifacts well, while smooth gradients show them quickly. Photos for print or heavy editing need higher quality than photos for a web page.

Every time you open a JPG, edit it and save it as JPG again, a new round of lossy compression is applied on top of the old one. Quality drops a little each time, a problem known as generation loss. Avoid it with a simple habit: keep your original file untouched, and always make exports from it, doing resizing and compression once at the end.

If you need several versions (say a large and a thumbnail), create each directly from the original rather than from one another.

Photos also carry metadata, such as camera settings, timestamps and sometimes location. It is usually small compared with the image data, but it can include an embedded thumbnail and, for edited files, extra profiles that add up. Browser-based tools that redraw the image and save it as a new file generally leave this metadata behind, which is also good for privacy. Our guide to EXIF data explains what is stored and why you might want to remove it.

1. Keep the original safe.
2. Crop away what you do not need.
3. Resize to the display size (about double for high-density screens).
4. Choose the format: WebP or JPG for photos, PNG for graphics.
5. Compress with quality near 80 to 85, then check the result at full size.
6. Compare the file sizes, and keep the smallest version that still looks right.

All of these steps run on your device, and nothing is uploaded.