converters · · Updated
How to Compress Images Without Visible Quality Loss
By Chirag
"Compress this image" usually means "make the file much smaller without anyone noticing". There are two very different ways to do that. One is resizing the image to the size it is actually shown at; the other is lowering the quality setting. Do them in that order.
Step 1: stop shipping pixels nobody sees
A 12-megapixel phone photo is 4000 × 3000 pixels. If it appears on a web page in a column 600 pixels wide, a 1200-pixel-wide version already looks sharp, even on high-density "retina" screens that use two device pixels per CSS pixel. The difference in pixel count:
- Original: 4000 × 3000 = 12,000,000 pixels
- Resized: 1200 × 900 = 1,080,000 pixels
- That's about 11 times fewer pixels to store.
At the same quality setting, file size roughly tracks the pixel count, so this one step often does more than any quality slider could. The Image Resizer takes exact pixel dimensions or a percentage, and its aspect-ratio lock keeps the photo from being stretched.
Work out the target size from where the image will be used:
- A content column or card on a website: twice its displayed width, to cover high-density screens.
- An email attachment someone will just look at: around 1600–2000 pixels on the long side is plenty for a screen.
- Print: keep the original. Print needs far more pixels than a screen.
Step 2: pick the quality setting by comparison, not by rule
For JPG and WebP, the quality setting trades detail for size. There isn't one right number, because it depends on the picture:
- Busy textures (foliage, gravel, fabric) hide compression artifacts well and tolerate lower settings.
- Smooth areas (sky, skin, studio backgrounds) show blocky "banding" first.
- Text and sharp edges show halos first. If an image is mostly text, it probably shouldn't be a JPG at all.
The Image Compressor starts at 70% for JPG and shows the original size, the new size, and the percentage saved before you download. A practical method: compress at 70%, look at the result at 100% zoom around edges and smooth areas, and move the slider down until you can see a difference, then step back up once. That lands on the lowest setting that looks right for that particular image.
Choosing the output format while compressing
- WebP usually gives the smallest file at similar visual quality, and every current browser displays it. Choose it for your own website.
- JPG when the file goes somewhere that might not accept WebP: email, older software, some upload forms.
- PNG is lossless. Choosing it re-encodes the image without discarding anything, which helps with some unoptimized screenshots. But the output can be larger than a PNG that was already optimized by another tool. When that happens, the compressor says so ("12% larger" instead of "12% smaller") rather than hiding it. For photos, PNG is the wrong choice.
Don't compress the same JPG repeatedly
Every JPG save throws away detail again, even at the same quality setting. Edit, crop, compress, then re-edit and compress again, and the artifacts compound ("generation loss"). Keep the original, and always compress from it rather than from a previously compressed copy.
What happens to the photo's hidden data
Photos from phones and cameras carry EXIF metadata: camera model, exposure settings, the date, and often the GPS coordinates of where the photo was taken. A photo of an item you're selling, taken at home, can reveal your home's location.
Because the compressor re-draws the image and encodes a fresh file, EXIF data (GPS included) isn't copied into the compressed output. The browser applies the photo's rotation first, so the result stays upright. If you want to strip the metadata without changing size or quality, the EXIF Data Remover does only that. It keeps PNG and WebP in their own format, and writes other photos as a high-quality (95%) JPG. Either way, the file you share no longer carries a location.
A worked routine for web images
- Start from the original photo.
- Resize to twice the display width, for example 1200 px for a 600 px column.
- Compress to WebP, starting at 70%, and adjust by eye.
- Check the before and after sizes the compressor shows, and open the result at 100% zoom before it replaces the original anywhere.
Everything runs in the browser, so the original photo is never uploaded. That matters most for exactly the photos whose location data you're trying to protect. For how the formats differ in the first place, see PNG vs JPG vs WebP vs BMP vs GIF.
Related guides
- PNG vs JPG vs WebP vs BMP vs GIF — Which Format to Use · How the five common image formats differ in compression, transparency and colour, with real file-size arithmetic, plus what changes when an image is converted in the browser.
- How Accurate Is Browser OCR? 14 Test Images, Measured · The same invoice line rendered 14 ways (small text, blur, tilt, low contrast, a busy background, JPEG artifacts, Hindi, two columns) and run through this site's OCR, with the accuracy of each and the three fixes that took the average from 88.5% to 99.5%.
- Converting CSV and XML to Excel Without Losing Data · Why double-clicking a CSV mangles PIN codes, long IDs and ₹ signs, what a careful CSV-to-Excel conversion does instead, and how XML records map to spreadsheet columns.