Two levers, two different effects
To shrink image file size without changing dimensions, use compression rather than resizing. Compression lowers the number of bytes in the file; resizing changes the width and height in pixels.
Image "size" has two meanings that are easily conflated. Resizing changes the pixel dimensions. Compressing changes the file size while keeping the pixel dimensions the same. A 2000×2000 pixel image can be 10 MB or 200 KB depending on the compression. A 500×500 pixel image can be 50 KB or 500 KB depending on the same. The two meanings:
- File size: the number of bytes the file takes on disk. This is what upload limits, email attachment limits, and storage quotas measure.
- Pixel dimensions: the width and height of the image in pixels. This determines how large the image can be displayed at full resolution without appearing blurry.
When you need to compress, not resize
You need compression (not resizing) when:
- A form specifies a maximum file size and the correct pixel dimensions are already set
- An image needs to remain at its current dimensions for a specific layout or display requirement
- You want to reduce storage or transfer costs without changing how the image appears at its display size
- The image will be further processed by a system that expects specific pixel dimensions
How compression reduces file size without changing dimensions
Lossy compression works by discarding image information that the human eye is unlikely to perceive. For JPEG and WebP, this primarily targets high-frequency color detail in complex regions — the subtle variations that appear in textured surfaces, foliage, hair, and out-of-focus backgrounds. At quality 80, a JPEG might be 60–70% smaller than at quality 100 while appearing identical at normal viewing sizes and at the same pixel dimensions.
Lossless compression (used in PNG and optionally in WebP) finds mathematical redundancies in the data — runs of identical pixels, predictable patterns — and represents them more compactly without discarding any pixel values. PNG files of photographs are hard to compress significantly with lossless methods because photographs have few redundant patterns. PNG files of screenshots and graphics with flat color areas compress well.
Choosing a quality setting
For JPEG and lossy WebP, the quality slider is the primary control. The right setting depends on how closely the image will be examined. Always compare the compressed output at the size it will be displayed, not at full pixel zoom: artifacts that are visible when examining individual pixels at 100% zoom are typically invisible when the image is viewed at its actual display size.
| Image type | Quality |
|---|---|
| Profile and document photos (viewed close) | 82-88 |
| Web content images | 75-82 |
| Thumbnails and small previews | 65-75 |
| Background or decorative images | 60-70 |
Format conversion as a compression technique
Switching formats can significantly reduce file size without changing pixel dimensions. A PNG photograph is losslessly compressed but still much larger than a JPEG of the same image because lossless compression cannot achieve the compression ratios of lossy encoding. Converting a PNG photograph to JPEG or WebP at quality 82 can reduce file size by 80–90% without any change to the pixel dimensions.
Converting JPEG to WebP at equivalent quality also reduces file size by 25–35% without any change to dimensions. If a JPEG needs to meet a file size limit, converting to WebP first may allow you to meet the limit at a higher quality setting.
When resizing is also necessary
If the image has far more pixels than necessary for its intended use, compression alone may not bring the file size down to the required limit without unacceptable quality loss. In that case, resizing first — reducing the pixel dimensions — and then compressing is more effective and preserves better quality.
A 4000-pixel-wide photograph at JPEG quality 90 might be 5 MB. Resizing to 1000 pixels wide first, then compressing at quality 82, might produce a 150 KB file that looks better than the original at quality 30, because the quality degradation is avoided.
Rule of thumb: if the file is much larger than the limit and the image is much larger than needed, resize first. If the dimensions are already appropriate but the file is too big, compress first and only resize if compression alone cannot meet the limit.
How to compress without changing dimensions in PhotoTools
Open the compress tool in PhotoTools and drop in your image. The quality slider reduces file size while keeping the same pixel dimensions. The output format selector lets you switch to WebP for additional file size reduction. The file size estimate updates as you move the slider so you can see when you have hit a specific limit.
Do not use the resize tool if your goal is only to reduce file size. The resize tool changes pixel dimensions. Use compress for file size only.