Image compression reduces the amount of data needed to store or deliver an image. The challenge is choosing a smaller file without making important details look damaged. A photograph, a screenshot, a logo and a transparent graphic can need very different settings. The best result is not always the smallest file; it is the smallest file that still looks and works the way the destination requires.
What image compression changes
Compression can remove redundant information from an image and reduce its file size. Lossy compression can discard some visual information to achieve a smaller file, while lossless compression keeps the image data recoverable from the compressed representation. The useful choice depends on the image and the purpose.
Choose the right starting file
Keep the original image as your source. Avoid repeatedly opening and re-saving a JPEG just to make it smaller because each lossy export can introduce more artifacts. Create the compressed version from the cleanest original you have.
How to compress an image with Tervilo
- Open the Tervilo Image Compressor.
- Select the original image.
- Choose an appropriate compression or quality setting if the tool provides one.
- Process the image.
- Compare the output visually with the source.
- Check the resulting file size before using it.
Do not compress before deciding whether the image is already too large
File size is affected by more than compression quality. A 4,000 × 3,000 pixel photograph may be much larger than necessary for a small web card even before compression is applied. If the destination only needs a smaller display size, resizing first can produce a larger improvement than pushing compression to an aggressive setting.
JPG, PNG and WebP behave differently
JPEG is often efficient for photographs. PNG is useful for graphics, screenshots and transparency, and can become comparatively large for photographic images. WebP supports both lossy and lossless encoding and transparency and can be efficient for web delivery. There is no universal “best” format; choose based on image content, transparency, destination compatibility and file-size goals.
How much quality should you sacrifice?
Use the lowest compression that meets the real requirement. Examine sharp edges, small text, skin tones, gradients and fine texture. Compression artifacts can be especially visible around text and high-contrast edges. If the image is used for a product listing or document scan, accuracy can matter more than saving the last few kilobytes.
Example: a website photo
Suppose an original photograph is 4 MB and 4,000 pixels wide, but the website displays it at about 1,200 pixels wide. Reducing the dimensions first and then applying moderate compression can often produce a much more useful file than aggressively compressing the 4,000-pixel original.
Example: a screenshot with text
A screenshot contains sharp text and interface edges. Very aggressive lossy compression can make letters look soft or create ringing around icons. A lossless format or a moderate-quality output may be more appropriate when legibility matters.
Compression is not the same as resizing
Compression changes how image data is encoded. Resizing changes pixel dimensions. You can do both, but they solve different problems. If a destination requires 800 × 800 pixels, simply compressing a 3,000 × 3,000 image does not satisfy the dimension requirement.
How to verify the result
After compression, check both file size and visual quality. Open the output at normal viewing size and inspect important details at a closer zoom. For a website, also test the image in its actual page layout because quality that looks acceptable on a desktop may appear differently on a small screen.
Common mistakes
- Trying to reach the smallest possible file size regardless of quality.
- Compressing an already compressed JPEG repeatedly.
- Ignoring image dimensions.
- Converting a transparent graphic to a format that cannot retain transparency.
- Replacing the original instead of keeping a clean source copy.
- Judging quality only from the file size number.
When to stop compressing
Stop when the file meets the actual upload, storage or delivery requirement and the visible quality remains appropriate. A further small reduction in size is not necessarily worth a noticeable loss of detail.
Quick checklist
- □ I kept the original image.
- □ I checked whether resizing is also needed.
- □ I selected a suitable format.
- □ I compared quality after compression.
- □ I checked the final file size.
- □ I verified the image in its real destination.
Quick answer
Compress images by starting from the original, choosing a suitable format and applying only as much compression as the destination allows. Check dimensions separately, compare visual quality with the source and keep the original so you can create another version later.