Image optimization is a set of choices. Format, quality, dimensions, and source content all affect the result. There is no fixed saving percentage that applies to every image.
Compression and resizing solve different problems
Compression changes how image data is represented. Resizing changes the pixel dimensions. A very wide image may still be heavy after compression because it contains more pixels than your layout needs. PikoZip offers both controls and makes resizing an explicit choice.
Lossy means some detail can change
PikoZip’s JPEG, WebP, and AVIF outputs use adjustable lossy encoding. Lower settings can reduce size but may affect textures, edge detail, or gradients. The comparison view lets you check what changed instead of trusting a percentage alone.
Lossless still has practical limits
Lossless PNG mode avoids palette quantization, but the browser decodes images into its working pixel representation before re-encoding. Color profiles, high bit depth, and metadata may not be preserved exactly as in a specialist archival workflow. Keep an original for archival or color-critical work.
A target is a goal, not a guarantee
A 100 KB target cannot always be reached at a useful size and quality. For supported lossy outputs, PikoZip searches lower quality settings when needed. If it cannot reach the goal, the result says so. You can then choose another format or a smaller width.
Know when to stop
Already optimized files sometimes do not get smaller. When format and dimensions are unchanged, PikoZip keeps the original if the output is larger. An explicit conversion can still increase size when compatibility is more important than byte savings.
Put the guide into practice.
Start with one representative image, inspect the result, then prepare the rest of the batch.
Open the compressor