What is lossy compression for images?

PERFORMANCE

Lossy compression reduces image file size by permanently discarding image data that is less perceptible to the human eye. The result is a smaller file at the cost of some image detail. JPEG uses lossy compression; WebP and AVIF support both lossy and lossless modes.

How lossy compression works

Lossy algorithms identify areas of the image where detail can be reduced without a visible quality drop. In JPEG, this is done through the discrete cosine transform and quantisation of frequency components. The compression quality setting controls how aggressively data is discarded.

At high quality settings (80–90), lossy compression is visually indistinguishable from the original on most displays. At low quality settings (below 50), compression artefacts — blockiness, colour banding, soft edges — become visible. Most web use cases work well at quality 75–85.

Lossy vs lossless for WordPress images

Photographic images benefit most from lossy compression. The human visual system is poor at detecting the high-frequency detail that lossy algorithms discard, so the file size saving (40–80%) is achieved with little perceptible quality loss.

For graphics with sharp edges, flat colours or text — logos, icons, screenshots — lossless compression is more appropriate. Lossy compression of these images introduces visible artefacts at the sharp colour boundaries.

Mediapapa uses smart lossy compression: the quality level is set to deliver the smallest file that is visually equivalent to the original.

Frequently asked questions

Can I reverse lossy compression?

No. Lossy compression is a one-way process. The discarded data cannot be recovered. Always keep a backup of original files before bulk compressing. Mediapapa does not currently store originals after compression.

How much file size reduction can I expect from lossy compression?

Typically 40–70% reduction on photographic JPEG files. WebP lossy at equivalent quality achieves 25–35% better compression than JPEG. Results vary significantly based on image content — a flat, simple image compresses more than a detailed photograph.