PDF knowledge base

PDF compression: what actually gets smaller?

A PDF can be large because of photos, embedded fonts, duplicated resources or object storage. Each cause needs a different optimization.

Structural optimization

The compressor on this site rewrites document objects using compressed object streams. It measures the output and returns the original if the rewrite is not smaller. It does not claim a minimum saving and does not replace an unchanged result with an invented percentage.

Image compression changes a different thing

Downsampling photos or lowering JPEG quality can reduce scanned-document size much more, but may blur fine type. That operation is different from the structural optimizer supplied here. If you control the scanner, use an appropriate resolution when making the original scan.

A practical check

Keep the original. Compare byte sizes, zoom in on small text and inspect every page. If a portal requires 200 KB and your scan is still 2 MB, changing object storage alone is unlikely to be enough. Do not repeatedly compress a lossy image and expect its quality to improve.

Try the related tools

Further reading

Updated 2026-10-18. These are implementation notes and practical explanations, not a conformance certificate.