PDF compression explained: lossless vs lossy
What actually happens when a PDF is compressed, and why the results vary so much between files.
Skip the reading. Use Compress PDF.Shrink it until the email goes through. Free, no sign-up, nothing uploaded.→A PDF is a container
A PDF bundles text instructions, fonts, images and metadata into numbered objects. Each object can be compressed separately, which is why two PDFs that look identical can differ in size by 20×.
Lossless: tidying the structure
Lossless compression shrinks the file without changing anything you can see. It packs objects into compressed streams, drops unused objects and strips metadata. Savings are usually modest, around 5–30%, but quality is untouched.
Lossy: re-encoding images
Lossy compression throws away detail you're unlikely to notice. Images are resampled to a lower resolution and re-encoded as JPEG at a chosen quality. This is where large savings come from on scans and photo-heavy files.
- Resolution (DPI): how many pixels per inch. Around 100–150 DPI is fine on screen; print usually wants 200–300.
- JPEG quality: higher keeps more detail. Below about 50%, blocky artefacts start to show around text.
Why I Hate PDF tries both
There's no single best method. A scanned contract shrinks dramatically with lossy re-encoding, while an exported report barely moves and would lose selectable text. So we run both methods and give you the smaller result, and we tell you which one won.