Reduce PDF file size without wrecking the quality

Two different things get called compression

Structural compression removes duplicated fonts, unused objects and uncompressed streams. It is genuinely lossless, and on a bloated export it can cut 10-30% with no visible change at all.

Image compression re-encodes the pictures inside the document at a lower resolution or JPEG quality. This is where the big savings live — 70% or more on a scan — and it is lossy by definition.

Choosing a level

• Light — for documents you will print or that contain small print, stamps or signatures.
• Medium — the right default for anything read on screen, including scanned contracts and receipts.
• Strong — for archiving or emailing when legibility of body text is all that matters.

Compression that makes files bigger

A PDF that is already optimised, or one that is pure vector text, can grow when its pages are rasterised. The tool always reports the before and after size, so if the number goes up, keep the original — nothing has been overwritten.

Step by step

1. 1Note your file's current size before you start.
2. 2Compress at the light level first and compare the result.
3. 3Step up one level at a time until you hit your target size.
4. 4Zoom into the smallest text in the output to judge the damage.
5. 5Download the version you are happy with — the original is untouched.

Frequently asked questions

Does compressing twice save more?

Almost never, and it degrades quality each time. Go back to the original and use a stronger setting instead.

Will the text still be selectable afterwards?

Not after image-based compression: pages become images, so text is no longer selectable or searchable. Keep the original if you need to copy text out of it later.