How does PDF compression actually work?
A PDF's size comes mostly from two things: embedded fonts and embedded images. Compressing a PDF reduces one or both of these, which is why results vary so much between documents.
Text-based vs image-based PDFs
- Text-based PDFs (created from Word, Google Docs, LaTeX, invoices, reports) store text as digital characters and fonts. The text itself is tiny; most of the size is fonts and simple layout. These compress very well, often shrinking by 50–90%.
- Image-based PDFs (scans, photos saved as PDF) store every page as a picture. There is little text to compress, so the only lever is image quality — these shrink less and lose more visible quality.
Why that matters for you
The biggest choice in PDF compression is how aggressively you reduce embedded images. Files created purely from text can usually take the strongest level with no visible change. For scans, a light or medium level balances size against legibility.
Lossy vs lossless
- Lossless compression reduces size without changing any visual content — ideal for archives and contracts where fidelity matters.
- Lossy compression trades a little quality for much smaller files — ideal for email, web upload, and storage where a tiny quality drop is invisible on screen.
A good tool, like Compress PDF, lets you choose a level or target size so you decide how much quality to trade.
When to compress
- Before attaching a PDF to email or a portal that caps file size (target sizes like 100 KB or 1 MB help here)
- Before uploading to a form, CMS, or applicant system
- To speed up sharing large manuals, reports, and presentations
When NOT to compress
- Before professional printing of image-heavy documents, where maximum quality counts
- When a document must be archived long-term and you want zero information loss
Verdict
Compression is the fastest way to make a PDF shareable. Use Compress PDF, match the level to whether your document is text or image based, and keep the original if you need a full-quality copy later.