PDF compression is a trade-off, not a magic switch. The best result depends on whether your file is mostly text, scanned pages, photos, vector artwork or a mixture of all four. Start by deciding what must remain readable and what can be simplified.
Start with the source
Keep an untouched original before you compress anything. Check whether text is selectable, whether pages are scans, whether the file contains forms or signatures, and whether images need print quality. A scanned contract behaves very differently from a digitally generated report.
Use the smallest sensible change first
If the PDF already contains selectable text, avoid converting every page into an image just to save space. When image downsampling is available, lower image resolution gradually and compare the result at 100% zoom. For screen reading, very high print-oriented image resolution is often unnecessary.
Watch for hidden quality losses
A smaller file can look fine at a glance but still have blurry charts, unreadable footnotes or flattened form fields. After compression, search for a known sentence, zoom into small text, check page order, test links, and compare a few image-heavy pages with the original.
Choose compression based on destination
Email attachments usually need a practical size, not the absolute minimum. Documents intended for printing need more image detail than documents viewed only on phones. If a portal publishes a hard upload limit, target a little below that limit so later metadata or signing does not push the file over.
Worked example
Example workflow
A 24 MB scanned report must fit a 10 MB upload limit. Keep the original, reduce scan-image resolution gradually, export a copy at about 8–9 MB, then check the smallest footnotes and several image-heavy pages before uploading.
Practical checklist
Before you finish
- Compress a copy, not the only original.
- Check the smallest text on the page after compression.
- Prefer one controlled compression pass over repeated conversions.
- If a PDF is mostly scans, test OCR separately rather than assuming compression creates searchable text.
Common mistakes
What to avoid
- Compressing the only copy of an important document.
- Rasterizing a text-based PDF before trying a lighter compression method.
- Checking only the cover page while small text or charts become unreadable later in the file.
- Running multiple lossy conversions in sequence.