Guides

Compressing a PDF without wrecking it

2026-09-22 · 5 min read

A PDF holds very different things: text and fonts, already compact; vector graphics, mathematical and cheap; and images, which dominate the weight of almost every heavy file. Meaningful compression therefore means re-encoding the images, and that sets the expectations.

Why results vary so much

A scanned document is nothing but full-page images and can shrink dramatically. A text-only contract exported from a word processor is already tight, and no compressor will make it meaningfully smaller. If a tool promises a fixed percentage, it is describing its average rather than your file.

What compression costs

It is lossy where it counts: image detail is genuinely reduced to buy size. For an email attachment or an upload limit that trade is almost always right. For an archival master or a print file, keep the original and compress a copy. Text is never touched, so the result stays selectable and searchable.

Getting under a 2 MB or 5 MB cap

• Compress the scan, not the born-digital form. The form is already small and has nothing to give.
• In an email thread with a dozen attachments, compressing the two scanned ones usually saves more than compressing all twelve.
• For archives, re-scan at a saner resolution rather than compressing an old scan repeatedly. Lossy passes stack.

The fifteen second check

Open the result and zoom to 200 percent on the page that matters most: a signature, a stamp, a table. If it reads cleanly the compression was free. If it smudges, run it again at a lighter setting. That beats resending a document a registry rejected as illegible.

Frequently asked questions