How to compress a PDF without turning it into mush

The phrase “compress PDF” hides two completely different jobs. One is throwing away pixels in a photograph of a page. The other is stripping unused objects from a born-digital export. Most free websites do the first job to everything, which is why a 12-page contract comes back looking like a fax.

Start with what the file actually is

Open the PDF and try to select a word. If you can highlight text, you have a text layer. Compressing that file by flattening it to JPEG pages will make it smaller and also make it worse — you will lose search, copy, and screen readers. If you cannot select text, you have a scan. Flattening is then the right tool, not a compromise.

Pick a quality like you pick an envelope

Strong is for email and chat: readable, light, not archival. Balanced is the everyday packet. High quality is for a deck that still has to survive a projector. If the “compressed” file is larger than the original, keep the original. There was nothing to gain.

Do the other work first

Split off the pages you do not need. Rotate the sideways scan. Then compress. A 60-page phone dump with three useful pages should never be compressed as a 60-page phone dump.

Keep the file in the tab

A compressor that requires an upload is asking you to trust a retention policy. A compressor that runs in the browser is asking you to trust the machine you already logged into. For anything with an account number on it, that is not a close call.