Loading your local toolbox…
Most "compress PDF" tools simply turn every page into a JPEG. That shrinks the file fast — and destroys the text layer. Here is how to choose properly.
No upload. No watermark. No sign-up. Every tool mentioned here runs inside your browser.
Structural: rewrite the file without redundant objects. Keeps text selectable. Shrinks 5–30%. This is our Safe mode.
Image re-encoding: re-compress the images already inside the PDF. Keeps text selectable, shrinks scans a lot. Best of both worlds, but it needs an image-aware engine.
Rasterisation: render each page as a picture. Biggest shrink, but the text stops being text. This is what Target size and Light/Strong modes do.
For a text document that must stay searchable: Safe mode.
For a scan you are emailing: Light mode (150 DPI) is usually indistinguishable on screen.
For a portal with a hard size limit: Target size, and accept that the text becomes an image.
Do not compress an already-compressed file repeatedly — each pass loses a little more. Always start from the original.
Do not compress a PDF just to merge it; merge first, compress once.
It was probably already optimised, or it is mostly text (which is small). Text-heavy PDFs have little to gain from rasterisation.
You can set a target and the tool will step quality down until it fits — but a target is a goal, not a guarantee.
Loading your local toolbox…