Compress PDF

Three levels, and what each one costs you

Maximum produces the smallest file and is the default. Balanced keeps noticeably more detail. Light barely touches image quality and is the right choice for anything you will print. The difference between them is how finely each page is re-rendered and how hard the images are compressed, so the trade is straightforward: smaller file, softer detail.

The honest catch most compressors do not mention

To get a real reduction on an image-heavy or already-optimised PDF, this tool re-renders each page as an image and rebuilds the document around it. That is what makes the file genuinely smaller — and it means the text in the compressed copy is a picture of text. You will not be able to select, copy or search it, and it will not be readable by a screen reader.

For a document you are emailing to be looked at, that is usually a fine trade. For anything that needs to stay searchable — a contract you will later need to find a clause in, a thesis, anything going into a document management system — keep the original and send the compressed copy only as a viewing version.

It also tries the lossless route first

Alongside the re-render, the tool runs a structural pass that strips out document metadata and repacks the internal objects more efficiently, without touching a single pixel. Whichever of the two results is actually smaller is the one you get, and when the structural version wins the tool tells you so — in that case your text stays fully selectable. If your PDF is already well optimised and the reduction comes out under 5%, it says that too, instead of handing you a "compressed" file that saved nothing.

Before and after

If the PDF is large because it contains many separate documents, merging is not the problem — try Split PDF to send only the pages that matter. If the size comes from photographs you assembled yourself, compressing the images first with Compress Image and then building the PDF with JPG to PDF usually beats compressing the finished file.