What actually gets smaller
A PDF is usually large for one reason: the pictures inside it. Text is vectors and weighs almost nothing — a hundred pages of writing is a few hundred kilobytes. A single phone photo dropped into a report can be four megabytes on its own, and a scanned document is nothing but photographs, one per page.
So compressing a PDF means finding those images, decoding them, and re-encoding them smaller. The text, the links and the form fields are copied across untouched. That is why a text-heavy PDF barely shrinks and a scan can lose 90% of its size: the tool is working on the part that is actually big.
- Images are re-encoded
- Each embedded image is decoded and written back at a lower quality. This is the part that is lossy, and the only part.
- Text stays text
- Words remain selectable and searchable. Compression does not flatten your document into pictures — that would make it larger, not smaller, and would break copy and paste.
- Structure is preserved
- Links, bookmarks and form fields survive, because they are cheap to store and expensive to lose.