Most PDF compression works by turning pages into images, which is very effective and quietly destroys something you may need.
Searchable text is what lets you find a clause in a contract, what lets screen readers read a document aloud, and what lets automated systems process an application. A PDF that has become a set of pictures loses all three at once, and nothing about it looks different on screen.
For a document that is mostly typed text, heavy compression saves surprisingly little anyway. The text was never the large part. You pay the full cost and get almost none of the benefit.
The better fix for a text heavy PDF that is too large is usually to find the one or two images inside it and replace them with smaller versions, then export again from whatever produced it.
Open the compressed file and try to select a sentence. If it highlights word by word, the text is intact.
The gentlest one is most likely to. The smallest setting rasterises pages, which does not.
No.