I think there are strong parallels between PhD research and invention (though most startups these days don't have that basis). A patent requires researching prior art (literature) and describing your idea formally (papers/thesis). A PhD must be original and substantial (novel and non-obvious). In both cases, you are need to obtain your examiner's approval. :-)

In other words, formally describing a new abstract idea in the context of what has gone before is common to both - but actually creating a working technology, that people want, is different from a PhD.

That said, the beginning part of a PhD is great fun.