So by saying "we're going to go with REST", you can determine about 90% of your API design more or less instantly, and it rarely gets in the way.
You still have to get your domain modeling right, and sometimes you need to make some extra resources. But you have a design that, basically, works.
Even for CRUD apps with simple master/detail data relationships it doesn't make sense to tie your domain design to network requests.
You mean like GraphQL? ;)
Then I read the actual documentation, and I felt like it was what I had been missing before.
Now I just need to find a decent implementation.
E.g. you can stick an off-the-shelf caching reverse proxy in front of a server.
E.g. you can split data centers and direct your hypermedia links there.
I think there are often more important things then just interoperability and flexibility, but those were the guiding principals of REST.
Most of the benefit of REST over SOAP was the fact that it used HTTP correctly instead of implementing a redundant protocol on top of it. Having machine-readable API contracts is still extremely valuable; we just found better ways of doing that.
It doesn't get you the "flexibility" of hypermedia/HATEOS, but it depends on your application if you need that. Most projects don't call for 15+ year APIs.