Docker's valuation didn't pan out because they tried to swim against Kubernetes.. compare that to CoreOS and openshift.

In this case it should be compared to valuation of tools like figma

Figma is a lot harder to replace than postman and it's primary users are not developers (although they often consume the result of the primary users). I think developers are more fickle than designers as a customer, especially for tools that are not central to the end result.

If I deploy my app and it works it does not really matter if I tested/documented it with postman, it's not central to the app at runtime.

I also think that postman's core product is a lot easier to replicate (they were not even the first to do it) than figma and there are a number of open source projects that do.

When it comes to their API gallery I would expect that any serious API would come with a non-proprietary definition or documentation (in something like OpenAPI or at least just good docs) too.

I don't know, I just have a hard time to see why this product (and by extension the company) is worth 5.6B.

Developers don't make the call. Companies would ask people to create and share a postman collection with API tests instead of checking it into git.

Valuations are crazy these days. But I don't see how this is crazier than other SaaS valuations. Why github when you can run your own server? Why use confluence when you can check in your docs into git?