But we chose to forego the "consultancy" jobs. Reason is that we optimize for learning (we're at that stage). And while consultancy may look like a good opportunity for learning, that it allows you to get paid for investigating the real problem domain of your customers it is not optimised for that.
Far from it. My decades of consultancy and freelancing experience has told me that it always takes longer to implement the solution than to identify the problem. Implementing is a great opportunity for learning the details of the problem, sure. But Implementing also eats vast amounts of time to "Horizontally center that Div on the Dashboard": any non-trivial, time-consuming but utterly uninteresting for our startup.
I really don't want to spend two weeks implementing another authz/authn integration and login flow for a consulting customer. When in those two weeks I can interview at least three other customers. Esp. because authn/authz is but a detail to our startup, and has nothing to do with our domain.