It sounds to me like you are trying to balance deep vs wide in an MVP. Wide being a large group of features/hypothesis, deep being more focus on fewer features

It sounds to me like you are trying to balance deep vs wide in an MVP. Wide being a large group of features/hypothesis, deep being more focus on fewer features. I think it depends on the situation, but for the most part I believe MVP should be neither of these things, or somewhere in the middle. It isn't about testing many hypotheses, it is about testing one hypothesis. "Is there a market for x". The answer isn't to build a robust release or build a facade for many tests, but to build the minimum amount to answer that question. You don't want to build deep into the technology of a single 'feature', and you don't want to build many features. What is the minimum number of features and the minimum amount of those features that can be built in order to answer the hypothesis.