If you hypothesise, design, prototype then test, there's a very high chance that you will have built something that isn't quite what is needed, by which time the team has fallen in love with the product and ploughs ahead trying to tweak it into success, leading to a lot of pain and headaches later.

If you hypothesise, test (interview), design, test (concept test), prototype, test, then build, you throw an awful lot of stuff away before you even have a high fidelity prototype, and have a far clearer idea of what is actually needed.