Comment by Filip Milovanović on What empirical evidence is there that the separation between analysis and design leads to better results than mixing them?

Depending on what you mean by this, this may not be a very controversial claim. E.g. this could just mean that, while technical details are important, what's more important is to nail down exactly what is it that the business actually needs, so that you don't build the wrong thing; decide on the technical details in the service of that. Often, instead of stating what the core problem is, clients will come asking for what they perceive to be a technical solution ("Rewrite in a fancy framework following all the best practices!"), and you need to look past that and dig deeper ("Ok, but why?").