A common assumption, which is sometimes even right, is that getting a significant number of users is harder than monetization. Put differently, your startup is most likely to fail (in this view) because it doesn't produce something that anyone wants to use, not even for free. So the main goal up front should be to figure out how to make something people will use, and then figure out how to reach them. Succeeding at that, but then failing to monetize the product, is a real failure mode too, of course. But many VCs are betting that failing to get users at all is the biggest early risk, and that it's easier to solve monetization later (if you ever get users) than it is to work on the monetization plan up front, and then later try to solve the but-we-have-no-users problem.

Part of this makes more sense if you're looking at it from the perspective of a VC betting on 100 companies, than from the perspective of a single company. The funnel they're looking for is: some subset of these companies will get a ton of users (hundreds of thousands, maybe millions), then a subset of those will be wildly profitable.