People doing startups are natural optimists, bold risk-takers. And to get the money, you have to sell people on a dream of giant scale, on a big vision of what is to come. If you can't create and sell that vision, nobody gives you the money.

To then turn around and say, "Ok, what's the minimum necessary?" is really hard. You have to throw out 99% of your vision. You have to turn into a risk-averse pessimist. After talking about making the best thing, you have to go and make nearly the worst thing. Then you user-test it, discover why it sucks for your core audience, and make it suck less, in hopes that this time you've got the minimum viable product.

Making that turn is definitely a skill, something you have to learn and practice. It's not something we learn at all just building things to spec, which is the normal experience. It's the kind of product management that in theory everybody should do, but that in practice you only really are forced to do if you don't have much money.