Building a startup is like climbing a mountain where you can only see forward, but not up. In fact, the most accurate way that I think I can describe building a startup is like this:
Every time you think you've made some good progress toward success, you discover a giant ravine as you realise that half the stuff you just spent all of your time and energy building was wrong. It got you to where you are now, but it won't get you any further up the mountain.
Each time you think "ok, I've got it now" - and then you hit the next ravine.
To some extent, this is a gift. If you were able to look all the way up and see just how high that mountain is - you'd probably never climb it in the first place.
Conversely, a large part of startup success is the ability to look up as far as possible. When people talk about founder "vision" - I like to think of it as the ability to look far up the mountain. To avoid as many ravines as possible and stay on a path which doesn't lead to a single, isolated peak.
The remaining part, I think, is how many times you're able to say "fuck." - and then just keep on going anyway.