The point about Lean Startup is close, but ultimately incorrect. The whole idea behind pivoting is to make a dramatic change when iteration starts to put you into a local maximum. Eric Ries explains it exactly in this way: when each additional iteration leads to less and less improvement, that is when it is time to pivot. Pivoting is not A/B Testing, it is not iterating. It's making a leap in a new direction.