If you are trying to improve the performance of existing operations in known markets, it is an analytical problem where it's just a question of aligning your execution engine in the right way. If it is about creating something new and different, you can't derive the right answer analytically.

I've never seen impeccable logic be sufficient to win both the heart and the mind.

Avoiding along some dimensions is easy. Create organization space and bring in some new talent so the innate cultural resistance is less material. Unfortunately, it's rarely that easy.

In the early stages of innovation, your goal is to learn as much as you can as quickly as you can.

The most important thing here is to largely ignore what customers say, and instead watch what they do or track where they spend money.

I think it is only in hindsight that you can determine whether something is a mistake or not.

I'm suggesting that principles meant to deal with uncertainty that occurs naturally can be useful to manage the uncertainty that characterizes any new idea.