Dan Coelho 1) Always start with why? 2) Understand the problem 3) Embrace uncertainity 4) Balance inputs, outputs, outcomes and learning 5) Iterate, Iterate