Measure Twice, Cut Once

The world and culture of “lean” startups has tended to emphasize doing. This is understandable. Lean is — as an approach and methodology — a reaction to what preceded it: waterfall approaches that began with requirements gathering, planning, development, beta releases and, finally, a product launch often many months and sometimes even years after the initial requirements phase. And nearly anyone familiar with startups today knows that steve blank and Eric Ries have led the revolutionary movement that has exposed so much of that historic approach to startups (and to software development) as wasteful, inefficient and ineffective.

But Eric’s innovation (and his book, “The Lean Startup”) was never really about meant to suggest that we simply “do” before we think or plan. This “Nike” approach or “just do it” approach to startups and software development has a certain appeal. It’s true, after all, that it’s a lot easier to steer a car once its moving.

Still, I’ve seen a lot of waste — time, effort and dollars — when people “do” before they think. And that reminds me of the old carpenter’s adage: “measure twice, cut once.” The point is obvious. Taking just a little more time meant that you avoided wasting a perfectly good piece of material — not to mention time and money. The lesson here is thoroughly consistent with what Steve and Eric have been saying: be smart about what you do; take steps to avoid waste that creeps into the system. Start with a pencil and paper or on a whiteboard before you hardcode your architecture.

Think. Just long enough, but no longer. Then do.