You don’t drive a car by planning every turn ahead of time. Getting to the right destination is a process of making a million tiny decisions, based on the new information you get every step of the way.
Yet this is exactly how most IT projects are run. They turn the exercise of driving down to the corner to get some milk into the moral equivalent of a NASA mission to Mars. And guess what? Most dev teams aren’t really up to planning a mission to Mars. They come up against the first unexpected obstacle, and have no idea how to get back to their plan. Or worse, they stick to their plan, disregarding the obstacles, and produce something that just doesn’t do what the business actually requires, even if it were a perfect implementation of the plan.
And just to put this into a little perspective: Let’s talk about the relative uncertainty of an activity like driving to the corner to get some milk, as compared with, say, launching a new startup to capture the enthusiasm (and dollars) of the social web, or tweens, or dog owners. I don’t think I’m being too controversial to suggest that the milk scenario doesn’t quite rise to the level of ‘conditions of extreme uncertainty’. And yet, even in this fairly bounded case, you’d be mad to plan your whole journey, and ignore all the little feedback along the way. 2 degrees off, and you find yourself scraping bits of stop sign out of your windshield. Ignore the competition, and you find the competition’s hood conveniently intermingled with your door frame. At least in the driving case, you have airbags. In the startup case, you have a lot of explaining to do to your investors.
In the real world, we check for feedback all the time. It’s how we make things like driving, or even walking, seem effortless.
The fact is, we can take an awful lot of risk out of the process of building startups by gathering as much feedback as we can.
And of course once we gather the feedback, treating it objectively is important. If it’s looking pretty likely that we’re going to drive into a ditch, we’re probably better off correcting our steering than building nicer powerpoint slides that say we’re going in the right direction.