Even very agile companies look very brittle in the short term. At GrubHub.com We have a weekly product update cycle. We keep very aggressive deadlines, try to keep meetings short and generally don't screw around. Sometimes we take a step back and say "Damn! we get a lot done"
So why is it that we look at the stuff we want to get done and despair that some of the important projects are always just over the horizon? We do a pretty good job of keeping things lean. So what gives? Well, its a matter of scope. When you are looking at implementation on too short a time scale, expectations grow out of whack. So I'm coming up with a new rule:
My implementation calendar is on a weekly basis.
My evaluation calendar is on a monthly basis.
Hopefully this will keep my strategery from experiencing the phenomenon known as thrashing in Computer Science. I'll let you know how it works out.
Comments