I’m a little down on the notion of continuous improvement these days. It’s not that it doesn’t happen – it does…sometimes. I simply fear that it promises too much. I think part of it is the terminology. You see in the real world continuous improvement is neither truly continuous or necessarily an improvement.
To begin with, I’ve critiqued the notion of continuously improving before from the point of view that keeping any process happening full time is ludicrous. Certainly once every sprint is nowhere near continuous. I guess maybe it is continuous when you compare it to other more plan driven methods, but that’s far from continuous in my book.
No, the part that I find most objectionable is the improving part. You see, it’s misleading. It suggests to me that every change is an improvement. That every effort is a step forward, not back. And that is simply not how it works. It would be better labelled periodic experimentation, or punctuated mutation. You see, in the real world, when we change something, we never really know if it’s going to work out or not. There are 50/50 odds that the change will actually make things worse!
Of course that’s a good thing. We learn a little either way. Hopefully.
The problem I have with continuous improvement is that it sets up an unreasonable expectation in those we sell it to. To restate the sales pitch: every change will be an improvement and they happen all the time.
If every change were really an improvement, I would be worried that I had been transported to an alternative universe. That I was being monitored by aliens. That there was a black helicopter hovering over my house. I’d be making a tin foil bunny suit. Fortunately I know what universe I’m in, that there aren’t any black helicopters over my house, and tin foil tends to chafe in the damnedest places. Not too sure about the aliens…
Fortunately, many of my efforts at improvement fail. And that’s the way it should be.