Metrics

One of my hard-won lessons is that what you choose to measure inevitably controls your thinking. That is, if you decide that you're going to measure, say, your car's mileage, you'll inevitably find yourself doing something with that number - either you'll seek to drive it down to as low a value as possible, to conserve fuel, or you'll seek to drive it up as high as possible, to see what your "high score" will be on the car. Or something - but people very seldom measure something and then do nothing with that information.

What that means is that you need to pick and choose your metrics very carefully, and just because something is easy to measure does not mean that you should measure it - if something is irrelevant then it's better not to know, so that you focus on other things that are important.

(Indeed, the effect of this can be readily seen in the blog: I've been tracking the number of books read, which has the effect that I've been pushed towards reading shorter books so I can clear them out. Had I instead chosen to count the number of pages read, that would have led to different behaviours.)

The upshot:

• Decide what is important, decide in what way it is important, and measure it (in the right way). Then review the results and take action.
• For things that aren't considered important, don't measure them!