You measure because you want to make better-informed decisions. But even simple, harmless-looking measures can be dangerous. For example, they can give you a nice, clear picture of an illusion. Do you want to base your decisions on illusions? Technical Editor Brian Lawrence advises that, before you dive into measuring anything, ask yourself, "Will measuring do more harm than good?"
In his article this issue, Rex Black says that if you, like Lord Kelvin of long ago, want knowledge beyond "a meager and unsatisfactory kind" you can decide to measure things. Maybe you've already done so. If you have, did it work out as you intended? I don't think we have much of a winning record for measurement in pretty much all of the software industry over all of its history. I believe we can do better, and I worry that what we've done so far is positively dangerous. Measurement has a way of creating self-fulfilling prophecies.