Measuring the Immeasurable

I’ve been an avid skier for more than half my life. When we get our first light October snowfalls here in Colorado, I start getting serious about getting in shape. Sadly, this winter my consulting practice has been so busy that I haven’t been able to give much more attention to skiing than I’ve been able to give to this blog. Which is not much at all. Given the state of the economy its certainly good to be busy, but realizing that the Colorado ski resorts start closing next month – and that the snow is already getting slushy – is downright depressing.

At 55, I’m a more careful skier than I was at 25. Bones and ligaments heal slower, and I’m not as strong as I was a quarter of a century ago. Someone has stolen much of the cartilage out of my knees when I wasn’t looking. I also outweigh my 25-year-old-self by – well, more than I’m willing to tell you. For all that, I’m a much better skier now than I was when I was 25. Then, it was all about the adrenaline: Go really fast and rejoice in being intact at the bottom of the run. The risks to a 55-year-old body have forced me to slow down and pay more attention to skills and technique. I’m sure I still look like a doof on the more technical terrain, but I’m better than I was.

At the top of any ungroomed slope I do a quick risk assessment: The angle of the slope, bumps, the likelihood of hidden obstacles, the condition of the snow, the quality of the light. Whether I’m warmed up or worn out.

Something could still go wrong, but my risk assessment reduces uncertainty enough that if I decide to ski the run I have an acceptable level of confidence that I will arrive at the bottom of the run unhurt and unembarrassed.

I said a couple of important things, in that last sentence, about risk assessment: I reduce uncertainty and I increase confidence.

In my previous post (man, I’m sorry that was so long ago!) I contrasted a count, which gives you an exact number, with a measurement, which is an approximation. In other words, a measurement is a reduction in uncertainty. Not an elimination of uncertainty (and this is an important point!) but a reduction in uncertainty. The more accurate we can make the approximation, the more uncertainty that’s eliminated. But there is some point, as I emphasized in the previous post, at which a more accurate approximation becomes unnecessary and perhaps unjustifiably expensive.

One of the questions I asked at the end of that post is, “How do we measure things that we might think of as intangibles, such as risk and security?” The first step, which seems obvious but in practice is often overlooked, is to define the “intangible” as precisely as possible within the context that we plan to use it. The question is simple: What do you mean by “risk?” What do you mean by “security?” What do you mean by “quality?” What do you mean by “performance?”

The answer is not so simple, or shouldn’t be. You and your project team could – and should – spend one or several meetings just coming up with the answer to “What do we mean by…?” It’s an exercise that puts specifics around a fuzzy concept, and what you will find is that as you begin listing the specifics, you begin realizing that most of those specifics are measurable.

Douglas Hubbard defines this exercise as a clarification chain:

1.     If it matters at all, it is detectable/observable.

2.     If it is detectable, it can be detected as an amount (or range of possible amounts).

3.     If it can be detected as a range of possible amounts, it can be measured.

As you come up with measurable specifics, uncertainty is reduced and confidence is increased.

The next question, which I’ll address in the next post, is: How do we determine an acceptable confidence level? Or as I asked it in my previous post, how do we know a measurement is close enough?

In the meantime, I’m skipping IETF in San Francisco next week to get a handle on my backlog of work. And maybe (don’t tell anyone) to duck up to the mountains for one last time before the snows melt.

Related:

Copyright © 2009 IDG Communications, Inc.

The 10 most powerful companies in enterprise networking 2022