3 criteria for good measurements

You can't improve what you don't measure.

(Or at least, if you don't measure it, you can't know if you've improved it.)

But how do you decide what to measure?

Outcomes, outputs, leading indicators, trailing indicators, expenditures — you could measure any of that, and more.

Except, of course, you can't measure everything.

Because even measuring requires an expenditure of time, effort, even money.

Here's a set of three criteria you can use in picking the measurements that will be most valuable to your work.

I recommend taking each of these in order. Start with the first criteria before moving on to the second and third.

1. Is it meaningful?

There's no point measuring something that doesn't matter to you. So to use this criteria, you have to think about what really matters to you.

Of course mission outcomes matter. They're the reason you're doing your work in the first place. So they would pass this criteria.

But other things are meaningful too. Leading indicators of success. Expenditures that might reveal an opportunity for, or the achievement of, increased efficiency.

When you hear yourself saying, "If I knew _________, then I could ...", that blank is something meaningful. Maybe it's worth measuring.

2. Is it measurable?

There are lots of things we'd like to know that are just hard to measure. For example, the actual impact you make in the world can be very hard to measure.

But often there are proxies that can be measured.

If you have access to solid research indicating that students who complete high school are less likely to get caught up in the criminal justice system, then the number of kids you've helped to complete high school is a proxy measurement for how many you've helped avoid jail time.

And for the things that you really can't find a way to measure, either directly or by proxy, the only honest answer is that you can't measure them right now.

And if something's not actually measurable, even by some approximation, isn’t it a little silly to say that you intend to measure it?

3. Is it manageable?

Finally, keep in mind that these measurements will require some level of diligence, attention to detail, and consistency.

And that level is a variable you can control.

Not every valid measurement has to be a randomized double-blind peer-reviewed study fit for publication in a national journal.

But for whatever level of detail you decide to measure, your effort has to be manageable, so that you can do it consistently over time, and so that your measurement will be more than a mere guess.

That means there's no need to get in over your head. You can pick a simple measurement scheme — which will provide a lot more value than no measurement at all — without it becoming a massive burden in your work.

Here's the thing:

Good measurements will provide valuable insights into your work and its effectiveness — and into ways that you can improve that work within your means.

Even the simplest measurement scheme, if it's meaningful to your work, measurable in some quantifiable way, and manageably consistent, really will get you some of those valuable insights.

All the best,
A.