Standing ovations

When you want to make an improvement, it helps to know your current metrics as a starting point. (Otherwise, how can you tell if you've improved — or if you're even moving in the right direction?)

But if you haven't been tracking metrics carefully, how can you even know where you're starting?

It's probably easier than you think.

Imagine you're the artistic director for your city's symphony orchestra, and you want to increase the quality of your performances (probably because you want to increase ticket sales and season memberships).

You could take steps to improve your performances, but how can you measure that?

Hire a big firm to conduct randomized surveys of a statistically relevant sample of your patrons? Ask patrons to rate ready performance from "poor" to "excellent" and compare the results over time?

Douglas Hubbard, in his How to Measure Anything, explains how the Cleveland Orchestra did it:

They counted the number of standing ovations.

Hubbard writes:

While there is no obvious difference among performances that differ by a couple of standing ovations, if we see a significant increase over several performances with a new conductor, then we can draw some useful conclusions about that new conductor. It was a measurement in every sense, a lot less effort than a survey, and—some would say—more meaningful. (I can’t disagree.)

Here's the thing:

For any outcome that you really care about, you could spend a lot of time and money measuring exactly that thing plus a long list of contributing factors leading up to it.

But there's probably a relatively easy proxy you could measure instead.

It wouldn't be an absolutely precise measurement — but nothing is, really (Neil deGrasse Tyson makes a good point).

Any measurement at all is still a lot better than no measurement.

What's so bad about a low-effort low-cost measurement that tells you most of what you need to know?

All the best,
A.