What gets measured...

Reading time: 3 min

Today’s high-signal drops:

• Makes-You-Think Tweet: good explanations > what you see
• Mind-Expanding Concept: what gets measured…
• Cool Quote or Question: an inherent flaw of setting goals
• High-Signal Content: the day the delusions died

Makes-You-Think Tweet

Mind-Expanding Concept

What gets measured…

The classic "what gets measured gets managed" is thrown around like hot frisbees. It's usually conjured when one wants to highlight that someone or some group is not making decisions based on data. It's normally attributed to Peter Drucker, the famous management consultant and thinker, probably one of the reasons it's so sticky. No one wants to challenge an OG like him.

A few problems though. One, Drucker never said it, according to the...Drucker Institute. But second and most importantly, it's so low-resolution that it becomes flawed in many instances. A few examples:

What gets measured can get (automatically) mismanaged. By measuring some things, we automatically and often unconsciously, optimize for them. And that can be counter-productive at worst or an empty-calorie vanity metric at best. Counting how many books you've read in a given period or how many countries you visited falls in here. If you track the number of books you read, chances are that you'll do a combination of (i) plowing through them, (ii) avoiding dumping bad books midway, (iii) resisting rereading impactful books, and (iv) listening to audiobooks at 2x speed. I'd argue that none of these is conducive to learning or a deep exploration of ideas. That's especially true if you share these numbers with friends or on social media. What to do then? Don't count them. Don't put yourself in a position to optimize for a useless metric. If you have the chance, you will.

What gets measured gets gamed. Goodhart's Law - when a measure becomes a target, it ceases to be a good measure. If media companies are paid by the number of clicks on their content, they'll write about the most sensationalist things in a dopamine-capturing way, often disregarding fact-checking on the way. If we measure how much children are learning by test scores, they'll eventually focus on memorizing content and learning how to better take tests, instead of actually focusing on mastering the content.

You can manage without measuring it. I can't plot on a graph the relationship between the time I spend with my daughter and how close we are. It's not knowable or measurable discreetly. Nonetheless, I'm sure there's a positive relationship between the two. Does it have diminishing returns after a certain point? Possibly. Will she, at some point, want to spend less time with me to spend more time with her friends? Likely. Still, when I notice I haven't spent time with her in a few days, I know I must course-correct. Not everything that matters can be measured.

I'm not arguing against data-driven decisions. I'm arguing for not using low-resolution statements like they're gospel.

Cool Quote or Question

High-Signal Content

“A friend of mine joked that she woke up on October 7th as a liberal and went to bed that evening as a 65-year-old conservative. But it wasn’t really a joke and she wasn’t the only one. What changed?”

See you next week,

Filipe