The trade-off no one wants to make
By jon
We’ve been quietly running a small experiment for the last quarter. The setup wasn’t dramatic and the results aren’t either, but a few people asked what we did, so this is the long version of what we did.
Notes for next time
The honest reckoning is that we didn’t realise we were optimising for the wrong thing until we’d been doing it for a year. Looking back, the signals were there. The signal-to-noise ratio just wasn’t.
There’s a question I’ve come back to a lot recently: what would the simplest version of this look like, if no one were watching? It’s a useful trick for stripping away the things you’re doing for performance rather than for the work.
The middle ground
If you’re considering a change like this, the part I’d warn you about is the middle. The first week is fun because everything is new. The last week is fine because you can see the end. The third week is when most of these efforts quietly die.