Diagnosis and Prediction Are Different Jobs
What is the difference between diagnosing engineering waste and predicting it?
Diagnosis measures the present: hours in coordination meetings, how long fixes take, and how often a change breaks something else. Translate those hours into currency with the average engineer cost and you have the current waste. Prediction takes that state and asks what each quarter costs over the next two years if the next features ship the fast way. A leader who sees the widening gap can decide differently from a leader who only sees the price of one feature.
Why is a falling velocity number a diagnosis of the system?
Velocity as an absolute number is a vanity metric, and story points are worse. If you track velocity for the same engineer across quarters and the number drops, the engineer did not get slower. The system got harder to maintain. That reading is the starting point for the projection. It is not the projection itself.
Get notified when new content drops