Variance Matters as Much
Would you rather be right on average or predictable?
The idea
Two processes with the same average can behave completely differently if one is far more variable.
In the real world
Two suppliers averaging five-day delivery, one always five days and one between one and twelve.
Going deeper
Two processes with identical averages can be completely different to work with, because what people plan against is usually the worst case rather than the typical one.
A supplier who always delivers in five days and one who ranges from one to twelve have the same mean and very different usable capacity. Everyone downstream of the second must schedule for twelve while paying for five, so the variability consumes buffer throughout the system. This is why reliability is often what a customer is actually buying, and why an average alone cannot price it.
Where it stops applying
Reducing variance usually costs something — buffer stock, spare capacity, slower committed times. The goal is matching predictability to what downstream processes actually need.
Why it matters
Reliability is usually what people are buying, and averages do not price it.
Try this today
For a process you manage, record the spread as well as the average.
Test yourself
Two suppliers both average five-day delivery. One is always five days; the other ranges from one to twelve. Why do customers prefer the first?
Show the answer
Because the second forces everyone downstream to plan for twelve while paying for five. Predictability lets others commit; the average is identical and the usable capacity is not, since the worst case is what schedules must accommodate.
Learn this in the feed Answering from memory, then again days later, is what makes it stick.