Charlie Munger
One hammer is not enough
The danger is the skill, not the ignorance
USC Business School, 1994
What he warned about was not ignorance. It was being very well trained in exactly one thing.
What actually happened?
At USC in 1994 he laid out the method of his life: you have to carry a number of models in your head and hang experience on that latticework. He put the count at eighty or ninety, drawn from mathematics, psychology, physics, biology and engineering. The warning attached to it is the line everyone knows: to a man with only a hammer, every problem looks like a nail — and the man with the hammer is not the ignorant one, he is the well-trained one.
The deeper the training, the better hidden the distortion
The hammer problem is not incompetence. It is that your discipline arrives with a viewfinder: the finance person sees a cost problem, the growth person a funnel problem. Each account is coherent, because any problem does have that face — and the coherence is what stops the search.
Churn was diagnosed as a pricing problem for three quarters by the only people who were asked, who all came from pricing.
Problems do not arrive by department
A pricing decision is economics and psychology and, because the competitor reacts, something close to biology. The latticework means laying several models over the same situation and seeing which one explains the most, rather than which one you already own.
The retention question turned out to be a queueing problem, and nobody in the room had ever been handed that model.
Load the models before you need them
A model you go and look up during the decision arrives too late to shape it. The models have to be in the head and automatic. There is no shortcut: take the two or three hardest ideas from each field and keep them.
He read about critical mass years before the network product, which is why the question came up in week one instead of year two.
How do I use it today?
Where you are: diagnosing a problem inside your own speciality.
Ask first: how would someone from a completely different field describe this? If you cannot produce a second account, you have one model, not a diagnosis.
Where it goes wrong: collecting models as trivia — the test is whether one of them changed a decision.
Lines to keep
The hammer man is not the ignorant one. He is the trained one.
Same situation, other people are asking
The more rules I add, the more people route around them.Everyone is chasing the same thing. Where is nobody standing?Every hour is booked and one delay wrecks the week.The numbers look great. Is that skill or is it leverage?I'm managing it and it's getting worse · all 14 questions →If this one named what you are going through,
send it to someone who needs it, or keep it somewhere you will find it again.