Fermi Estimation

Fermi estimation builds an approximate answer from a defensible decomposition, rough component estimates, and consistent units. It exposes assumptions and helps identify the information most worth refining.

Fermi estimation turns a poorly known quantity into an explicit calculation built from approximate components. The decomposition should follow a mechanism and preserve units, such as people times uses per person times time per use. Its value comes from making assumptions inspectable and revealing which uncertainty matters, not from avoiding evidence altogether.

Errors do not automatically cancel. Several optimistic factors can multiply into a large overestimate, and a missing component can dominate the result. Use ranges, an independent decomposition, and sensitivity analysis to determine whether the estimate is good enough for the decision. Refine the influential uncertain input rather than every number equally.

When to use it

When precise data is unavailable but approximate answers would inform decisions; when decomposition would make an impossible estimate tractable; when quick order-of-magnitude estimation would be sufficient; when the most uncertain sub-estimate needs identification for targeted research.

How it can help

Define the quantity and needed precision, estimate components with ranges, check the mechanism and units, and test sensitivity or an independent route.