Fermi Estimation

Breaking a question that looks unanswerable into smaller parts that can each be estimated, then recombining them into an overall answer. Named for the physicist Enrico Fermi, the move trades one impossible guess for a few tractable ones, exposing the assumptions behind the number along the way.

Why it Matters

Faced with a hard quantity, people either refuse to answer or blurt a number pulled from nowhere. Decomposition turns the problem into a chain of more knowable sub-estimates, so the final figure rests on visible, debatable assumptions rather than a single opaque hunch, and errors in the parts often partly cancel.

Signals

• A question dismissed as impossible to estimate.
• A confident headline number with no breakdown behind it.
• Disagreements that cannot be located because no one has decomposed the estimate.

Benefits

Tractable sub-questions in place of an intractable whole, explicit assumptions that can be challenged one at a time, and a structured estimate that is easier to update as any part changes.

Risks

A decomposition that omits a dominant factor; false confidence from a tidy calculation built on shaky inputs; spending effort breaking down a question that a reference class would answer faster.

Tensions

Decomposition costs time and can over-engineer a quick judgment call; the skill is knowing which questions are worth the breakdown.

Examples

Estimating a market's size by population times adoption rate times spend, rather than guessing a total; splitting a delivery date into component tasks before summing, instead of naming a date outright.