The Piano Tuners Question

Nobody expects the candidate to actually know how many piano tuners work in Chicago. The question, popularized by physicist Enrico Fermi and a staple of consulting and quant interviews, tests something else entirely: can you take a question with no available data and decompose it into a sequence of smaller estimates that are each individually plausible, so that the final number, even if wrong, is defensibly in the right order of magnitude?

The standard approach chains together population, household counts, ownership rates, and service frequency. Chicago has roughly 2.7 million people, or about 1 million households. Suppose 1 in 20 households owns a piano — that's 50,000 pianos. Each piano needs tuning roughly once a year, and a tuner can service about 4 pianos a day, working perhaps 250 days a year, so one tuner handles about 1,000 pianos annually. Dividing 50,000 pianos by 1,000 pianos-per-tuner gives roughly 50 piano tuners in Chicago.

The actual number (from business directories, when checked) tends to land within a factor of two or three of this kind of estimate — close enough that the exercise validates the method, not the precision. What interviewers are really scoring is whether each intermediate assumption was stated explicitly and defended with a rough justification, rather than pulled from nowhere, and whether the candidate caught and fixed an obviously wrong intermediate number (like assuming every household owns a piano) before it propagated to the final answer.