Market Sizing Fundamentals
Estimating the size of a market from a few defensible assumptions and clean arithmetic, and knowing which number the question is really asking for.
What it is
Market sizing is the skill of getting to a credible number for how big something is - annual demand, total spend, unit volume - when nobody hands you the data. You build it from a small set of anchor facts you can reasonably know (a population, a household count, a usage rate) and a transparent chain of assumptions that turn those anchors into the target number. The point is not precision to the dollar; it is a defensible order-of-magnitude estimate and, more importantly, a visible reasoning path the interviewer can follow and challenge. A good market size is one where someone could disagree with a single assumption and see exactly how the answer moves.
When to use it
Use it whenever you need to bound an opportunity: how big is this market, how many units could we sell, is this even worth pursuing. It is common as a standalone warm-up and as a step inside larger strategy cases. Be clear which market the question wants - the total addressable market (everyone who could conceivably buy), the serviceable market (those you could actually reach), or a realistic near-term capture. Do not over-engineer it; three or four clean assumptions beat ten shaky ones. And do not use a sizing exercise to answer a question that is really about attractiveness or feasibility - a big market can still be a bad one.
How to apply it
Pick a starting anchor and a direction. Top-down starts from a large known quantity (a national population) and narrows with filters; bottom-up builds from a unit (spend per customer times number of customers). State each assumption out loud with a round, defensible number, and keep the units consistent as you multiply - people per household, purchases per person per year, dollars per purchase. Do the arithmetic in clean round numbers so you can track magnitude. Then sanity-check the result against something you know, and name the one or two assumptions the answer is most sensitive to. If two different paths land in the same ballpark, your confidence should rise.
Common mistakes
The most damaging error is losing the units: multiplying a per-week figure by an annual one, or people by households, and getting a number off by a factor of fifty. Another is hiding assumptions - blurting a final number with no visible chain, so it cannot be checked or defended. People also chase false precision (assuming 37.4% when 40% is fine and clearer), forget to sanity-check against reality, and confuse total addressable market with realistically capturable demand. And treating the final number as the deliverable, rather than the reasoning, misses what is actually being tested.
Example
You are asked to size the annual market for replacement bicycle tires in a mid-sized country of 40 million people. Bottom-up: assume roughly one in four people owns a usable bike, so about 10 million bikes, each with two tires, so 20 million tires in service. Riders replace a tire maybe once every three years on average, giving about 6-7 million replacement tires a year. At an assumed 20 currency units per tire, that is roughly 130-140 million per year. The number itself matters less than the visible chain - an interviewer can push on 'one in four' or 'every three years' and immediately see the answer shift, which is exactly the point.
Practice it
Practice on standalone sizing prompts and on the sizing step inside strategy cases, narrating every assumption and holding your units straight as you go. The skills to build are clean round-number arithmetic, an instinct for which assumption the answer hinges on, and the habit of sanity-checking the final figure against something you already know.