Market Sizing Cases: How to Estimate Anything Under Pressure

Master market sizing cases with a repeatable framework. Ex-Bain manager explains the approach that impresses McKinsey, BCG, and Bain.

Why Market Sizing Trips Up So Many Candidates

Market sizing questions sound deceptively simple. “How many piano tuners are there in Chicago?” or “What’s the annual revenue of the U.S. pet grooming industry?” You’d think these would be the easy part of a consulting interview. They’re not.

Most candidates stumble because they try to recall a number instead of building one. The interviewer doesn’t care whether your final answer is exactly right. They care about how you think through it, how you structure your assumptions, and whether you can do rough math on the spot without freezing up.

I’ve seen candidates with perfect case prep bomb a market sizing question simply because they panicked when the numbers got messy. And I’ve seen candidates with imperfect math skills absolutely nail it because their structure was airtight.

The Two Approaches That Actually Work

Top-Down Estimation

This is the most common approach, and for good reason. You start with a big number you’re fairly confident about and then narrow it down through logical filters.

Let’s say you’re asked to estimate the number of gas stations in the United States. You might start with the U.S. population of roughly 330 million people. Then you estimate how many people drive regularly, how often they fill up, how many cars each station can serve per day, and you work your way to a number.

The beauty of top-down is that it feels intuitive. You’re slicing a big pie into smaller pieces. The risk is that each assumption compounds, so if your first filter is off by a lot, everything downstream gets skewed.

Bottom-Up Estimation

This approach works in the opposite direction. You start with a small, observable unit and scale it up. For the gas station example, you might think about how many gas stations you’ve seen in your own city, estimate your city’s share of the national population, and then multiply.

Bottom-up tends to produce more grounded estimates because you’re anchoring to something you’ve actually observed. It’s particularly useful for questions about local services, retail, and consumer behavior.

Which Should You Pick?

Honestly, the best candidates can do both and sometimes even cross-check one against the other. If your top-down estimate says 120,000 gas stations and your bottom-up says 150,000, you’re probably in the right ballpark. That kind of triangulation impresses interviewers.

Building Your Assumption Toolkit

Here’s what separates good market sizing from great market sizing: having a mental library of useful anchors.

You don’t need to memorize obscure statistics. But you should know a handful of foundational numbers cold. The U.S. population. Average household size (about 2.5 people). Average household income (roughly $75,000). Life expectancy. Number of households. These base numbers let you derive almost anything else.

For instance, if someone asks about the U.S. diaper market, you can start with 330 million people, estimate roughly 4 million births per year, figure each child uses diapers for about 2.5 years, calculate daily usage, estimate price per diaper, and you’ll land somewhere reasonable.

The candidates who struggle are the ones who try to guess the final number directly. Don’t guess. Build.

Common Mistakes That Cost You Points

Getting Lost in Precision

I’ve watched candidates spend three minutes debating whether the average American drives 12,000 or 13,000 miles per year. It doesn’t matter. Round aggressively. Use 10,000 if it makes the math cleaner. Your interviewer wants to see speed and structure, not decimal-point accuracy.

Forgetting to State Your Assumptions

Every single assumption you make should be spoken out loud. “I’m going to assume the average household has 2.5 people” is music to an interviewer’s ears. It shows you know that your estimate depends on certain inputs and that you’re being transparent about which inputs you’re choosing.

Skipping the Sanity Check

Always take ten seconds at the end to gut-check your answer. If you’ve calculated that the U.S. has 2 million dentists, something went wrong. That’s roughly one dentist for every 165 people, which feels way too high. Catching your own mistakes is a huge signal of maturity and practical intelligence.

Not Segmenting When You Should

Some questions practically beg to be segmented. “What’s the market size for eyeglasses in the U.S.?” You could just estimate total population times some percentage that wears glasses. But a stronger answer segments by age group, since children, working adults, and seniors have very different rates of glasses usage and very different purchasing behavior.

Practice Strategy That Builds Real Skill

Don’t just practice market sizing problems from a book. Practice estimating things in your daily life. How many cups of coffee does this Starbucks sell per day? How much revenue does your local gym generate annually? What’s the total square footage of office space in your city’s downtown?

The goal isn’t to verify your answers. It’s to build the muscle of breaking big questions into small, answerable pieces. After a few weeks of this, you’ll walk into your interview and find that market sizing feels almost fun. You’ll have an instinct for which approach to use and which assumptions to make, and you’ll move through the math quickly and confidently.

That fluency is exactly what top firms are looking for. They don’t need you to be a human calculator. They need you to be someone who can make reasonable estimates quickly and explain your reasoning clearly. That’s the skill that translates directly to client work, where you’ll be making exactly these kinds of back-of-the-envelope calculations every single week.