Market Sizing Questions: A Case Interview Guide

Market sizing questions trip up candidates who rush straight into the math. Interviewers are not primarily testing whether you can calculate quickly. They want to see how you structure a problem, make assumptions, and communicate your reasoning under pressure.

Strong candidates slow down first. They clarify the prompt, outline a framework, state assumptions clearly, and sanity-check the result before landing on an answer. The math is usually simple. The thinking is what matters. For more on the broader interview pattern, see our case interview prep guide.

In this guide, we’ll break down what market sizing questions actually evaluate, walk through a four-step framework that works for nearly any prompt, and solve a complete example step by step.

What market sizing questions actually test

A market sizing question asks you to estimate the size of something, usually a market, with no data and no calculator, expressed in either volume or dollars. What makes them hard isn't the math. It's the ambiguity. You have to define what you're sizing, decide how to break it down, and pull plausible numbers from thin air while talking through your reasoning out loud.

Interviewers care more about the path than the number. They're watching three things: whether you ask clarifying questions before solving the wrong problem, whether you build a structure someone can follow, and whether your assumptions hold up to a sanity check. Market sizing shows up in case interviews constantly because consultants run back-of-the-envelope estimates with clients all the time. If you can't break an unfamiliar problem into pieces in an interview, no one's trusting you to do it on a project.

The four-step framework for any market sizing question

Almost every workable answer to a market sizing question follows the same four steps. Get these in your bones and you'll have a default approach for any prompt the interviewer throws at you.

Step 1: Clarify the question. 

Before you write a single number, ask one to three clarifying questions. Are we estimating volume or value. What geography. What timeframe. What customer segment. The lipstick market in the U.S. last year is a different question from the global premium matte lipstick market over five years. Confirm the scope and restate the question in your own words. This buys you ten to fifteen seconds of thinking time and signals structured thinking before you say anything else.

Step 2: Lay out your structure. 

Tell the interviewer how you plan to solve it before you do any math. A simple verbal sketch works: "I'll size this by estimating the U.S. population, narrowing to the relevant segment, then multiplying by purchases per person per year." Drawing an issue tree on paper or whiteboard is even better when you can. Get the interviewer's nod on the approach before you commit to numbers. If they want a different angle, you can adjust now, not after you've calculated for three minutes down the wrong path.

Step 3: Make assumptions and calculate. 

Use round numbers. The U.S. population is 330 million; round to 300 million. People drink coffee five days a week; round to one cup per day per coffee drinker. State each assumption out loud as you make it. If a calculation gives you a messy number, round before you use it in the next step. The point is not to be precise. The point is to be transparent and to get to a defensible final number quickly.

Step 4: Sanity-check the answer. 

Once you have the final number, pause. Does it pass the smell test. If you estimated the U.S. coffee market at $50 billion and it intuitively feels right (Starbucks alone runs in the tens of billions), say so. If you got $5 trillion, you missed a decimal somewhere; back-track and find it. End with a sentence that says what the number means: "So my estimate is around $50 billion, which feels reasonable given how many people drink coffee daily." That close shows judgment, not just arithmetic.

The prompts will change, but this structure rarely does. Once you can clarify, frame, calculate, and sanity-check without panicking, market sizing questions start to feel far more predictable and manageable.

Top-down vs bottom-up: which approach to use

Most market size questions can be solved two ways. Knowing which to pick saves time and makes your reasoning cleaner.

Top-down starts with a large number, usually population or total industry size, and narrows it with filters: total U.S. population, percent who own dogs, average annual spend per dog. It works best for broad consumer markets where you have a reasonable anchor for the top line.

Bottom-up starts with a unit-level estimate and rolls it up: coffee shops in NYC, cups per shop per day, days per year. It works best when the market is local, the unit economics are knowable, or the customer base is a single dense segment.

When in doubt, use top-down. Most prompts are consumer markets and top-down is the cleaner fit. But the strongest candidates pick the approach that fits the prompt and say why. "I'll go bottom-up because the customer base is a single hospital system and we can estimate per-bed consumption directly" is a credible signal that you understand both 

Market sizing questions with solutions: a worked example

The fastest way to internalize the framework is to watch it run on a real prompt. Below is the kind of full walkthrough you should be able to produce in three to five minutes once you've practiced enough.

WORKED EXAMPLE: "HOW MANY CUPS OF COFFEE ARE SOLD IN NYC EACH DAY?"

‍More market sizing question examples to practice

Fluency comes from repetition. Practice the framework across different prompts and solve under time pressure, not just by reading examples.

Here are market sizing questions with answers worth working through, sourced from actual McKinsey, BCG, and Bain interview reports:

• How many smartphones are sold in India each year

• What is the U.S. market size for residential light bulbs

• How many croissants are sold per day in Paris

• What's the annual revenue of takeaway coffee in New York City

• How many flat-screen TVs were sold in the U.S. last year

• Estimate the global market for electric scooters

• What's the size of the U.S. pet food market

• How many gas stations are in downtown Dallas

• What's the annual market for cell phone services in NYC

• How many cars are sold in Mexico each year

Time yourself. Aim for three to five minutes per question once the framework is automatic. Practicing market sizing questions and answers under a clock is the only way to make the structure feel natural when you're actually in the room.

Common mistakes to avoid

Three mistakes show up in most weak market sizing answers, and all three are fixable once you know what to watch for.

• Skipping the clarifying questions. Candidates who jump straight into math often answer the wrong question. Sizing the global lipstick market when the interviewer wanted just gloss in the U.S. is a fail no matter how clean the arithmetic is. Always confirm scope, units, geography, and timeframe before you start.
• Hiding the assumptions. If you do all the math in your head and announce only the final number, the interviewer cannot follow your logic. They cannot reward what they cannot see. State every assumption out loud as you go: "I'm assuming the U.S. has 300 million people, of whom roughly forty percent drink coffee daily." Make the work visible.
• Forgetting to sanity-check. Candidates who get to a number and stop are skipping the most important step. A sanity check is what separates a candidate who can do math from one who can do business judgment. If your estimate of U.S. annual diaper sales is $2 trillion, something is wrong, and saying "that feels too high, let me retrace" is far better than defending the wrong answer.

A fourth mistake worth flagging: trying to sound impressive by using clever methods or precision math. Interviewers are not looking for a savant. They're looking for someone they could put in front of a client tomorrow. Clean, simple, well-explained beats elaborate every time.

How to practice market sizing under real interview conditions

Reading frameworks doesn't build the skill. Reps under pressure do. Most candidates can write a clean four-step answer alone at a desk with no clock running. The same candidates fall apart when an interviewer is staring at them, the prompt is unfamiliar, and they have ninety seconds to start talking.

Do this instead:

1. Run timed reps, five per week. Pick a prompt, set a five-minute clock, and talk through it out loud. Silence doesn't count as practice.

2. Practice with pushback. Talking to a wall is not the same as defending your assumptions to someone pushing back in real time. Cook'd AI runs interview simulations the way a McKinsey or Bain interviewer would, then gives you feedback on structure, assumption quality, and whether your sanity check landed.

3. Track where you break down. If clarifying questions are weak, drill that first. If your sanity check feels rushed, work on closing strong. The framework is the same for everyone. The reps are what separate candidates who walk out confident from the ones replaying the prompt in their head all night.

The candidates who get offers aren't the ones who memorized the best framework — they're the ones who ran it enough times that it became instinct.

Stop reading. Start practicing with Cook’d AI.

Pick a prompt from this guide, set a five-minute clock, and talk through it out loud right now. Most candidates spend weeks reading about market sizing and never once defend an assumption to someone pushing back in real time. That gap is exactly what breaks them in the room. The framework only works if you've run it enough times to trust it under pressure.

Cook'd AI runs interview simulations the way McKinsey, BCG, and Bain interviewers actually do it: unfamiliar prompts, real pushback, and feedback on whether your structure held up. No scheduling, no waiting, no friendly roommate who lets your weak assumptions slide. 

Just realistic reps until the framework is instinct. Your next case round is closer than you think, so start today with Cook’d.