
This tests your ability to break down a complex problem into manageable parts, make reasonable assumptions under uncertainty, and perform mental arithmetic, all core consulting skills for case interviews at McKinsey.
Start by clarifying the question and defining terms. Use a top-down or bottom-up framework, segment the population logically, make clear assumptions, show your math, and end with a sanity check. Practice mental math daily.
This is a market-sizing case, and the strongest response follows a clear structure rather than jumping straight to a number. First, clarify scope: confirm the given definition (premium equals PHP 150 or more per cup), confirm that 'annual market size' means total revenue rather than unit volume or shop count, and confirm the geography is Metro Manila only. Second, name the framework out loud: a supply-side, bottom-up approach, starting from the number of premium coffee outlets and multiplying by average transactions per shop per day and average ticket size, is usually cleaner here than a demand-side approach that tries to estimate individual consumption habits across Metro Manila's whole population. Third, walk through the math transparently with round, clearly labeled assumptions, for example: assume roughly 1,000 premium coffee outlets across Metro Manila's malls and business districts (an assumption to flag as needing validation against real store-count data), each doing about 150 transactions a day at an average ticket of PHP 200, operating 350 days a year; multiplying these gives an estimated annual market size. Fourth, sanity-check the result against a rough top-down cross-check, such as whether the implied per-capita premium coffee spend looks plausible for Metro Manila's income distribution, and present the answer as a range rather than false-precision point estimate, while flagging which assumption would most change the answer if it turned out to be wrong. The specific input numbers above are illustrative starting assumptions used to demonstrate the method, not verified market data.
Common mistake: blurting out a random number or getting stuck on mental math. Instead, state your approach first, then do the math out loud, and don’t be afraid to say “I’ll round to 1 million for simplicity.” Avoid saying “sorry” if your number seems off.
Situation
In a mock case interview for my application to McKinsey, I was asked to estimate the annual market size for premium coffee shops in Metro Manila.
Task
I needed to define the boundaries of 'premium coffee shops' and produce a structured, reasonable estimate using logical assumptions and basic mental math, all within a few minutes.
Action
I started by clarifying the definition: premium means at least PHP 150 per cup, and shops include chains like Starbucks and independent cafes. I then segmented the market by population: Metro Manila has roughly 13 million people. I assumed 60% are coffee drinkers (7.8 million). Among those, I estimated 40% (3.12 million) buy coffee out at least once a week. Of those, about 35% (1.09 million) choose premium options. I multiplied that by 52 weeks per year to get 56.68 million premium coffee purchases annually. I assumed an average spend per visit of PHP 200 (drink plus possible pastry). That gave an annual market size of approximately PHP 11.3 billion. I then sanity-checked: that seemed reasonable given the number of Starbucks outlets in Metro Manila (around 300) each doing probably PHP 30-40 million annually, which fits.
Result
I provided a clear, step-by-step estimate of around PHP 11 billion, showing my assumptions and calculations aloud. The interviewer nodded and asked me to discuss one key sensitivity, which I identified as the frequency assumption.
Market-sizing questions test structured thinking, not exact numbers: clearly state each assumption and walk through your math step by step.
Write your own answer, then get instant AI feedback graded against:
Get AI feedback on your answer — free.
3 free AI-graded answers + 1 free mock interview, no card needed.
Sign Up FreeAlready have an account? Log in
Sign in to join the conversation.
No answers shared yet — be the first to show how you'd approach this.