The interviewer is not grading whether you once read a polished solution about pizza or school buses. They are listening for whether your first numbers sound like someone who has lived with real demography for a week, not someone who invents 400 million Americans mid-sentence.
Thesis: the scarce resource in market-sizing prep is not another bank of worked answers. It is a short, dated set of primary-source anchors you can defend under pressure, then a handful of official firm cases that show how sizing sits inside a real interview, then active reps. Everything else is optional reading.
After this piece you will know which statistical tables to open first, fifteen numbers worth committing to memory as of September 2026, which firm pages actually contain sizing arithmetic, and how CoachNed's drill and tools fit without pretending they replace the data.
For the method itself, use the sibling guide on market sizing step by step. For prompt variety without a competitor roundup, see guesstimate interview questions.
Rank the primary sources, not the prep blogs
Rank sources by how often a live case starts from their numbers, and by how durable the table is. Open these after you attempt an estimate, not instead of attempting one. Exact lookup during the interview is rare; memorized anchors and a clean chain of logic are what get scored.
| Rank | Source | What to open first | Why it belongs high |
|---|---|---|---|
| 1 | U.S. Census Bureau Vintage 2025 population estimates plus ACS income/household tables | Population (July 1 vintage), households, persons per household, median household income | Most US consumer and B2C prompts start here |
| 2 | Eurostat population news / Statistics Explained | Population on 1 January for EU-27; country tables for Germany, France, Italy | Default for EU market and pan-European cases |
| 3 | ONS mid-year population estimates | UK mid-year total; England / Scotland / Wales / NI when the city is British | London and UK-office interviews lean on this |
| 4 | World Bank Open Data — SP.POP.TOTL, NY.GDP.MKTP.CD | World, China, India population; world and US GDP (current US$) | Cross-border, emerging-market, and "TAM of X globally" prompts |
| 5 | BLS Employment Situation (CPS Table A) | Civilian labor force, employed, unemployment rate, participation rate | Workforce, employer, and "number of workers who…" cases |
| 6 | National statistics offices (INSEE, Destatis, Istat, Statistics Canada, ABS, …) | Country population, households, labor force for that geography | Use when the prompt is explicitly France, Germany, Canada, Australia, etc. |
Census sits first because US cases dominate most MBA and undergrad pipelines outside Europe. Eurostat and ONS jump ahead of World Bank when your office slate is London, Paris, or Munich. BLS is fifth only because pure consumer sizing often never needs employment; when the prompt is "how many truck drivers" or "how many nurses," it becomes first.
Fifteen anchors to memorize (as of 2026-09-24)
Memorize rounded interview versions. Keep the precise official figure in a note for debriefs. If a number moves, the rounded form usually survives a recruiting season.
| # | Anchor (say this) | Official figure | Source / table |
|---|---|---|---|
| 1 | US population ~342 million | 341.8 million (July 1, 2025, V2025) | Census Vintage 2025 press release |
| 2 | US households ~129 million | ~129 million (ACS 2020–2024) | Census ACS / household estimates |
| 3 | ~2.5 people per US household | ~2.5 persons per household | Census ACS household estimates |
| 4 | US median household income ~$82k | $81,604 (ACS 2024, 1-year) | Census ACS brief ACSBR-025 |
| 5 | US civilian labor force ~170 million | 169.8 million (Aug 2026, sa) | BLS Employment Situation, Table A |
| 6 | US employed ~163 million | 162.7 million (Aug 2026, sa) | BLS Table A |
| 7 | US unemployment rate ~4% | 4.1% (Aug 2026, sa) | BLS Table A |
| 8 | EU population ~450 million | 450.4 million (1 Jan 2025) | Eurostat population news |
| 9 | Germany ~84 million | 83.6 million (1 Jan 2025) | Eurostat |
| 10 | UK population ~69.5 million | 69,487,000 (provisional mid-2025) | ONS UKPOP series |
| 11 | World population ~8.2 billion | 8.22 billion (2025 midyear) | World Bank SP.POP.TOTL |
| 12 | India ~1.46 billion | ~1.46 billion (2025) | World Bank SP.POP.TOTL |
| 13 | China ~1.41 billion | ~1.41 billion (2025) | World Bank SP.POP.TOTL |
| 14 | World GDP ~$118 trillion | $118.35 trillion (2025, current US$) | World Bank NY.GDP.MKTP.CD |
| 15 | US GDP ~$31 trillion | $30.77 trillion (2025, current US$) | World Bank NY.GDP.MKTP.CD |
That set covers the openings of most geographic scopes you will face: US, UK, EU, China, India, and "global." Age splits, urban share, and category prices still require judgment; the population and income floor should not.
Ned's rule. If your opening population is more than about 10 percent off the primary figure for that geography, I stop trusting the tree until you correct it. Interviewers forgive soft frequency and price assumptions. They do not forgive inventing a country.
A worked check: annual spend on takeaway coffee in Cambridge
Use Bain's public Coffee Shop Co. setup as the prompt shape, then run original arithmetic with anchors you own. Scope: Cambridge, England residents only; cups per year; retail price to consumer.
- UK population ~69.5 million (ONS). Cambridge is a university city; treat resident population as ~145,000 for the estimate (a figure you would confirm or accept from the interviewer in a live case).
- Share who buy takeaway coffee weekly: say 40 percent → 58,000 buyers.
- Cups per buyer per week: 3 → 174,000 cups/week.
- Weeks per year: 50 (allow holidays) → 8.7 million cups/year.
- Price £3 → market ~£26 million/year of takeaway coffee among residents.
Sanity check against Bain's published Coffee Shop Co. solution path: their worked answer lands near 7 million cups for the resident segment they use, with break-even near 205,000 cups at £3 / £1 contribution. Your estimate above is deliberately on the high side of frequency; if the interviewer pushes "heavy users only," cut cups per week from 3 to 2 and you drop to about £17 million. The point of the exercise is not matching Bain's internal assumptions. It is showing you can start from ONS, state every lever, and flex one lever without rebuilding the tree.
Reproduce the arithmetic on paper once. Then change only the price to £2.50 and restate the market in one sentence.
Firm pages that actually show sizing
Do not collect anonymous "21 market sizing questions" lists. Use the firms' own published cases, which embed estimation inside a client problem the way the interview usually does.
| Firm page | What you practice | What it is not |
|---|---|---|
| McKinsey Electro-Light | Share, break-even volume, and market context for a US beverage launch (price $2 per 16-oz bottle, $40M fixed cost, $1.90 variable) | A standalone "how many X in the US" guesstimate |
| McKinsey Diconsa | Aggregation of monthly savings across Mexican rural families (pesos, percentages of population by distance to a bank) | US Census arithmetic |
| Bain Coffee Shop Co. | Explicit resident market size for Cambridge coffee, then break-even cups | A global TAM |
| BCG interview process / case prep and the public recruiting brochure examples | Structured problem-solving with market and share logic (e.g., household and ADSL-style share math in older brochure cases) | A dedicated sizing question bank |
McKinsey's interviewing hub presents problem-solving cases, not a separate "market sizing round." Bain's Coffee Shop Co. is the clearest official standalone resident-sizing drill among MBB public materials. BCG's careers guidance stresses structuring, clarifying questions, calculation accuracy, and synthesis more than memorized frameworks. Treat firm pages as calibration for how sizing appears in a conversation, not as a substitute for primary anchors.
CoachNed is not affiliated with McKinsey, BCG, Bain, Deloitte, or any firm named above.
CoachNed practice layer (plain description)
Once anchors and one firm case are in muscle memory, you need fresh prompts and a score, not another static PDF.
/drills/market-sizing— timed prompts that force clarify → approach → assumptions → math → sanity check. Use this when you already know the method from the step-by-step guide and need reps under the clock./tools/market-size-calculator— a structured calculator for top-down or bottom-up chains. Use it in debrief to compare your spoken estimate to a transparent formula, not as a crutch during a live mock./tools/market-sizing-questions— a bank of prompts you can filter and attempt before revealing any solution path. Write or speak the answer first; then check.
Everything on CoachNed is open for seven days with no card; then $120 for a recruiting season or $49 a month. The free forever quick-math path sits at /drills/quick-math if arithmetic, not assumptions, is the leak.
Practice this today (45 minutes)
- Write the fifteen anchors from memory. Misses go on a one-page card you review before every mock.
- Attempt Bain's Coffee Shop Co. out loud without looking at their solution, using UK population from ONS as your first number.
- Run one CoachNed market-sizing drill. In the debrief, tag every number as "primary anchor," "stated assumption," or "guess I should have flagged."
- Optionally open Electro-Light and solve only the break-even share question, narrating units aloud.
If you freeze on structure rather than anchors, switch to a structure drill after one more sizing rep. If you know the tree but stumble on multiplication, isolate math for twenty minutes, then return.
Frequently asked questions
Should I memorize exact Census figures or rounded ones?
Rounded interview figures. Say "about 340 million" or "roughly 342 million," then move. Exact seven-digit recall does not earn points; a wrong order of magnitude loses them.
Do McKinsey interviews still ask standalone market sizing?
McKinsey's published interviewing materials center on full problem-solving cases such as Electro-Light and Diconsa, where quantitative estimation appears inside the case. Treat standalone "how many gas stations" prompts as more common at firms that publish explicit sizing practice (Bain's Coffee Shop Co. is the clear public example) and as possible but not the default McKinsey format.
How many primary anchors are enough?
Fifteen geography and macro anchors cover most openings. Add two or three category-specific prices (a cup of coffee, a movie ticket, a monthly SaaS seat) for your target office's common industries. More than that becomes a trivia hobby.
When should I look up data during prep versus inventing an assumption?
Look up after the attempt. During the attempt, state the assumption, mark it as soft, and proceed. Looking up mid-practice trains you to stall; looking up in debrief trains calibration.
Is a market-sizing calculator cheating?
In a live interview, yes if you would use it on the call. In prep, use /tools/market-size-calculator only after you have spoken an answer, to see where your chain diverged. The scored skill is the spoken chain.
Sources
- U.S. Census Bureau: U.S. Population Growth Slows (Vintage 2025) — US population 341.8 million as of July 1, 2025. Checked 2026-09-24.
- Census ACS brief: Household Income in States and Metropolitan Areas: 2024 — median household income $81,604 (ACS 2024). Checked 2026-09-24.
- Eurostat: EU population increases for the 4th consecutive year — EU 450.4 million and Germany 83.6 million on 1 January 2025. Checked 2026-09-24.
- ONS: United Kingdom population mid-year estimate (UKPOP) — UK mid-2025 provisional 69,487,000. Checked 2026-09-24.
- World Bank Open Data: World — world population and GDP. Checked 2026-09-24.
- World Bank indicator NY.GDP.MKTP.CD — world, US, China, India GDP (current US$). Checked 2026-09-24.
- BLS Employment Situation Summary Table A — US labor force, employment, unemployment rate. Checked 2026-09-24.
- McKinsey: Electro-Light practice case — embedded launch economics and market share. Checked 2026-09-24.
- McKinsey: Diconsa practice case — aggregation of rural benefit-collection savings. Checked 2026-09-24.
- Bain: Coffee Shop Co. practice case — Cambridge resident coffee market and break-even. Checked 2026-09-24.
- BCG: Case Interview Preparation — official guidance on case skills. Checked 2026-09-24.