Top-Down vs Bottom-Up Market Sizing Framework (2026)

The right market sizing approach is the one that anchors on the binding constraint. Top-down (demand-side) starts from total population or GDP and filters down. Bottom-up (supply-side) starts from one store, one driver, or one transaction and scales up. Top-down fits ~80% of consumer-market questions per Management Consulted. Bottom-up wins for B2B, infrastructure, and capacity-constrained markets. Stating your choice (and why) is what interviewers score before any arithmetic. This article focuses narrowly on the decision: when to use which approach, how to recognize the pattern in 30 seconds, and how to cross-check both estimates against each other for a higher score.

For the full methodology (clarify, segment, calculate, sanity-check, state implications) with reference data and 5 worked examples, see the market sizing step-by-step guide. This page assumes you already know the 5-step method and zooms in on approach selection.

TL;DR — What you need to know

• Top-down anchors on a population or GDP number, then filters by demographics, behavior, and price. Use for consumer markets where the binding constraint is demand.
• Bottom-up anchors on supply (number of stores, machines, doctors), then scales by throughput and price. Use for B2B, physical locations, and capacity-constrained markets.
• Decision rule: ask what is the binding constraint. If demand (pool of customers), go top-down. If supply (locations, capacity), go bottom-up.
• Cross-check: when time allows, run both. A 2 to 3x divergence signals a structural gap in one model. Reconciling the two estimates impresses interviewers (IGotAnOffer).
• Frequency: estimation appears in ~30 to 40% of MBB final-round and partner interviews (IGotAnOffer). Bain dedicates a full first-round session to it in many countries.

What is top-down market sizing?

Top-down (demand-side) starts with a large, known anchor (typically total population or GDP) and applies successive filters to narrow toward the target market. The mental model is a funnel: start wide, narrow by demographics, apply a penetration rate, then scale by volume or price.

When to use it: consumer markets, national or regional questions, any question where the end customer is an individual person or household. Per Management Consulted, this is the preferred approach for most consulting interview market sizing questions.

Template formula:

Market size = total population x % in target demographic x penetration rate x purchase frequency x average price

Example: US gym memberships

• US population: 330M
• Adults (18 to 70): ~200M (60%)
• Exercise regularly enough to consider a gym: ~40% = 80M
• Choose gym vs home/outdoor: ~45% = 36M members
• Average annual membership: ~$500/year
• Total: ~$18B (industry estimates put gym memberships at $30 to $35B including premium studios; our estimate is in range for traditional gyms only)

This connects to broader customer segmentation framework work, where the demographic and behavioral filters drive segmentation logic.

What is bottom-up market sizing?

Bottom-up (supply-side) starts from a single micro-unit (one store, one transaction, one day of operations) and scales up using the number of those units in the market. The logic flows from observed micro behavior to aggregate market size.

When to use it: B2B markets, physical-location businesses, infrastructure, capacity-constrained industries. Per a PrepLounge community discussion on top-down vs bottom-up, supply-side works best when clients have physical locations and you are measuring market share or unit counts.

Template formula:

Market size = number of supply units x average output per unit x price

Example: US coffee shops

• US coffee shop locations: ~38,000 (Starbucks ~16,000 US locations; chains and independents 2.5x that)
• Average daily transactions: ~180 (urban ~250, suburban ~150, rural ~100)
• Average ticket: $6.50 (coffee $4 to $5, ~40% food attachment)
• Daily revenue: $1,170. Operating days: 350.
• Total: ~$15.6B (industry data places dedicated coffee shops at $20 to $25B; our estimate is directionally correct)

How do I decide which approach to use?

Ask one question: what is the binding constraint in this market?

• If demand is the natural anchor (a defined population of customers), go top-down.
• If supply capacity defines the market (number of stores, drivers, beds, machines), go bottom-up.

State your choice before calculating. Per IGotAnOffer's market sizing guide, the narration is what interviewers are listening for: the strategic reasoning that precedes the math.

What is the proxy-based variant?

When there is no clean population anchor and supply is hard to count directly, use a proxy: an indirect metric you can estimate to reach the target.

The classic example is the Fermi estimation question on piano tuners in Chicago:

• Chicago population: ~2.7M, ~1M households
• % with a piano: ~5% = 50,000 pianos (proxy)
• Tuning frequency: once per year = 50,000 tunings
• Tuner throughput: 4 tunings/day x 250 days = 1,000 tunings/year
• Chicago piano tuners: 50,000 / 1,000 = ~50 tuners

The proxy here is the number of pianos, estimated from households and a reasonable ownership rate. Per PrepLounge market sizing basics, the proxy method is essential for markets where direct demand estimation is impossible.

How do I cross-check top-down with bottom-up?

The highest-leverage move in a market sizing answer is running both approaches on the same question, then reconciling. If the two estimates diverge by more than 2 to 3x, something is wrong in one of your formulas (often an excluded segment).

Worked cross-check: US rideshare market

Top-down (demand-side):

• US urban/suburban population: ~230M (70% of 330M)
• % using rideshare monthly: ~20% = 46M
• Rides per user per month: ~4 (light users 1 to 2, heavy 10+)
• Average fare: ~$18
• Annual market: 46M x 4 x 12 x $18 = ~$40B

Bottom-up (supply-side):

• US rideshare drivers: ~2M
• Rides per driver per week: ~30
• Operating weeks: 50
• Average fare: $18
• Annual market: 2M x 30 x 50 x $18 = ~$54B

Reconciliation: the estimates bracket each other ($40B vs $54B), so the true market is likely $40 to $50B. Uber's US gross bookings alone were ~$37B in 2023, validating directional accuracy. The gap likely reflects heavy users (business travelers, urban non-car-owners) that the top-down model under-weights.

This kind of bracketing is a calling card of strong case interview frameworks work. For the full step-by-step method that produces both estimates cleanly, see the market sizing step-by-step guide.

How do top-down and bottom-up compare on the same question?

The cleanest way to internalize the decision is to see both approaches run on a single market and observe where each one shines.

Question: US enterprise SaaS license market

Top-down (demand anchor: enterprise employees)

• US firms with 500+ employees: ~20,000 (US Census SUSB)
• Average employees per firm: ~5,000 (skewed distribution)
• Total enterprise employees: 100M
• SaaS licenses per employee: ~8 paid tools
• Total: 100M x 8 = ~800M licenses

Bottom-up (supply anchor: license vendors)

• ~50 dominant enterprise SaaS vendors (Microsoft, Salesforce, SAP, ServiceNow, etc.)
• Average paid US enterprise licenses per vendor: ~15M
• Total: 50 x 15M = ~750M licenses

Both arrive in the same range. Cross-check against revenue: at $150 to $250 per license per year, 750 to 800M licenses imply $115 to $200B revenue, aligned with Gartner estimates of US enterprise SaaS at $150 to $200B. Either approach works; running both gives you a confidence interval.

What are the most common approach-selection mistakes?

Five failure modes appear repeatedly:

1. Top-down when supply is the constraint. Sizing US gas stations from population x trips per year is the wrong anchor. Bottom-up (locations x throughput) gives a tighter answer.

2. Bottom-up when demand is the constraint. Sizing streaming music by counting vendors produces a noisy estimate. Top-down (adults x % using x ARPU) is cleaner.

3. Picking silently. Not narrating costs structure points. Say "I'm going top-down because demand is the constraint" before any math.

4. Running both with no time. Cross-checking takes ~2 extra minutes. With 4 minutes left, pick one approach and defend it. Don't half-finish two trees.

5. Proxies when a direct anchor exists. Proxies add noise; only use them when there is no clean population or supply anchor.

For broader sizing failure modes, see the step-by-step guide common mistakes. Drill the approach call live on our free drill set, or step into the full prep toolkit when you are ready for end-to-end practice.

How does approach selection connect to the rest of the case?

Market sizing feeds market entry framework (TAM/SAM/SOM go/no-go), profitability framework (contextualizing a revenue trend), customer segmentation framework (segment-level prioritization), and M&A case framework due diligence. The MECE principle applies to your tree: branches mutually exclusive, collectively exhaustive. For arithmetic speed, see mental math for case interviews and run timed math drills before interview week.

Can I test my approach-selection instincts?

Test yourself

1 / 3

Question 1 of 3

A candidate is asked: 'How large is the US market for yoga studios?' Which approach is most appropriate?

• ATop-down demand-side, anchoring on US adults who practice yoga
• BBottom-up supply-side, anchoring on number of yoga studios nationally
• CEither approach is valid; announce your choice and justify it
• DProxy-based, using gym membership data

Frequently Asked Questions

What is the difference between top-down and bottom-up market sizing?

Top-down starts from a large anchor (population or GDP) and applies filters to reach the target customer. Bottom-up starts from one supply unit (one store, one user, one transaction) and scales up. Top-down fits consumer markets ~80% of the time; bottom-up wins for B2B, infrastructure, and physical-location businesses.

When should I use top-down vs bottom-up?

Ask what the binding constraint is. If the constraint is the number of potential customers, go top-down. If supply capacity defines the market, go bottom-up. State your choice and justify before calculating.

Should I run both top-down and bottom-up?

Yes when time allows. Running both gives a cross-check. A 2 to 3x divergence signals a structural gap (often an excluded segment). Reconciling the two estimates is one of the highest-leverage moves in a market sizing answer.

What is a proxy-based market sizing approach?

A proxy approach uses an indirect metric (like the number of pianos) to reach the target (number of piano tuners). Use proxy when there is no clean population anchor and supply is hard to count directly. The classic example is the Fermi piano tuners question.

How does this article connect to the main market sizing guide?

This article focuses narrowly on the top-down vs bottom-up decision and the cross-check move. For the full step-by-step methodology (clarify, choose, segment, calculate, sanity-check, implications) with reference data and 5 worked examples, see the market sizing step-by-step guide. For a 20-question practice bank, see market sizing questions.

Sources (checked April 28, 2026)

IGotAnOffer market sizing guide, IGotAnOffer 21 market sizing questions, Management Consulted, PrepLounge basics, PrepLounge top-down vs bottom-up, US Census SUSB, Gartner press releases, McKinsey careers, BCG case prep.

Related guides

• Market sizing step-by-step guide (full methodology)
• Market sizing questions: 20 practice examples
• Mental math for case interviews
• MECE principle explained
• Case interview frameworks complete guide