Market Sizing and Research Deep Dive for Case Competitions

Good case teams do not just present numbers. They present decisions supported by numbers that are coherent, transparent, and actionable. This guide covers how to build a full market-sizing and research workflow that holds up in judging panels and Q&A.

If you want to apply these techniques against live deadlines, start by selecting a target brief from current CaseCrest competitions. A real prompt exposes weaknesses in your process faster than any generic drill.

1) Why market sizing fails in student teams

Market sizing usually fails for one of four reasons:

• the team answers the wrong market question
• the numerator and denominator come from inconsistent definitions
• assumptions are hidden or contradictory
• outputs are not linked to the recommendation

A polished chart does not solve any of these issues. Only process discipline does.

Before calculating anything, define the decision objective in one sentence. Then define what market variable directly informs that decision. For instance:

• If choosing expansion geography, you need addressable demand and entry friction.
• If choosing channel strategy, you need acquisition economics by channel.
• If choosing product mix, you need segment willingness-to-pay and margin structure.

Your market-sizing output should always map to a decision variable.

2) Build a question hierarchy before you build a model

Use a three-level question hierarchy.

Level 1: strategic question

• “Should the client enter segment X over segment Y?”

Level 2: analytical questions

• “What is segment size?”
• “What is segment growth?”
• “What is feasible share capture?”
• “What are unit economics by segment?”

Level 3: evidence questions

• “How many potential buyers exist?”
• “What frequency and spend assumptions apply?”
• “What friction reduces conversion?”

When teams skip hierarchy, they collect data fragments that cannot be assembled into a defensible answer.

3) Choose top-down, bottom-up, or hybrid intentionally

Top-down sizing starts from macro totals and applies filters. Bottom-up sizing starts from unit behavior and scales up. Hybrid combines both and reconciles differences.

Use top-down when:

• macro data quality is high
• market structure is stable
• segmentation logic is clear

Use bottom-up when:

• behavior-level assumptions are key
• market boundaries are unclear
• you need scenario sensitivity

Use hybrid when:

• stakes are high
• data confidence varies by layer
• judges are likely to test credibility

A hybrid approach is usually best in competition finals because it provides both speed and defensibility.

4) Define TAM, SAM, and SOM correctly

Most teams misuse these terms. Keep definitions strict:

• TAM: total available market under broad category definition
• SAM: serviceable available market within practical target boundaries
• SOM: serviceable obtainable market under realistic execution constraints

Your recommendation should never rely on TAM directly. Strategy decisions should be linked to SAM and SOM under explicit assumptions.

Also define time horizon:

• current annual market
• three-year projection
• first-year capture target

Judges care about transition logic, not just static market snapshots.

5) Assumption design principles

Every assumption should pass three tests:

• traceable: source or rationale is visible
• plausible: consistent with context
• impactful: materially affects output

Assumptions that do not materially affect the decision can be simplified. Assumptions that materially affect the decision should be stress-tested.

Use assumption tables with columns:

• variable name
• base value
• confidence level
• source type
• sensitivity impact

This turns your model into a decision instrument rather than a black box.

6) Unit economics in market sizing

Market size without economics is incomplete. Build at least one unit economic lens:

• revenue per customer
• contribution margin per customer
• acquisition cost by channel
• payback period

If a large market has poor unit economics, recommendation viability is weak. If a smaller market has superior economics and execution fit, it may dominate strategically.

Tie market potential to economic quality:

• “Large and profitable”
• “Large but low margin”
• “Small but high margin”
• “Small and unattractive”

This matrix helps judges see why your recommendation is selective rather than simplistic.

7) Growth modeling and scenario bands

One-point forecasts are fragile. Build scenario bands.

At minimum, use:

• conservative scenario
• base scenario
• accelerated scenario

For each scenario, vary only key drivers:

• adoption rate
• conversion rate
• retention
• pricing realization
• channel cost

Then show recommendation stability. If your strategy only works in the accelerated case, it is risky. If it remains viable in conservative conditions, confidence increases.

8) Source triangulation method

Do not rely on a single source for critical claims. Triangulate with at least two independent perspectives:

• industry reports
• regulatory filings
• company disclosures
• benchmark datasets
• credible operator interviews or proxy insights

For each critical metric, annotate:

• source A estimate
• source B estimate
• chosen estimate and reason

This signals analytical maturity and reduces vulnerability in Q&A.

9) Handling sparse or conflicting data

Student teams often freeze when data is incomplete. Instead, use bounded estimation:

• create upper and lower bounds
• explain why each bound is reasonable
• position base case within that band

If sources conflict, do not hide it. Show conflict explicitly and justify your selection logic.

A sentence like “Sources vary from 4% to 9% CAGR; we use 6% base with 4% downside and 8% upside due to category maturity and channel concentration” demonstrates control over uncertainty.

10) Segmentation that actually informs strategy

Avoid demographic-only segmentation unless the prompt requires it. Prefer behavior and economics-based segments:

• need intensity
• willingness to pay
• acquisition channel response
• retention propensity
• cost-to-serve profile

Then map strategy to segment priorities:

• primary segment for launch
• secondary segment for expansion
• excluded segment with rationale

This makes your go-to-market recommendation actionable.

11) Geography and channel overlays

Many cases require market selection across geographies or channels. Build overlays rather than separate disconnected analyses.

For geography, compare:

• demand depth
• competitive intensity
• regulatory friction
• logistics complexity
• talent availability

For channels, compare:

• CAC
• conversion quality
• speed to scale
• operational dependency

Then produce a prioritized matrix. Recommendation should specify sequence, not just target.

12) Model transparency and auditability

Judges value clarity more than sophistication. Your model should be auditable quickly.

Rules:

• separate inputs, calculations, outputs
• avoid hard-coded values inside formulas
• label units for every variable
• include a assumptions tab with source notes
• highlight key sensitivities

In slides, you do not show all formulas. You show model logic and key drivers. Keep detailed structure in backup.

13) Converting analysis into narrative

Analysis wins only when translated into story. Use this structure:

• What is attractive in the market?
• What is difficult in the market?
• Where is the profitable entry point?
• Why is this option superior now?
• How do we execute with controlled risk?

Each section should reference explicit evidence from your sizing work.

Avoid data dumping. Every chart should answer a specific question that advances the recommendation.

14) Common judge challenges and prepared responses

Prepare for recurring challenges:

• “Why this growth assumption?”
• “Why this share capture target?”
• “What if CAC rises?”
• “What if competitor retaliation occurs?”
• “What capabilities are missing?”

Prepare concise responses with one evidence point and one mitigation.

Example response pattern:

• direct answer
• evidence
• risk acknowledgement
• mitigation action

This pattern prevents rambling and improves perceived command.

15) Time allocation under competition constraints

Do not spend all your time modeling. Use a balanced allocation:

• 20% problem framing and hypothesis setup
• 35% research and sizing
• 20% synthesis and recommendation design
• 15% deck construction
• 10% rehearsal and Q&A prep

If your team enters final rehearsal without a tested recommendation, reallocate immediately. Presentation polish cannot rescue weak logic.

16) Quality checklist before submission

Run this final checklist:

• Does every major claim have source support?
• Are TAM, SAM, SOM definitions consistent?
• Are units and periods consistent across charts?
• Are scenario drivers explicit?
• Is recommendation robust in downside case?
• Is implementation linked to market realities?

If any answer is no, fix that gap first.

17) Reusable templates your team should maintain

Maintain shared templates for:

• market sizing input sheet
• assumption confidence matrix
• scenario sensitivity table
• segment prioritization matrix
• channel economics comparison

Template reuse improves speed and reduces errors. Over time, your team quality improves through process, not luck.

18) Apply this on live rounds now

Pick a current brief and execute this workflow from setup to final. Use the public competition feed to select one with timelines that force discipline.

When your sizing process is coherent, your recommendation becomes harder to attack and easier to trust. That is the difference between a decent submission and a finals-ready one.