Unstructured Case Interview: How to Solve Open-Ended and Non-Standard Cases (2026)

Every McKinsey candidate prepares extensively for profitability cases, market entry cases, and growth cases. Then they get to the final round and encounter: "If you were the CEO of a mid-sized city, how would you improve it?"

No framework applies. The candidates who pass are the ones who stay calm, structure an approach from first principles, and reason clearly through uncertainty. The ones who fail are frozen waiting for a familiar structure that isn't coming.

What Makes a Case "Unstructured"

The difference isn't subtle:

Interviewers use unstructured cases in later rounds to see how you think when you can't fall back on preparation—which is a reasonable proxy for how you'll perform when a client gives you a messy, novel problem with no precedent.

The 5 Most Common Unstructured Case Types

Type 1: Open Strategy Problems

Example: "Our client is a century-old encyclopedia company. It's 2025. What should they do?"

There's no right framework. You must impose your own structure:

1. Current state: What does the business look like today? What assets does it have (brand, content, distribution)?
2. Core problem: Why is the business struggling? (Digital disruption of physical media)
3. Strategic options: Content licensing, digital platform pivot, education consulting, brand licensing
4. Evaluation criteria: Which option leverages existing strengths? Which has the fastest path to viability?
5. Recommendation: Explicit choice with explicit reasoning

The key: you supply all of this structure yourself. The interviewer won't tell you which framework to use because there isn't one.

Type 2: Brain Teasers and Logic Problems

Example: "You have 9 balls. 8 weigh the same; 1 is heavier. You have a balance scale. What is the minimum number of weighings to find the heavier ball?"

Answer: 2 weighings.

• Weighing 1: Put 3 balls on each side of the scale. Leave 3 aside.
  • If one side is heavier, the heavy ball is among those 3.
  • If balanced, the heavy ball is among the 3 you left aside.
• Weighing 2: Take the group of 3 containing the heavy ball. Put 1 on each side, leave 1 aside.
  • If one side is heavier, that's the ball. If balanced, it's the one you left aside.

The interviewer isn't asking you to know the answer—they're watching how you reason through the problem aloud. Speak your logic explicitly: "I'm trying to eliminate the maximum number of possibilities with each weighing. If I split 9 into groups of 3, one weighing tells me which third of the balls contains the heavy one..."

Type 3: Estimation Without Clear Anchors

Standard market sizing has clear anchors (US population, age distribution, etc.). Brain-teaser estimation is harder:

Example: "How many piano tuners are in Chicago?"

First-principles approach:

1. Chicago population: ~2.7M people
2. Average household size: ~2.5 people → ~1.1M households
3. % of households with a piano: ~1.5% (rare urban amenity) → ~16,500 pianos
4. Tuning frequency: pianos need tuning 1–2 times/year → ~25,000 tunings/year
5. Tuner productivity: one tuner can do ~4 tunings/day × 250 working days = ~1,000 tunings/year
6. Chicago piano tuners: 25,000 ÷ 1,000 = ~25 tuners

(Actual estimate: roughly 50–60 according to piano tuner associations—our estimate is in the right ballpark.)

Type 4: Business Design Problems

Example: "Design a new product for a traditionally non-tech retailer to reach Gen Z consumers."

No standard framework. Build your own:

1. User insight: What do Gen Z consumers actually want from this retailer? (Don't assume—ask clarifying questions)
2. Technology scan: What technology makes new solutions possible? (AR/try-on, social commerce, creator partnerships)
3. Business model: How does the product generate revenue or retain customers?
4. Feasibility: What capabilities does the retailer need? What partnerships?
5. Risk: What could fail? What's the minimum viable test?

Type 5: Judgment / Leadership Dilemmas

Example: "You're a junior consultant on an engagement. You discover that your partner has made a recommendation to the client that you believe is wrong. What do you do?"

This isn't a business case—it's a values and judgment test.

Strong answer structure:

1. Don't assume malice: There may be context you don't have
2. Build your case: Document specifically why you think the recommendation is wrong
3. Raise it constructively: Bring the concern to the manager (not directly to the client, not directly to the partner) with your analysis
4. Escalate if unresolved: If the manager dismisses it and you still believe it's serious, bring it to the partner—respectfully and with evidence
5. Never compromise your integrity: If the firm ignores clear client harm, that's a firm-level problem

What the interviewer is evaluating: Do you speak up? Do you do it constructively? Do you understand hierarchy? Are you driven by evidence, not ego?

The First-Principles Framework for Any Unstructured Case

When no framework applies, use this 5-step approach:

Framework

First-Principles Approach for Unstructured Cases

1. 01

   1. Clarify the core question

   Before anything else: what is the fundamental question this case is asking? Restate it in your own words. If it's ambiguous, ask one clarifying question.

2. 02

   2. Map the key factors

   What are the 3-5 most important dimensions you need to understand to answer the question? These become your structure. They should be MECE where possible.

3. 03

   3. State your hypothesis early

   Give a direction—'my initial hypothesis is X because...' This shows confident thinking and gives the interviewer something to react to.

4. 04

   4. Explore one factor at a time

   Go deep on each dimension in turn. If new information changes your hypothesis, update explicitly: 'Given that data, I'd revise my earlier hypothesis to...' Show flexibility without being a pushover.

5. 05

   5. Synthesize a recommendation

   End with a clear recommendation, its main rationale, the key uncertainty, and what you'd do to resolve it. Never leave an unstructured case open-ended.

Structural Techniques That Work on Any Case

MECE bucketing: Even without a framework, you can create MECE categories. For "improve a city": Infrastructure, Economy, Quality of Life, Governance. These may not be the perfect buckets, but they're MECE and cover the space.

2x2 matrices: When you have two competing considerations, build a quick matrix. For "design a new product": high effort vs. low effort × high impact vs. low impact. Focus on high impact, lower effort.

The 5 Whys: For root cause problems, asking "why" five times gets you from symptom to root cause. "Revenue is declining → why? → prices are falling → why? → competitors entered → why? → patent expired → why? → etc."

Analogous industries: "This problem looks like what happened to newspapers in the 2000s" is a powerful framing device—it helps you import pattern knowledge from adjacent situations.

Preparation Strategy: Building the Muscle

You can't prepare for unstructured cases by memorizing frameworks. You prepare by developing mental flexibility.

Checklist

Execution checklist

• Practice 10+ estimation cases (Fermi problems) without looking up answers first

  Estimation under uncertainty is the core skill—builds the habit of structured first-principles reasoning

• Read McKinsey and Harvard Business Review articles weekly

  Exposure to diverse business contexts gives you the pattern recognition to find analogies quickly

• Practice articulating structure under time pressure

  30 seconds of structured thinking beats 5 minutes of unstructured talking—practice the pause

• Do 3-5 deliberately unconventional cases from PrepLounge

  PrepLounge's 'unusual' case category is specifically designed for unstructured prep

• Have someone give you random business problems with a 60-second response requirement

  Forces hypothesis-first thinking instead of framework-first thinking

Common Mistakes Under Uncertainty

Connect to Core Case Skills

Unstructured cases test the same underlying skills as structured ones—just without scaffolding:

• MECE principle for creating your own structure
• Hypothesis-driven consulting for leading with a direction
• Case interview synthesis for delivering a recommendation under uncertainty
• Case interview communication tips for thinking out loud clearly
• Case interview types for the complete taxonomy of case formats

Test Your Knowledge

Test yourself

1 / 3

Question 1 of 3

You receive an unstructured case: 'How would you monetize a free park?' Your first action should be:

• AImmediately list every revenue option you can think of
• BAsk for time to structure your approach, then identify 3-4 MECE dimensions before answering
• CAsk the interviewer which framework they'd like you to use
• DDecline to answer until you have more information

Sources and Further Reading (checked March 25, 2026)

• PrepLounge — Unstructured Case Interviews: preplounge.com/en/articles/unstructured-case-interview
• Hacking the Case Interview — Unstructured Cases: hackingthecaseinterview.com/pages/unstructured-case-interviews
• Management Consulted — Unstructured Case Interview: managementconsulted.com/unstructured-case-interview
• Caseinterview.com — Weird Case Interview Questions: caseinterview.com/weird-case-interview-questions
• Strategy Case — Brainstorming in Case Interviews: strategycase.com/brainstorming-in-a-case-interview