How to prepare for a case interview: frameworks that actually work

You're sitting across from a McKinsey interviewer. They describe a regional airline whose profitability has dropped 15% over two years. They want to know why, and what to do about it. You have 30 minutes. No prep. No notes. Just your thinking, out loud, while someone evaluates every word.

That's a case interview. And if you've never done one before, it sounds terrifying.

It should. Case interviews are hard. Top consulting firms accept roughly 1% of applicants, according to Management Consulted's 2025 recruiting data. But among candidates who reach the case stage, preparation quality is the single biggest predictor of who gets offers. Candidates who complete 30+ practice cases are three times more likely to receive an offer than those who do fewer than 10.

The fear is real. The difficulty is real. And the path through it is practice, not talent.

What you're actually being tested on

Case interviews simulate the real work of consulting: analyzing ambiguous business problems, structuring your thinking, and communicating recommendations under pressure. Unlike behavioral interviews where you draw from past experience, cases ask you to solve a problem you've never seen, in real time, with an audience.

A typical case runs 30 to 45 minutes:

1. You listen to the prompt. The interviewer describes a business scenario.
2. You structure your approach. One to two minutes to outline how you'll break down the problem.
3. You analyze data. The interviewer shares data points, charts, or tables as you work through your structure.
4. You do math. Almost every case involves quantitative analysis. No calculator.
5. You make a recommendation. Synthesize findings into something clear and actionable.

Firms use this format because it mirrors what consultants do every day. If you can't structure ambiguous problems, work with data, and communicate clearly to a client, you won't survive your first engagement. The interview is a direct preview of the job.

The four case types

Most cases fall into one of four categories. Knowing the patterns helps you recognize what you're looking at quickly. But real cases blend types, so don't get rigid about categories.

Profitability cases

Setup: Profits are declining. Figure out why and fix it.

Break profit into revenue and costs. Then decompose:

• Revenue = Price x Volume. Has price changed? Has volume changed? Which products or segments are affected?
• Costs = Fixed + Variable. Which categories increased? Are increases driven by volume or rate changes?

The trap here is running mechanically through both sides. If the interviewer tells you revenue has been flat, stop asking about revenue. Follow the data.

Example: "Your client is a national pizza chain. Profits dropped 20% last year despite flat revenue." You'd quickly establish costs are the issue, then dig into which categories (labor, ingredients, rent, marketing) drove the increase and whether it's an industry trend or specific to this client.

Market entry cases

Setup: A company wants to enter a new market. Should they, and how?

Work through:

• Market attractiveness: Size, growth, profitability, competition
• Company fit: Capabilities, brand, distribution, resources
• Entry strategy: Build, buy, or partner
• Financials: Investment required, expected returns, payback period
• Risks: Regulatory, competitive response, execution

Example: "A US luxury skincare brand wants to enter South Korea." You'd analyze the Korean beauty market (large, competitive, sophisticated consumers), assess brand fit, evaluate entry options (local distributor vs. direct), and model the financials. The South Korean beauty market is one of the most competitive in the world. That context matters.

Mergers and acquisitions

Setup: Should this company buy that company?

Cover:

• Strategic rationale: Revenue synergies, cost synergies, market access, talent, technology
• Target evaluation: Is the target healthy? Strengths and weaknesses?
• Financials: What's it worth? What premium is reasonable?
• Integration: How difficult will merging the two companies be?
• Risks: Cultural mismatch, customer churn, regulatory approval, overpayment

Growth strategy cases

Setup: A company wants to grow revenue. Where should they focus?

Map the options:

• Existing products, existing customers: Increase usage, reduce churn, raise prices
• Existing products, new customers: New segments, geographies, channels
• New products, existing customers: Adjacent products, upselling, bundling
• New products, new customers: Diversification (highest risk)

Evaluate each option against the client's specific situation, capabilities, and market position.

Why memorized frameworks will fail you

This is where most case prep goes wrong. Guides teach you to memorize frameworks (the profitability tree, the market entry checklist, the M&A structure) and apply them mechanically. This works for straightforward cases. It fails on anything complex.

Three reasons.

Cases don't fit neatly. A profitability case might actually be a competitive response case. A market entry case might hinge on an acquisition decision. If you're locked into one framework, you'll miss the real issue.

Interviewers have seen the frameworks. They've watched thousands of candidates walk through the same profitability tree. What impresses them is a candidate who tailors their approach to the specific problem instead of applying a template.

Frameworks don't prioritize. Knowing you should look at revenue and costs doesn't tell you which to investigate first. Prioritization comes from listening to the prompt, asking smart clarifying questions, and following the data.

What to do instead: build your own structure every time

MECE thinking is the underlying skill. Mutually Exclusive, Collectively Exhaustive. Break a problem into buckets that don't overlap and that cover everything.

When you hear a case:

1. Identify the key question. What exactly is the client trying to decide?
2. Break it into sub-questions. What do I need to know to answer this?
3. Check for MECE. Do my sub-questions overlap? Am I missing anything?
4. Prioritize. Based on what I've heard, where should I start?

This is harder than applying a pre-built framework. It's also much more impressive to interviewers, and much more useful in actual consulting work.

Market sizing questions

Sometimes standalone, sometimes embedded in a larger case. They test your ability to make reasonable assumptions and do mental math.

Example: "How many tennis balls are sold in the US each year?"

Walk through it:

1. US population: about 340 million. Roughly 5% play tennis, so 17 million players. Maybe 30% play regularly enough to buy balls: 5 million regular players and 12 million casual players.
2. A regular player might buy a can of 3 balls every two weeks during a 6-month season. That's about 12 cans, or 36 balls per year. A casual player might buy 3 to 4 cans.
3. 5 million regular players x 36 balls = 180 million. 12 million casual players x 10 balls = 120 million. Add institutional purchases. You're in the range of 300 to 350 million.

The exact number doesn't matter. Your structure, assumptions, and arithmetic do.

Math under pressure

Quantitative analysis appears in almost every case. Market sizes, profit margins, break-even points, ROI figures. No calculator. This is where many candidates fall apart, not because they can't do math, but because pressure makes simple arithmetic feel impossible.

Round aggressively. Use $1 billion instead of $987 million. Interviewers expect this.

Show your work out loud. "If the market is $5 billion and we capture 3%, that's $150 million in revenue."

Estimate before you calculate. Before multiplying 47 x 23, say "that's roughly 50 times 20, so about 1,000." Then do the precise math.

Practice daily. Spend 10 minutes a day on mental math. Multiply two-digit numbers, calculate percentages, convert fractions. This muscle atrophies fast.

Don't panic over mistakes. Catch the error, correct it, move on. Your process matters more than perfection.

How to practice

Weeks 1-2: Solo foundations

• Read 10 to 15 case examples from books like "Case in Point" and "Case Interview Secrets." Study the structures, not the answers.
• Practice a market sizing question every day. Pick a random product, estimate its market size, time yourself to five minutes.
• Drill mental math until two-digit multiplication feels automatic.
• Watch case interview videos. Seeing strong candidates work through cases teaches pacing and communication style better than any book.

Weeks 3-6: Partner practice

You need another person. Solo practice builds foundations, but cases are conversations.

• Find a case partner. Other consulting candidates are ideal. Check consulting forums, MBA classmates, LinkedIn groups.
• Practice both roles. Interviewing someone else teaches you what good and bad looks like from the other side.
• Record your sessions. Review for filler words, structure clarity, math accuracy.
• Get feedback on communication, not just content. Did you lay out your structure clearly? Did you check in with the interviewer? Did your recommendation follow logically from your analysis?

Ongoing: AI practice to fill the gaps

Between partner sessions, AI mock interviews let you run a full case start to finish with feedback on structure and communication. You can repeat as many times as you want without coordinating someone else's schedule.

The 30-case benchmark

Aim for 30 full cases before your interviews. That's about 5 per week over 6 weeks. Distribute them:

• 8 to 10 profitability cases
• 6 to 8 market entry cases
• 4 to 6 M&A cases
• 4 to 6 growth strategy cases
• 4 to 6 market sizing exercises

The mistakes that sink people

Jumping to solutions. The most common mistake. Candidates skip structure and dive straight into an answer. Interviewers want to see your thought process. Take two minutes to build your framework before analyzing anything.

Ignoring the interviewer's hints. Cases are collaborative. If the interviewer says "let's focus on the cost side," they're pointing you toward the interesting data. Follow their lead.

Data without insight. Calculating that costs increased 15% is analysis. Explaining that the increase is driven by a 40% spike in raw material prices due to supply chain disruption, and recommending a dual-sourcing strategy, is insight. One gets a nod. The other gets an offer.

Losing the thread. Keep track of where you are. Summarize periodically: "So far we've established that revenue is flat and the profit decline is driven by a 15% increase in variable costs, specifically raw materials. I'd like to dig into why raw material costs increased."

Weak recommendations. "The client should reduce costs" is weak. "The client should renegotiate raw material contracts and diversify suppliers to reduce input costs by 10% to 12%, recovering roughly $8 million in annual profit" is strong. Be specific. Be actionable. Tie it to your analysis.

What a great performance looks like

After observing thousands of cases, hiring managers at top firms describe the ideal candidate the same way:

They ask thoughtful clarifying questions before diving in. Their structure is logical, MECE, and tailored to the specific problem. They prioritize based on available information rather than following a checklist. Their math is fast and accurate, or they catch and correct their own mistakes. They turn data into insights, not recitations. Their recommendation is crisp and directly supported by their analysis. They communicate clearly throughout, checking in with the interviewer and adjusting based on new information.

That description sounds like a lot. It is. But every piece of it is trainable. Case prep is one of the most time-intensive parts of consulting recruitment. The skills it builds, structured problem-solving, quantitative reasoning, clear communication under pressure, are valuable far beyond the interview room. Start early and practice consistently. There's no shortcut, but there's also no mystery.