10 Tips to Beat the Case Interview

The case interview is where most consulting candidates lose the offer, and almost none of them lose it for the reason they think. They assume they got cut because they didn't know the right framework. They got cut because they memorized one and tried to staple it onto a problem it didn't fit.

When I went through McKinsey's on-cycle recruiting, what surprised me was how little the partners cared about framework names. They cared whether I could take a vague business problem, break it into a few clean drivers, do the math without losing the thread, and tell them what it meant. Here are the ten things that actually move a case from a rejection to an offer, in roughly the order they matter.

What do consulting interviewers actually test in a case?

Five things: structure, quantitative reasoning, business insight, communication, and how you respond under pressure. None of them is "which framework did you pick." The interviewer wants to watch you think through a problem the way a consultant would in front of a client. A candidate who reasons clearly through a custom structure beats a candidate who recites a textbook framework every time.

Is the case the only thing that matters in consulting interviews?

No. The behavioral round and fit matter more than candidates expect, and at McKinsey the Personal Experience Interview is its own evaluated component. But the case is the filter that cuts the most people, so it deserves the most prep hours. Get the case to a reliable pass and your behavioral stories carry the rest.

1. Build your own structure instead of memorizing frameworks

The single highest-leverage habit is building a custom structure for the specific case in front of you, not pulling a memorized template off the shelf. Interviewers can tell within thirty seconds whether your structure was made for their problem or borrowed from a book. The borrowed one is almost always too generic to be useful.

The fix is to practice structuring out loud. Take any business question, pause, and lay out three or four MECE buckets that actually fit it. Frameworks like profit equals revenue minus cost are starting points, not answers.

Drill structuring as its own skill: ten different prompts, three minutes each, until building a clean tree feels automatic.

2. Identify the case type before you structure

Almost every case maps to one of five types, and naming the type first tells you where to point your structure. The five are profitability, market entry, mergers and acquisitions, operations or cost reduction, and market sizing. Get the type right and the structure follows naturally.

A profitability case wants you to isolate whether the problem is revenue, cost, or both before proposing fixes. A market-entry case wants market attractiveness, the client's ability to win, and the financial case. Misreading the type sends you down the wrong tree from the first sentence.

In the first minute, silently classify the case into one of the five types before you build anything.

3. Restate the objective and ask before you dive in

Taking thirty seconds to restate the question and confirm the objective is the cheapest way to avoid solving the wrong problem. Candidates lose entire cases because they answered "should we enter this market" when the real question was "should we enter profitably within two years."

Repeat the prompt back in your own words. Confirm the success metric. Ask one or two clarifying questions that actually change your approach, not filler questions to look engaged. Then ask for a moment to structure.

4. Drill mental math as a separate skill

Quantitative pressure fails more candidates than weak structure, and it is the most fixable problem on this list. Cases require multiplying large numbers, working percentages and fractions, and computing breakevens, all while talking and staying calm.

You cannot improve this during case practice alone. It needs its own reps. Twenty minutes a day of timed arithmetic, no calculator, for two to three weeks before interviews will noticeably change how you perform.

Practice math out loud while someone watches, because doing it silently on paper is not the skill the interview tests.

5. Lead with the answer, then the reasoning

Strong case communication is top-down: state your conclusion first, then give the two or three reasons behind it. Consultants call this the "so what." A partner reading your recommendation should get the punchline in the first sentence and the support after.

Candidates who narrate every calculation before reaching a point sound like they are thinking out loud rather than concluding. Pause, decide what your answer is, say it, then justify it.

End every section of the case with a one-sentence "so what" before the interviewer has to ask for it.

6. Practice live with a partner, not just by reading

Reading case books builds recognition; live reps under pressure build the actual skill, and only one of those gets you the offer. You can read every page of Case in Point and still freeze when a real person is watching the clock.

Find a practice partner or a case-prep platform and run live cases where someone pushes back. PrepLounge is built for partner practice, and Crafting Cases have strong worked examples, and IGotAnOffer publishes detailed case walkthroughs. The firms publish their own cases too: McKinsey, BCG, and Bain all post practice cases on their careers sites.

Aim for at least twenty to thirty live cases before your first real interview, with the last ten under realistic pressure.

7. For market sizing, show the logic, not the number

In a market-sizing case the interviewer is grading your assumptions and arithmetic, not whether your final number is correct. Asked how many gas stations are in the US, nobody expects you to nail it. They want to see a clean top-down or bottom-up build with stated assumptions.

Pick a starting population, ratio down with explicit assumptions, and sanity-check the result at the end. Say your assumptions out loud so the interviewer can follow and challenge them.

8. Isolate the cause before you propose the fix

The most common profitability-case mistake is jumping to recommendations before isolating where the problem actually sits. If profits are down, first determine whether it is a revenue issue, a cost issue, or both, and then which specific line is moving.

Diving into "here's what they should do" before you've found the driver signals weak discipline. Walk the tree. Narrow to the actual cause. Then, and only then, talk solutions, naming the specific mechanism rather than a vague "cut costs."

9. Stay calm when the interviewer pushes back

Pushback is usually a test of composure, not a signal that you are wrong, and how you handle it is itself being scored. When an interviewer says "are you sure about that," they want to see whether you can defend sound reasoning and update on genuinely new information.

Don't crumble and reverse a correct answer at the first challenge. Don't dig in stubbornly when they hand you new data that should change your view. Acknowledge the point, reason through it openly, and adjust if the data warrants it.

Treat every "are you sure?" as a request to show your reasoning, not as proof you made a mistake.

10. Prep the firm-specific format and case style

Each MBB firm runs a different pre-interview assessment and a different case style, and walking in blind to those differences costs easy points. McKinsey uses the Solve game and an interviewer-led case where the interviewer steps you through. BCG uses the Casey chatbot and a candidate-led case where you drive. Bain uses its own online test and a candidate-led style with heavy emphasis on the closing synthesis.

Know which assessment you are facing and which case style the firm uses before the round. The reasoning skills transfer, but the choreography does not, and the firm-published prep pages tell you exactly what to expect.

Match your prep to the firm: interviewer-led reps for McKinsey, candidate-led reps for BCG and Bain, and the right assessment for each.

What about the candidates who say they "just aren't case people"?

There is no such thing. The case interview looks like a personality test and behaves like a trainable skill. Every habit on this list, from structuring on the fly to staying composed under pushback, gets measurably better with reps. Candidates who believe they are bad at cases are almost always candidates who have done fewer than ten live ones. The people who walk in calm have usually done forty.

Where to start

If you have a month, spend week one on structure and case-type recognition, week two on mental math and live reps, week three on full timed cases with pushback, and week four on the firm-specific assessment and your behavioral stories. If you have only a week, drill mental math every day and run as many live cases as you can find a partner for, because live reps under pressure are the closest thing to the real round.

Build the structure habit first. It is the skill the other nine depend on, and it is the one that separates a candidate who sounds like a consultant from one who sounds like they read a consulting book. Everything else — the math fluency, the top-down communication, the composure when the interviewer pushes — gets layered on top of a clean structure.

The candidates who beat the case interview are not the ones who memorized the most frameworks. They are the ones who can take a problem they have never seen, structure it cleanly, run the numbers without panicking, and tell the interviewer what it means. That is a learnable skill, and the reps are the whole secret.

Stephen Turban is the co-founder of Wall Street Guide and Lumiere Education. He graduated Magna Cum Laude from Harvard College in Statistics and worked as a Business Analytics Fellow at McKinsey & Company. He founded WSG to give ambitious students the same insider access to finance and consulting recruiting that top-school students take for granted.