Case Interview Secrets

Victor Cheng, a former consultant and case interviewer at McKinsey & Company, draws on his experience on both sides of the interview table to offer a systematic guide to passing the case interview, a specialized evaluation format used by top strategy consulting firms such as McKinsey, Bain & Company, Boston Consulting Group (BCG), Oliver Wyman, A.T. Kearney, Monitor, and Roland Berger. Cheng recounts how his initial 40-minute case interview ended after only 3.5 minutes, despite his strong academic credentials, prompting him to spend over 100 hours preparing, including 50 practice sessions. He went on to pass 60 out of 61 case interviews, received six job offers, and accepted a position at McKinsey, where he was promoted to associate as one of only 10 out of 100 in his starting class. This background, spanning his roles as candidate, consultant, and interviewer, forms the basis for the book's advice.

Cheng identifies seven evaluation tools consulting firms use, grouped into two categories. Quantitative assessments include standardized tests such as McKinsey's Problem Solving Test (PST) and estimation questions, which require candidates to estimate figures like market sizes using pen-and-paper math. Hypothetical-situation case interviews encompass five formats: the candidate-led case, in which the candidate drives the analysis after receiving an ambiguous prompt; the interviewer-led case, in which the interviewer controls the direction; the written case, the group case, and the presentation-only case.

For estimation questions, Cheng presents six core skills: simplifying large-number arithmetic into smaller computations, rounding intelligently in an offsetting fashion to keep estimates within a reasonable margin of error, finding a proxy (a correlated substitute number), identifying the proxy's imperfections, segmenting estimates into sub-estimates, and solving sub-estimates via assumptions.

Cheng argues that case interviews simulate the daily reality of consulting. Interviewers assess whether a candidate can be sent to a client site with minimal supervision and handle the work independently. Key interviewer priorities include avoiding "boiling the ocean" (doing only what is necessary under tight time and resource constraints), being accurate enough rather than precisely accurate, supporting every claim with factual justification, and communicating diplomatically so clients can follow the reasoning. The "airplane test" gauges whether an interviewer would want to sit next to the candidate on a three-hour flight, serving as a proxy for the interpersonal skills critical to client management. Firms also prize process excellence, the ability to follow a structured analytical approach consistently, over arriving at the right answer through an unrepeatable method.

The book's central section introduces four core problem-solving tools. The first is the hypothesis: A candidate states an educated guess about the client's problem, tests it with data, and iterates. Cheng's Five-Minute Hypothesis Rule states that if a candidate has not offered a hypothesis by the fifth minute, the candidate should state one immediately, because further delay risks omitting one altogether.

The second tool is the issue tree, a logical structure that lays out conditions which, if proven true, prove the hypothesis. A framework is an issue tree template for commonly recurring business problems. Cheng outlines three validity tests: the tree must test the specific hypothesis, it must be MECE (mutually exclusive and collectively exhaustive, meaning categories have no overlap and together cover all possibilities), and it must be "conclusive," meaning that proving all branches true should make the opposite conclusion unimaginable. He sharply criticizes "framework robots," candidates who mechanically apply memorized frameworks without considering whether the framework tests their hypothesis, calling this interviewers' single biggest complaint.

The third tool is drill-down analysis, a process-of-elimination approach in which the candidate works down each branch of the issue tree, gathering data to support or contradict the hypothesis and revising it as needed. Cheng advises starting with the branch that eliminates the most uncertainty and using quantitative and qualitative analysis in roughly a 70/30 ratio.

The fourth tool is synthesis, a top-down communication structure distinct from a chronological summary. The preferred format is to state the action-oriented conclusion first, provide up to three supporting points, and restate the conclusion. Cheng notes that roughly 80 percent of the feedback he received from his McKinsey managers related to communication skills rather than analytical ones, underscoring the importance of concise delivery.

Cheng presents three core frameworks covering roughly 70 percent of cases. The profitability framework deconstructs profits into revenues (unit price multiplied by units sold) and costs (fixed plus variable), enabling candidates to isolate whether a problem is revenue-driven or cost-driven. He emphasizes always comparing metrics to historical data or competitors, because absolute numbers lack meaning without context. The business situation framework provides qualitative understanding through four branches: customer, product, company, and competition. Once a key insight emerges, candidates should synthesize and revise the hypothesis rather than completing the framework for its own sake. The mergers and acquisitions (M&A) framework applies the business situation framework independently to each company in a potential transaction and then to the combined entity, focusing on either strategic value or cost savings.

To demonstrate flexible application, Cheng walks through an extended case about Omega & Omega, a fictional billion-dollar advertising agency with flat sales and profits. Using the profitability framework, the candidate determines the stagnation is company-specific. Switching to the business situation framework, the candidate discovers that digital advertising is growing rapidly while the client remains entirely dependent on traditional advertising. The case transitions into an M&A analysis, and the candidate evaluates three acquisition targets, concluding that eCool is the strongest candidate based on its size advantage in U.S.-based Fortune 500 accounts, comparable financial ratios, and alignment with the client's customer base. Throughout, the hypothesis drives the analysis, with frameworks applied selectively.

For the candidate-led format, Cheng provides a five-step opening: stall briefly to let the initial panic subside, clarify terminology, request time to organize thoughts, state the hypothesis, and communicate the case structure. For the interviewer-led format used by McKinsey, five modular phases (introduction/hypothesis, problem structuring, quantitative analysis, brainstorming, and synthesis) make the case easier overall but shift differentiation to issue tree quality and synthesis. Cheng also addresses written, group, and presentation-only formats, noting that group cases test client management skills and that presentation slides should feature titles answering the "so what?" question rather than merely labeling data.

Cheng argues that securing offers depends on disciplined practice, not merely knowledge. Of candidates who received offers at top firms, approximately 90 percent invested 50 to 100 hours in preparation. He outlines four steps to mastery: building knowledge, finding role models, practicing in live settings (ideally around 50 cases), and seeking assessment from a mentor who can identify subtle mistakes. On confidence, Cheng notes that most of his McKinsey colleagues were introverts whose confidence derived from technical competence rather than personality, and he identifies three sources of confidence: extreme competence, correct mental perceptions (such as viewing the interview as a two-way assessment), and extensive practice.

The book concludes with the ten most common mistakes in interview order: failing to state a hypothesis, not linking the framework to the hypothesis, insufficient mutual exclusivity in the issue tree, omitting a key factor, missing insights from insufficient quantification, missing insights from lack of qualitative questioning, making math errors (which result in automatic rejection roughly 95 percent of the time), jumping around instead of drilling down linearly, pursuing unnecessary analysis, and delivering an activity-based summary instead of a synthesis. Cheng stresses that 90 percent of rejections involve at least one of these mistakes and that most candidates who commit them understand the principles but have not practiced enough to eliminate bad habits. The book's closing message is that multiple offers require following a proven process, using available resources, and working hard.