Key Takeaways
- McKinsey runs interviewer-led cases: they steer you question to question, unlike the candidate-led cases at BCG and Bain. Your job is to nail each discrete question, not to drive the whole case.
- Roughly half the evaluation is the Personal Experience Interview (PEI) — structured behavioral stories on personal impact, leadership, and drive. Most candidates under-prepare it badly.
- The case still rewards structure, clean math narrated out loud, and a clear recommendation — the universal consulting skills.
- McKinsey weighs communication and synthesis heavily: say the answer first, then support it.
- You can't cram either half. Both reward out-loud reps with feedback, not silent reading.
"McKinsey case interview" gets searched far more than it used to, because consulting recruiting keeps pulling its timelines earlier and candidates have less runway to prepare. The mistake most of them make is preparing for a case interview generically, when McKinsey's format has two specific features that change how you should train.
What makes a McKinsey case different
Most case prep teaches the candidate-led model: the interviewer hands you a prompt, you build a framework, and you drive the analysis where you think it should go. That's BCG and Bain. McKinsey is interviewer-led. They walk you through a sequence of distinct questions — "how would you structure this?", "what could be driving the cost increase?", "the client has this data, what does it tell you?", "what's your recommendation?" — and your job is to answer each one cleanly, not to seize the wheel.
This changes your prep. You're not rehearsing one long framework monologue; you're rehearsing tight, well-structured answers to discrete prompts, and you're staying responsive to where the interviewer is taking you. Over-driving a McKinsey case — barrelling ahead with your framework while they're trying to direct you — reads as not listening.
Two more McKinsey emphases worth knowing. They reward synthesis: when asked for a recommendation, lead with the answer in one sentence, then support it — don't narrate your way toward it. And they weigh communication unusually heavily, because the job is partly client-facing from day one.
The half everyone forgets: the PEI
Here's what generic case prep skips. At McKinsey, a large share of the evaluation is the Personal Experience Interview — structured behavioral questions, usually probing three dimensions: personal impact (a time you influenced or persuaded someone), leadership (a time you led a team through difficulty), and entrepreneurial drive (a time you drove something hard against resistance).
These are not warm-up small talk. Interviewers go deep on one story — "what exactly did you say?", "how did they react?", "what would you do differently?" — for ten or more minutes. A shallow story collapses under that questioning. Candidates who spend all their prep on cases and none on the PEI routinely get rejected on the half they ignored.
Prepare two or three genuinely strong stories per dimension, each detailed enough to survive five layers of follow-up. Know the specifics: the numbers, what you personally did versus the team, the tension, the outcome. Vague stories die in the PEI.
⭐ Pro Tip
Build your PEI stories before you grind a single case. They take longer to develop than framework recall, they're half the score, and they're the part you cannot improvise. Most rejected candidates were strong on the case and thin on the PEI.
How to prepare for both halves
For the case: get the frameworks down first — our consulting case interview framework guide covers the structures (profitability, market entry, pricing, MECE). Then shift to the format-specific skill: answering discrete interviewer-led questions cleanly, leading with synthesis, and narrating math out loud. The general method for that — record yourself, review, fix one thing per rep — is in our guide on practising cases out loud.
For the PEI: write the stories, then tell them out loud and record the follow-ups you'd dread. The PEI is an interview about your life that you'll still fumble if you've never said the stories aloud. Rehearsing them is the difference between a story that deepens under questioning and one that crumbles.
The thread across both halves is the same: McKinsey is evaluating you live, on communication and structure, in real time. Silent reading prepares neither. Out-loud reps with someone (or something) pushing back prepare both.
Rehearsal · practice that compounds
McKinsey evaluates you live. Practice live.
The case and the PEI are both performances under questioning — and both punish candidates who only prepared on paper. Rehearsal's AI interviewer runs interviewer-led prompts and deep behavioral follow-ups, so the real round isn't the first time you've said it out loud.
The one-line version
McKinsey is interviewer-led and half PEI. Prepare the behavioral stories first, drill clean answers to discrete prompts, lead with synthesis, and rehearse both halves out loud — because they're scoring how you perform, not what you've read.