Run a full case interview against an AI interviewer that pushes back like a real MBB engagement manager — then see exactly where your structure, quantitative work, and synthesis lost points.
Free to start · No credit card required · No webcam needed
Profitability case · 32 min
Overall
Biggest gap · Synthesis
Your recommendation listed findings before stating the answer. Lead with the recommendation, then support it.
A complete case from prompt to recommendation, run in real time. You get the prompt, ask clarifying questions, lay out a structure, work through the quantitative section out loud, and deliver a synthesised recommendation under time pressure.
Cases span the formats consulting firms actually use: profitability, market entry, market sizing, growth strategy, M&A, and operations. Interviewer-led cases follow the McKinsey style, where you are steered question to question. Candidate-led cases follow the BCG and Bain style, where you drive the structure and decide what to analyse next.
The point of practising out loud is that reading case books does not surface the failure modes that actually cost offers: a structure that is not MECE, arithmetic that falls apart when you are talking, and a recommendation that never answers the question asked.
A sample of what the mock asks, and what a strong answer has to do.
Profitability cases reward a clean profit tree before any hypothesis. Flat revenue with falling profit points at cost, but you still have to prove it by splitting revenue into price and volume first.
Market sizing is scored on the transparency of your assumptions, not the accuracy of the number. State each assumption, keep the arithmetic round, and sanity-check the result against something you know.
The synthesis is where most candidates lose an otherwise strong case. Lead with the answer, give two or three supporting reasons, then name the biggest risk and the next step.
Choose profitability, market entry, market sizing, growth, or a random draw. Set interviewer-led or candidate-led to match the firm you are targeting.
The AI interviewer gives you the prompt, answers clarifying questions, hands over data when you ask for it, and pushes back when your logic slips.
Within minutes you get a transcript annotated with scores per dimension, the specific moments that cost points, and what a stronger answer would have sounded like.
Almost every candidate who fails a case interview understood the frameworks. What they had not done is perform under time pressure while a stranger interrupts them. Silent preparation trains recognition; the interview tests production.
Practising out loud surfaces the gap immediately. You hear yourself lose the thread mid-structure, drop a decimal place while narrating, or finish a case without ever stating a recommendation. Those are the errors that decide outcomes, and they only appear when you are speaking.
The other constraint is partner availability. Case partners are hard to schedule, inconsistent in quality, and rarely willing to sit through your fifth profitability case in a week. An AI interviewer removes the scheduling problem entirely and gives you the same scoring rubric every time, so improvement between sessions is actually measurable.
Most successful MBB candidates report somewhere between 25 and 50 practice cases before their first round. The number matters less than the distribution: a spread across case types, a mix of interviewer-led and candidate-led formats, and enough repetition on your weakest dimension that it stops being the thing that sinks you.
Front-load structure work, then shift to quantitative drills once your frameworks are reliable, then spend the final stretch on synthesis and delivery. Track scores per dimension across sessions rather than counting cases completed — the count is a vanity metric.
Consulting recruiting runs on a tight calendar. Applications for full-time and internship roles typically open in July and August, with first-round interviews clustering from September through October and final rounds close behind. Firms also run online assessments before the case rounds, so the screening starts earlier than most candidates expect.
Working backwards, serious case practice should start six to eight weeks before your first round. Starting in August for an October interview leaves room to build structure, drill maths, and still get a dozen full cases in at realistic pace.
Your first mock interview and scored report are free. Most candidates find their weakest dimension in the first session.
Yes. You can run a full case interview and get the scored report without paying. You only need an account so your transcript and feedback are saved.
A full case runs 25 to 35 minutes, matching a real first-round case. The scored report is ready within a few minutes of finishing.
Yes. Interviewer-led cases follow the McKinsey style where the interviewer steers you question to question. Candidate-led cases follow the BCG and Bain style where you drive the structure and decide what to analyse. You pick the format when you start.
Yes. You can run a standalone market sizing case rather than a full case if you want to drill the quantitative section specifically.
Case partners are inconsistent and hard to schedule, and most are not calibrated to a scoring rubric. The AI interviewer is available whenever you are, applies the same criteria every session, and gives you a per-dimension score you can track over time. It works best alongside peer practice, not instead of it.
A full transcript annotated with scores for structure, quantitative work, business judgement, communication, and synthesis, plus the specific moments where you lost points and what a stronger answer would have sounded like.