The BCG case interview, and how it differs

What does a BCG interview usually look like end to end?

Candidates preparing for Boston Consulting Group often arrive with McKinsey's format already drilled, then feel slightly off balance in the room, because BCG runs a related but different exam. The overall shape has stayed consistent enough across recent hiring cycles to prepare for: an application, a resume screen that for some candidates includes a chatbot-based case screen, one or more live interview rounds, and a fit conversation that earns its own slice of the score.

With that caveat carrying full weight, here is the generally reported pattern, drawn from what BCG publishes on its own careers site and from what is widely and consistently described by candidates who have been through it.

• An online application, screened first on the resume and any assessment BCG requires for that role
• For a subset of candidates, a chatbot-based online case, commonly known by the name Casey, that runs before any live interviewer is involved
• A first round, commonly reported as two interviews, each pairing a case with a shorter conversation about your background
• A final round with more senior interviewers, often partners or managing directors, where the cases run broader and the questioning goes deeper
• For some offices and roles, most often reported in parts of Europe, a written case: a data packet, a fixed amount of time to build a recommendation, and a short presentation

On the online case, keep expectations general until you reach it. Casey is common enough, and specific enough in its own right, to earn its own guide, which covers what the chatbot actually asks, how the format differs from a live interviewer, and where candidates lose points without realising it.

One disclosure, plainly: MBB Ready is not affiliated with, endorsed by or connected to Boston Consulting Group. What follows is drawn from BCG's own published materials plus what is publicly and consistently discussed about the format. It is not an internal document and does not speak for the firm.

What does candidate-led actually mean when you're in the room?

BCG is generally described as running candidate-led cases: you are given a prompt and a question, and the next stretch of the conversation is yours to organise. You clarify what is being asked, propose a structure, name where you want to start, and ask for the data that branch needs. The interviewer holds information back until you ask for it, then hands it over, often as an exhibit that was ready the whole time.

That changes the texture of the room compared with a tightly interviewer-led case. Long stretches can pass with the interviewer saying very little, because the case is doing what you built it to do. Silence here is not disapproval and not a cue to fill the air with narration. It means you are driving, and nobody is going to take the wheel back for you.

What good looks like

Visible ownership without wandering is the discipline. Say what you are about to do before you do it: "I want to check whether this is a volume problem or a price problem, so could I see revenue split both ways." Then finish the branch you opened before starting another. A structure that survives contact with real data, and gets amended out loud when the data contradicts it, reads better than one defended out of loyalty.

BCG is also widely described as leaning collaborative within that candidate-led shape. Interviewers tend to engage as a thinking partner rather than a silent examiner, occasionally offering a steer, a genuine "what if", or a nudge toward an angle you have not considered. Take those. Accepting a real hint and building on it well is not a sign of weak structuring. It is the two-way conversation BCG describes itself as running.

The trap: waiting to be led

The most common failure looks like this. A candidate lays out a strong structure, states which branch they will start with, then stops and waits, as if for permission. Nothing comes. They wait a little longer. The interviewer eventually prompts them, and in doing so quietly takes back a decision that was supposed to be scored as the candidate's own. Repeat that across thirty minutes and the transcript reads like someone being led through material rather than driving it, whatever the individual answers were worth.

How do the maths and exhibit questions work in a candidate-led case?

The quantitative moments in a BCG case follow from who is holding the map. Because you are driving, you typically decide when a calculation is needed and which branch of your structure it belongs to. The interviewer is not going to interrupt with a number you did not ask for. Build a structure with a cost branch and a demand branch, never ask for the data behind either, and nobody stops the case to hand it to you unprompted, the way a more tightly interviewer-led moment might.

Exhibits behave the same way. A chart typically appears because you asked a question that made it relevant, not on a schedule the interviewer is running. That is a genuine skill in its own right: knowing which question, at which point in your structure, is worth spending an exhibit request on, rather than asking for everything up front and drowning in data you have not earned the time to read carefully.

Why the sense-check sentence matters just as much here

When you choose your own inputs, it is tempting to assume you already have a feel for whether the output is plausible, because you picked the numbers. That feel is less reliable than it seems, especially under time pressure, and this is exactly where candidates confidently announce a figure that is off by a factor of a thousand. A number you requested is still a number you can be wrong about.

So the sentence after the arithmetic still carries the substance of the mark. Compare the answer to something you already know: another figure in the case, a rough sense of the client's scale, an earlier number from the same exhibit. Say why the result is plausible, or why it is surprising enough to flag. State units every time; revenue or profit, per store or per year, millions or thousands, said aloud while you write the number down.

How is a BCG case actually different from a McKinsey one?

The comparison candidates ask about most is McKinsey, because the two firms are widely described as sitting on opposite ends of the same spectrum. McKinsey is generally reported as interviewer-led: the interviewer works from a prepared sequence of questions and controls when exhibits appear. BCG is generally reported as candidate-led and more conversational: you propose the path, and the interviewer responds to it rather than dictating it. Bain sits closer to BCG on this than to McKinsey, though accounts differ on exactly how close.

None of that changes what is being judged. A structure has to fit the problem either way. A number has to be sense-checked either way. What changes is how much of your evidence for each skill comes from you proposing versus you responding, and which failure mode is waiting for a candidate who only ever rehearsed one format.

It also is not as clean a split as the shorthand suggests. A BCG interviewer will still ask a pointed question with a single right answer, especially on the maths. A McKinsey interviewer will still hand you a question broad enough to structure unaided. Individual interviewers carry their own habits, and later rounds at both firms tend to loosen up relative to first rounds. Treat the labels as where the centre of gravity sits, not as a guarantee of the room you walk into.

What is the BCG fit interview looking for?

The fit interview, sometimes folded into the same 45-minute slot as the case and sometimes run as its own conversation in later rounds, is not filler. BCG's own careers site names five qualities it evaluates in candidates: integrity, intellectual curiosity, creative thinking, a collaborative mindset and drive. Those five map fairly directly onto what a behavioural interviewer is actually listening for while you tell a story.

Expect at least one version of "why BCG", and treat it as a genuine question rather than a formality. Interviewers are listening for whether your reasons are specific to this firm rather than interchangeable across all three. Reasons drawn from an actual conversation with someone who works there, or a specific piece of BCG's public work you found interesting, read better than reasons that would fit any of the three firms equally well.

Expect stories to be pushed on. A single story followed for several minutes, with three or four questions about what exactly happened, who exactly objected, and what you would do differently, is a more common shape than a rapid tour across many stories. A story built to be told once cleanly tends to come apart under that kind of pressure. A lived one does not.

What is the interviewer writing down while you talk?

Interviewers do not leave with an overall impression and reverse-engineer reasons for it afterward. They leave with a filled-in grid: named skills, a score against each, a line of evidence for why. Our rubric has three dimensions and fourteen skills: analytical thinking at 50%, presence and communication at 25%, track record and drive at 25%. Each skill is marked 1 to 5, where 3 means you met the first-round standard rather than average.

In a candidate-led format, more of the evidence for a few particular cells comes from what you choose to do rather than what you are asked to do. Framing and creativity get more airtime, because the structure you propose actually governs the next twenty minutes rather than being one answer among several. Synthesis gets more weight too, because nobody prompts your close the way an interviewer-led case's final question typically does.

BCG's own materials describe its consulting approach as hypothesis-driven: starting from a working answer and refining it as data arrives, rather than exhaustively mapping every branch before touching any one of them. That preference shows up in the case room. A structure organised around a stated hypothesis, tested and revised out loud as exhibits land, tends to read stronger than an exhaustive checklist worked evenly across every branch regardless of what the data is saying.

How many rounds are there, and how long does the process take?

Exact timelines are the least reliable thing to promise, because they depend on office, role, season and how a specific hiring class is moving that year. With that caveat, several details recur consistently enough across independent accounts to describe as a general shape rather than a single confident number.

• First round interviews are commonly reported as two interviews, roughly 45 minutes each, split between a case and a shorter conversation about your background
• Final round interviews are commonly reported as two to three interviews with more senior staff, often run back to back on the same day
• The gap between rounds is commonly reported as ranging from about a week to several weeks, longer around graduate recruiting seasons when volumes are highest
• A decision after the final round is commonly reported to arrive within a few business days, sometimes the same day for on-campus or condensed hiring events

We are deliberately not turning any of that into a single confident figure. Prep sites that quote a precise week count or a precise pass rate are quoting an average across offices and years that may have nothing to do with your specific process. Your recruiter has the actual shape of your timeline in front of them, and it is the only version worth planning around.

How should you prepare for this specific format?

Generic case practice gets you most of the way, because the analytical skills are the same everywhere. What it will not give you is the habit of proposing rather than waiting, and that is worth deliberate rehearsal, because it is the exact opposite of what a strict interviewer-led format trains.

Our default session is a 30-minute case-only run, which cannot assess track record and drive because it has no fit portion. The 60-minute version at /full includes one, and that is the one to use once your case fundamentals hold up. Adrian Caldwell-Pierce, one of our AI coach personas, is built as an ex-BCG partner and runs candidate-led cases, so it is the closest single rehearsal to the room described in this guide. The first case is free. Our coaches are AI personas, designed by ex-MBB interviewers, and we say so everywhere: you are talking to a rubric with a voice around it, not a person.