Bain Associate Consultant — India EV Charger Sizing 2030

It tests whether you can build a defensible estimate, not whether you land the exact number. The interviewer grades how you structure the problem before calculating, whether each assumption has a one-line justification and a round number, how cleanly you do mental math out loud, whether you sanity-check against an independent anchor, and whether you close with a range and a recommendation. Bain cases are interviewer-led, so the round also tests whether you drive the case and recover when an assumption is challenged rather than waiting to be prompted.

Pick bottom-up or top-down explicitly and say why. A common bottom-up path starts from India's projected vehicle parc by 2030, applies an EV penetration rate, splits by vehicle type since the parc is two-wheeler heavy, applies an EV-to-charger planning ratio, separates highway corridor charging from city destination charging, and converts charge points to physical stations. Lay the equation out loud before you multiply anything, and only segment where a split materially changes the answer.

Jumping into multiplication before stating a structure, pulling an EV-to-charger ratio from nowhere with no defence, never sanity-checking the final number, sanity-checking with the same inputs that built the estimate, calculating silently so the interviewer cannot follow the logic, and ending on a bare number with no range and no so-what. Confusing the PM E-DRIVE target of about 72,000 government chargers with the total market need is another frequent trap.

It behaves like a real round-one interviewer in the ways that matter: it steers mid-case, interrupts, pressure-tests your single most load-bearing assumption, and never coaches you on the method. It differs in that it is consistent and available on demand, it never reacts to accent or delivery style, and it produces a transcript-backed scorecard naming the exact moment your structure or an assumption broke. It will not tell you how you did during the conversation.

Your transcript is scored against weighted dimensions covering how you decompose the problem, how you justify and defend assumptions, your mental-math hygiene, your sanity-check discipline, whether you drove the case, and the quality of your closing recommendation. Being within roughly two to three times the true figure is acceptable if the logic is sound, so structure and defensibility carry far more weight than precision. The scorecard quotes specific moments rather than giving a single vague grade.

Take a breath and do not start multiplying. Restate the question scope in one line, decide whether you are going bottom-up or top-down and say why, then lay out your decomposition out loud as a structure before you touch a single number. Flag the one or two assumptions you expect to be most load-bearing so the interviewer knows you know where the estimate is fragile. Signpost that you will sanity-check at the end.

Do not rationalise the number you already wrote. Recalculate. State why you picked the ratio, acknowledge the range it could plausibly sit in, pick a defensible point and show how the final answer moves if the ratio changes. The interviewer is testing recovery, not whether your first guess was perfect, so the worst response is defending a weak number; the strongest is reworking the estimate calmly and stating the new range.

It opens with an explicit choice of approach and a reason, walks through a clean decomposition before any arithmetic, states each assumption with a short justification and a round number, segments only where it changes the answer, and sanity-checks the result against something independent like the current installed base rather than the inputs that built it. It closes answer-first with a confident range and a one-line implication about what the number means for the question asked.

No. The round rewards reasoning, not recall. You are not expected to know that India had roughly 29,000 public stations in 2025 or that one industry body estimates about 1.32 million by 2030. What matters is that your assumptions are internally consistent and defensible and that your sanity-check is independent. Knowing a rough anchor helps you catch an order-of-magnitude error, but a clean estimate built from stated assumptions scores well even without memorised figures.

Because a market-sizing answer with no sanity-check signals that the candidate cannot tell when their own number is implausible, which is a real risk on client work. A valid sanity-check compares the estimate to an independent reference such as the current installed base or a per-capita figure, never to the same inputs that produced it, since reusing those inputs is a circular reference that proves nothing. The interviewer will explicitly probe whether your check is independent.

It runs about twenty minutes as a single spoken case with no slides and no exhibit. It opens with the prompt and a short structuring pause, moves into the core build where the interviewer pushes on assumptions, escalates with pressure on the weakest part of your logic, and closes with a short reflection on what you would change. The whole conversation is interviewer-led, so expect to be interrupted and steered rather than left to monologue.