A case interview opening is the first two to three minutes of the case, in which you restate the problem, confirm what the client is actually trying to achieve, ask the questions you need answered, and present the structure you will work through. It is the only part of a case that follows the same shape every time, which is what makes it the part you can make reliable. Interviewers use it to judge how much of the rest of the case they will have to steer.
This guide covers the four moves an opening consists of, what each one is for, and the mistakes that make an otherwise strong candidate look unprepared in the first sixty seconds.
The four jobs of an opening
| Move | What it is for | Roughly how long |
|---|---|---|
| Restate the prompt | Catch a misheard fact before you build on it | 20 seconds |
| Confirm the objective | Establish the measure your recommendation is judged against | 15 seconds |
| Ask clarifying questions | Resolve ambiguities that would change your structure | 30–60 seconds |
| Present the structure | Give the interviewer a map of where you are going | 30 seconds |
The order matters more than the timing. Each move depends on the one before it: you cannot confirm an objective you have misheard, and you cannot know which clarifications matter until you know what you are optimising for.
Restate the prompt before you touch it
Read the prompt back in your own words and stop. Do not analyse, do not hypothesise, do not add. The purpose is narrow — to confirm the facts you are about to build on.
“So the client is a regional grocery chain with forty stores in one province, profit has fallen over the last two years, and they want to understand why. Have I got that right?”
Three things get caught here: numbers you misheard, scope you assumed (national when the interviewer said regional), and the direction of the change, which candidates invert more often than they expect under pressure.
Restating in your own words rather than repeating the interviewer’s does the real work. Verbatim repetition proves you can hold a sentence in memory. Paraphrase proves you understood it, and it surfaces the places where you did not.
Confirm the objective, and get it in a number
“Improve profitability” is a direction, not an objective. Ask what the client is measuring and by when.
“Is there a specific profit level they are targeting, or a timeframe they have in mind?”
The answer changes the case. A client that wants to recover four million dollars of operating profit within twelve months cannot get there by entering a new segment, so entire branches of your structure become irrelevant before you have drawn them. A client with a three-year horizon and no stated number is asking a different question and deserves a different structure.
Two objectives are worth asking about explicitly because they are so often unstated: whether there is a constraint the client will not break — a plant they will not close, a brand they will not discount — and whether the measure is absolute profit or margin. Those two questions cost fifteen seconds and reshape the case more often than any other.
Ask fewer clarifying questions, and better ones
The test for a clarifying question is simple. If the answer came back either way, would your structure change? If not, do not ask it. Three good questions beat eight that establish colour.
Questions that usually earn their place:
- Scope. Which parts of the business are in play? A profit decline across four business units is a different case from a decline in one.
- Comparison. Falling against what — last year, budget, competitors? The benchmark determines what counts as the problem.
- Constraint. What is the client unwilling or unable to do? This removes branches you would otherwise spend the case on.
- Definition. Where a term in the prompt is genuinely ambiguous — expansion, digital, the market — ask what it means here.
Questions that usually do not earn their place: anything you could reasonably assume and state as an assumption, anything whose answer is “it depends”, and anything asked to appear thorough. Interviewers notice the difference, and the second kind consumes the time you needed for structuring.
If a question would help but the interviewer says the information is not available, say what you will assume instead and move on. “I will assume the decline is spread across all forty stores rather than concentrated in a few, and I will test that early” is a complete and confident recovery.
Build the structure, then name it
Take the time out loud: “May I take two minutes to put a structure together?” Then use them. Write the structure down; do not compose it while speaking.
The order that separates strong candidates from rehearsed ones is problem first, framework second. Start from what this client is trying to achieve and decompose that. If the result happens to resemble a profitability tree, say so — the label is useful shorthand. If it does not, you have still built something defensible.
Announcing the framework first inverts this, and it shows. A candidate who opens with “I will use the four Ps” has committed to product, price, place and promotion before establishing that the case is about marketing at all. When it turns out to be about a supply constraint, the structure has to be abandoned in front of the interviewer. The McKinsey 7S framework guide makes the same argument at more length: a framework is a prompt for investigation, not a script.
Three properties make a structure work in an interview:
- It is driven by the objective. Every branch, if investigated, could plausibly change the recommendation.
- The branches do not overlap. Two branches that could each claim the same cost make the arithmetic ambiguous later.
- You can say which branch you would start with, and why. This is the sentence that turns a structure into a plan, and many candidates leave it out.
Present it in thirty seconds
State the number of branches, name them, then say where you want to begin.
“I would like to look at this in three parts. First, whether the profit decline is coming from revenue or from cost. Second, within whichever of those is driving it, which segment or cost category accounts for the change. Third, what the client can realistically act on given the twelve-month horizon. I would start with the first, because until we know which side it is on the rest is guesswork — could we look at how revenue and costs have moved over the two years?”
That last clause matters. It hands the case back to the interviewer with a specific request, which is how an opening ends cleanly instead of trailing off.
Common failure modes
| What it looks like | What the interviewer reads | Fix |
|---|---|---|
| Diving into analysis with no restatement | Has not checked the facts | Restate every time, even when the prompt seems simple |
| Six or seven clarifying questions | Cannot tell what matters | Apply the “would it change my structure” test |
| Framework announced before the problem is understood | Rehearsed, not thinking | Decompose the objective, label afterwards |
| Long silence with no request | Stuck | Ask for the time and say how much |
| A structure with no starting point | Lists rather than prioritises | End with the branch you would open and why |
| Objective left as “improve performance” | Will have nothing to test the answer against | Ask for the measure and the timeframe |
Practise the opening on its own
The opening is the most drillable part of a case, because it does not depend on the case. Take ten prompts, do only the first three minutes of each, and stop. Twenty minutes of that is worth more than one full case, because you get ten repetitions of the move you will make in every interview instead of one.
Record yourself and check three things: whether the restatement changed any fact from the prompt, whether every clarifying question actually moved a branch, and whether you named a starting point. Those three are the difference between an opening that buys you control of the case and one that merely gets through it.
When you are ready to run the whole case rather than the opening, the worked case studies walk through complete answers from prompt to recommendation, and the case interview preparation hub sets out the order to work through them in.
