Published 9 min read
The Case Interview Recommendation: How to Close Like a Consultant
A case interview recommendation is the closing move where you state your answer in one sentence, back it with two or three reasons pulled from the analysis you actually did, then name the biggest risk or the next step before the interviewer has to ask for it. Lead with the answer, not a recap. Most candidates do the opposite: they walk back through everything they calculated, in order, and let the interviewer guess what they actually think. That single habit, summarizing instead of concluding, is the fastest way to turn a strong case into an average one in the final ninety seconds.
I sat through more than a hundred interviews at McKinsey, and the recommendation was where I made up my mind if I was still on the fence. A candidate could structure cleanly, do clean math, and still lose me here by rambling. The ones who closed well did three things every time, in the same order, without fail. This article breaks down exactly what those three things are and how to practice them so the close becomes automatic instead of something you're improvising under pressure.
Why the close carries so much weight
Interviewers are grading you as a future consultant giving a client an answer, not as a student who solved a math problem. A partner who spent forty minutes on analysis and then says "well, I found a lot of interesting things" gets fired. A partner who says "my recommendation is X, here is why, and here is what I'd want us to watch" gets asked back. The case interview recommendation is a compressed test of that exact moment, and it's graded on communication and judgment as much as on the accuracy of your numbers.
If you got the earlier parts of the case wrong, a strong close cannot fully save you. But if you got them right and then fumble the close, you lose credit for work you already did. That asymmetry is why I want you to practice this piece on its own, not just as an afterthought tacked onto case practice.
The three-part shape
Every strong case interview recommendation follows the same shape. Learn the shape once and it works for profitability cases, market entry cases, and anything else you'll see.
- The answer. One sentence, stated as a decision, not a hedge. "I recommend the client enter the market" or "I recommend against the acquisition at the current price."
- The reasons. Two or three supporting points drawn directly from the data you gathered, not new information you're introducing for the first time.
- The risk or next step. One sentence naming what could go wrong with your recommendation, or what you'd want to investigate next if you had more time.
That's the whole structure. It takes thirty to forty-five seconds to say out loud. Practice it until it feels short, because candidates almost always run long the first few times they try it.
Answer first, always
This is the single hardest habit to build, because your instinct after forty minutes of analysis is to show all your work. Resist it. State the recommendation in the first sentence, before any reasoning.
Compare these two openings for the same case. Weak: "So we looked at the market size, then we looked at the cost structure, and the margins were pretty tight, and then we also considered the competitive response, which seemed manageable, so..." Strong: "I recommend the client proceeds with the acquisition."
The weak version makes the interviewer wait for the point while doing mental work to extract it from a narrative. The strong version tells them exactly what you concluded, and now everything you say after is support for a claim they already understand. This is the same synthesize-first instinct you used when you opened the case, and it applies again at the close. If you want a refresher on that opening move, I cover it in how to open a case in the first sixty seconds.
Picking your two or three reasons
You did forty minutes of analysis. You have far more supporting detail than you need, and dumping all of it is a common way to undo a good answer. Pick the two or three points that carry the most weight, not the two or three that happen to be freshest in your memory.
A good filter: which findings would change your recommendation if they came back differently? Those are your reasons. A finding that was directionally supportive but wouldn't have flipped your answer either way is not worth a mention here, save it for if the interviewer asks a follow-up.
Keep each reason to one sentence with a number in it where you have one. "Margins in the new segment run at 22%, versus 14% in the client's current core business" does more work than "the margins looked favorable." Numbers signal that your conclusion is grounded in the case, not a gut feeling dressed up as analysis.
The risk or next step that shows judgment
The third piece is the one candidates skip most often, and it's the one that separates a good close from a great one. After your answer and your reasons, name one thing that could go wrong, or one thing you'd want to check next if you had more time on the case.
This single sentence does a lot of work. It shows you're not naive about your own recommendation. It shows you can think past the immediate question to what a real client would ask next. And it gives the interviewer a natural place to follow up if they want to keep going, which most of them do.
Examples: "The main risk is the competitor response, if they cut price to defend share, our margin assumption drops." Or: "Before I'd finalize this, I'd want better data on customer switching costs, since our recommendation leans on customers actually moving to the new offering." One sentence is enough. Do not turn this into a second case.
A worked example, start to finish
Take a profitability case where the client's revenue is flat and costs have crept up. Say your analysis found that a single underperforming product line accounts for most of the margin decline, and it also happens to be the highest-volume line.
A full case interview recommendation, spoken out loud, sounds like this: "My recommendation is that the client restructures or divests the underperforming product line rather than cutting costs across the board. Two reasons. First, that line alone accounts for roughly 70% of the margin decline, so broad cost cuts would hurt profitable lines to fix a problem concentrated in one place. Second, the line's input costs have risen 18% while its price has stayed flat, which means the economics have shifted structurally, not temporarily. The main risk is customer concentration, if the client's biggest account only buys because of that product, we'd want to check contract terms before making a final call."
Notice the shape: one sentence for the answer, two numbered reasons with figures attached, one sentence of risk. Under a minute, and it sounds like a person who is confident in a conclusion, not a person still processing out loud.
Recommendation habits that cost points
| Habit that costs points | What to do instead |
|---|---|
| Summarizing the whole case chronologically | Lead with the one-sentence answer first |
| Hedging with "it depends" and no actual call | Make the call, then note the condition as your risk line |
| Listing five or six supporting points | Cut to the two or three that would flip the answer if wrong |
| Introducing new data you never mentioned earlier | Only use findings you already surfaced during the case |
| Ending on the risk with no clear recommendation before it | Answer, reasons, risk, always in that order |
| Rushing through in ten seconds, sounding unprepared | Take a breath, thirty to forty-five seconds is normal |
That table covers the failure modes I saw most in real interviews. If any of these feel familiar from your own practice, they're worth drilling specifically rather than assuming better case reps will fix them on their own. I go deeper on the broader set of case-day errors in the ten most common case interview mistakes.
Practicing the close on its own
You don't need a full case to drill this. Take any business scenario, even a made-up one, and force yourself to state a one-sentence recommendation, two reasons with numbers, and a risk, out loud, in under a minute. Do this five times a day for a week and the shape becomes automatic, which matters because during a real case you'll be tired and a little adrenalized by the time you reach the close.
It also helps to keep a clean page of notes throughout the case specifically so you can build the close from what's written down rather than from memory alone. I cover the note-taking system that supports this in how to take notes in a case interview.
Interviewer-led cases change the timing, not the shape
In an interviewer-led case, you may not build to your own recommendation gradually, the interviewer can jump you straight to a final slide or ask for your view with less runway than you'd like. The three-part shape still applies, you just may have less material to draw your reasons from because the interviewer controlled which paths you explored. Don't let a shorter runway push you into skipping the answer-first habit, it matters even more when time is tight. The differences between the two formats, and how they change your approach throughout the case, are covered in interviewer-led versus candidate-led cases.
If your structure was solid heading into the close, the recommendation should feel like a formality, you're just naming the conclusion your own analysis already pointed to. That's what a strong case structure buys you: a close you don't have to invent under pressure.
The bottom line
A case interview recommendation is a short, three-part habit: state the answer first, back it with two or three reasons that have numbers attached, then name one risk or next step. Practice it on its own, separate from full case reps, until the order feels automatic. If you're still building the earlier skills that feed into a strong close, start with what a case interview actually is and work forward from there.
Go deeper
The full CaseMap method walks through the close alongside structure, math, and every other stage of the case, so the whole thing works as one system instead of separate tricks.
Get the complete Cut to the Case course →
One payment, lifetime access, 30-day guarantee.
Frequently Asked Questions
What is a case interview recommendation?
It's the closing statement of a case where you give your final answer to the client's question. A strong one states the recommendation in one sentence first, supports it with two or three reasons drawn from your analysis, and names a risk or next step, all in under a minute.
How long should a case interview recommendation be?
Thirty to forty-five seconds when spoken out loud. One sentence for the answer, two or three sentences for supporting reasons, one sentence for the risk or next step. Longer than a minute usually means you're re-summarizing the whole case instead of concluding it.
Should I state my recommendation first or explain my reasoning first?
State the recommendation first, always. Interviewers are grading you as a consultant who gives clients a clear answer, and leading with reasoning before the conclusion reads as unsure or unpracticed, even when the underlying analysis was strong.
What if I'm not fully confident in my recommendation?
Make the call anyway, and put your uncertainty into the risk line instead of hedging the answer itself. A recommendation with a clearly stated risk shows judgment, while a vague or wishy-washy answer with no real position shows the opposite.
Do I need new data for the risk or next step?
No. The risk should come from a gap or assumption already inside the case, something you noticed but didn't have time to fully verify. Introducing an entirely new consideration at the last second usually reads as scrambling rather than as insight.
Does the recommendation matter as much as the math in a case interview?
It's graded on its own dimension, communication and judgment, separate from analytical correctness. Strong math with a rambling close loses points that solid math with a clean close would have kept, since interviewers remember how a case ended almost as much as how it was solved.