Protocol Career
A blank cardstock sheet debossed with a four-section grid sits on a grey surface in soft light. Title: “How to Practice Cases Without Wasting the Repetitions”, over the Protocol Career wordmark.

How to Practice Cases Without Wasting the Repetitions

Firms publish the material and two timing constraints. None publishes how to practice, so here is a session structure with the arithmetic behind its numbers.

8 min read
In short
  • McKinsey publishes four full cases with suggested answers and says practicing them builds familiarity with the types of questions it asks.
  • Bain publishes the only structuring clock we found: one to two minutes before you speak.
  • EY publishes the only volume figure we found in any firm source, and it is about stories rather than cases: prepare 10 to 15 examples you can adjust on the fly.
  • EY also forbids the obvious shortcut. Practice go-to stories out loud, and do not read responses from a script.
  • The session structure and the review rule below are ours, and the page names the weights they come from.

The learning happens in the twenty minutes after a case rather than the forty minutes inside it. The firms publish the material and two timing constraints; the session structure below is ours, and it exists because a case you cannot score is a case you cannot learn from. Sources checked 30 September 2026.

Where the practice cases legitimately come from

Before any method, the supply question. Management consulting practice cases are published by the firms themselves, free and with no account required, and five sources carry them.

Source What it gives you Limit
McKinsey's four practice cases Full cases with suggested answers Four cases, one firm's style
Bain's two published cases A worked break-even and a revenue case, with good and better answers Two cases
Deloitte's LeadAuto case One case with a two-tier answer key, good then great One case, US recruiting site
BCG's case preparation page Five named tasks and twelve behaviors No case to work
A university career office Workshops, drop-ins, review clinics Your own institution only

Say which is which, because the four names people search for point at four different objects. The McKinsey practice cases are four full cases with suggested answers attached. The Bain practice cases are two, a break-even and a revenue case, each printed with a good answer and a better one. The Deloitte practice case is one, LeadAuto, and its answer key splits into a good tier and a great tier, which tells you where the bar sits rather than only whether you cleared it. There are no BCG practice cases to work through: its preparation page carries five named tasks and twelve behaviors and no case, and its job simulation modules live elsewhere, covered in the BCG X guide.

Add them up: four from McKinsey, two from Bain, one from Deloitte, so seven cases in total across three firms. That is the honest supply figure, and a plan built on forty repetitions has to find the other thirty-three somewhere other than a firm.

McKinsey's interviewing page recommends reviewing its cases and says they contain logic, thought processes and suggested answers, and that practicing them builds familiarity with the types of questions the firm asks. That is an invitation, published by the firm, to use its own material.

One boundary applies to everything else you might find. McKinsey's assessment integrity expectations prohibit sharing information about the content of an assessment or interview, including specific questions and problem-solving scenarios. Material circulating from someone else's sitting sits on the wrong side of that line, whatever site it appears on.

Reading a worked example is not producing one.

Score my answer

How to practice a case, in four blocks

Ninety minutes per full case, of which the case itself is forty. This split is ours.

Block Minutes What happens
Setup 5 Paper out, timer set, calculator away
The case 40 Uninterrupted, spoken aloud, no pausing to look things up
Scoring 20 Score your own artifacts against one rubric
Repair 25 Redo the single worst moment, three times

The scoring block is the one candidates delete when short of time, and deleting it converts the session into exercise. The repair block is the one nobody does at all.

Bain's FashionCo case sets the clock inside block two. Take one to two minutes to think about how you would approach the problem, it says, then walk your interviewer through your structure and analysis step by step. Practice with that window enforced rather than approximated, because the difference between ninety seconds and four minutes is invisible from the inside.

What to score, and against what

Score artifacts rather than feelings. Four things survive a case on paper and each maps to a weight in our own rubric code, published on our method page.

  1. The structure you wrote. Score Coverage first, weighted 45 of 100, because a missing branch costs more than an untidy one.
  2. The arithmetic chain. Answer carries 40, Working 25, Units 15, Interpretation 20. Score units before speed.
  3. The exhibit sentence. So what carries 30 against Observation at 25, so a correct description with no conclusion scores under half.
  4. The closing thirty seconds. Answer first carries 35 on the synthesis rubric, and an answer arriving in sentence four scores nothing on it.

Those weights are ours, read from our own code. No consulting firm publishes a mark scheme, and the closest published thing is Bain labeling one model answer good and a fuller one better on its own case page.

The repair block, which is the whole point

Twenty-five minutes on one moment. Pick the worst artifact from the scoring block and redo only that.

  1. Say the same structure again, in ninety seconds. Most first attempts are three minutes of hedging around two useful minutes.
  2. Redo the arithmetic out loud, with units in every line. Per bottle, per gallon, per household. The unit error is the one that survives repetition because the numbers look right.
  3. Rewrite the exhibit sentence as a decision. What would you test next, given that reading.
  4. Deliver the synthesis cold, twice. Answer, three reasons, one risk, one next step.

EY's interview tips page endorses the mechanism and warns about its failure mode in the same passage. It tells candidates to practice go-to stories out loud, and then says not to memorize or read responses from a script, because you do not want to sound too rehearsed. Repeating a structure until it is fast is useful. Repeating a wording until it is fixed is the thing the firm is warning against.

One more scoring habit is worth building. Write the score down with a date beside it. A single session tells you almost nothing, and four sessions across two weeks tell you which dimension is actually moving. Without the dates you are left with an impression, and impressions after a hard case are reliably worse than the artifacts.

Can you hold the practice under a clock?

How often to practice cases, and how much

No firm publishes a case count. The only volume figure we found in any firm source is EY's, and it is about behavioral material rather than cases: prepare 10 to 15 examples you can adjust on the fly to suit a question.

Do the arithmetic on that, because it sets a floor for the other half of your preparation. McKinsey suggests two personal examples in each of four themes, which is eight distinct stories. EY's range starts at 10. So the binding number is 10 to 15, and eight is the minimum rather than the target.

For cases, our working rhythm is below. It is ours, and no source endorses it. The drill hours are case math practice and exhibit reading in short blocks, which is why they outnumber the full cases in every row.

Weeks out Full cases per week Drill hours per week
8 or more 1 4
4 to 8 2 4
2 to 4 3 3
Final week 2 2

The count falls in the last week on purpose. A candidate three days out gains more from rehearsing four structures cold than from meeting a fifth industry.

The mistakes that waste case interview repetitions

BCG's case preparation page lists six behaviors to avoid: panicking under pressure, rushing your response, overcomplicating solutions, ignoring feedback, staying silent and interrupting. Four of those six are practice artifacts rather than character.

  1. Practicing silently. Staying silent is on BCG's avoid list, and a case worked in your head trains the opposite of what the room requires.
  2. Pausing the timer to look something up. The case becomes a reading exercise and the clock stops meaning anything.
  3. Keeping a calculator in reach. McKinsey's integrity page allows only pen and paper for calculations in its interviews.
  4. Ignoring the feedback you collected. BCG names this one directly. An unread scoring sheet is worse than none, because it cost you twenty minutes.
  5. Chasing case volume in the last week. Familiarity peaks early and delivery peaks with rehearsal.

Two of those five have the same cause, which is that a full case feels like work and a scoring sheet feels like homework. The case is the cheaper half. Cases are free from two firms and from our own library, while twenty minutes of honest scoring is the part nobody will do for you.

The calculator habit deserves separate attention because it hides. A candidate who drills arithmetic with a phone in reach is practicing a different task: reading a number off a screen rather than holding a chain in working memory. McKinsey's rule is explicit and it applies in the room, where the phone is gone.

What a method cannot supply

It cannot supply cases. Four from McKinsey, two from Bain and one from Deloitte is seven, and the firms publish no more than that. Our case library exists for exactly that gap, and the free prep resources page lists everything official we found at no cost.

It cannot supply a second voice. The scoring block substitutes for a partner and it substitutes badly on one dimension, because nobody interrupts you. Our case practice partner page covers how to make a non-expert useful, and practicing cases alone covers what degrades when you cannot find one.

It cannot make the session happen. Ninety minutes is a real commitment and the honest planning question is how many of those you have between now and the round, which our application tracker answers by holding the dates in one place.

And it cannot tell you when you are ready. No firm publishes a threshold, a score or a passing bar, so any page that offers you one is offering a number nobody has.

See your answer scored, dimension by dimension.

Reading a worked example and producing one are different skills. Work a full case and see it scored, dimension by dimension.

Sources

Primary sources only: the firm, the test vendor, or the institution concerned. Each carries the date we last read it.

case-practicepreparationmethodcase-interview
Written and reviewed by
Elias MarchandStrategy editor

Edits the firm and case guides on this site. Reads each claim about a recruiting process back to the firm's own page before it is published, keeps the source ledger dated, and owns the corrections.

How this guide was made

  • Written against a published editorial process, not from memory. Every claim about a firm's process is either linked to that firm's own page or absent. We say what we could not verify rather than filling the gap.
  • Each page carries a dated source ledger, below. When a firm changes a programme or a deadline, the date tells you how far to trust what you just read.

We look for first-hand candidate accounts for every firm guide. Where we found none we could source, the page says so and carries the firm's own words instead. Nothing here is dressed up as testimony, and if your own experience contradicts what you read, tell us and we will correct it and date the change.

Something wrong or out of date here? Tell us and we will correct it and date the change.