Protocol Career
A glass sand timer rests beside a tidy stack of blank cards on a grey stone surface. Title: “Case Interview Prep Guide: The Confirmed Window Is 16 Days”, over the Protocol Career wordmark.

Case Interview Prep Guide: The Confirmed Window Is 16 Days

One posting publishes real dates: 39 days from application, 16 from confirmation. Case interview prep plans from there, forty hours against what firms assess.

9 min read
In short
  • One Accenture posting publishes dates. Applications close 18 October 2026, outcomes arrive by 10 November, and the selection bootcamp runs 26 and 27 November. That is 39 days from application and 16 from confirmation.
  • Sixteen days at an hour a day is 16 hours, which is 49.6 percent of the 32.25 hours needed to work all 43 published cases once. Thirty-nine days covers it.
  • Check first whether you will face a case at all. Deloitte's UK early careers process has four stages and none of them is a case interview.
  • Two firms publish opposite models of preparation. L.E.K. names exposure with no exercise, and Simon-Kucher names simulated conditions including time awareness.
  • Forty hours allocated against our own weights puts 18 on structure, because Coverage carries 45 of 100 in our structuring rubric.

One job posting publishes real dates, and they settle the planning question. Applications close 18 October 2026, outcomes arrive by 10 November, and the selection bootcamp runs on 26 and 27 November. Checked 30 September 2026. Sixteen days between knowing and doing is the window most candidates actually get.

Whether you need case interview prep, by country and role

Before building a plan, check the plan is needed. One firm's graduate process does not contain a case interview.

Deloitte's UK early careers page at deloitte.com publishes four stages plus the outcome: Apply Online, Immersive Online Assessment, Job Simulation, Final Stage Assessment, Outcome. It states: "Regardless of whether you're applying for a Summer Vacation Scheme, Industrial Placement, or one of our full-time Graduate or Brightstart programmes, our assessment process follows the same structure". The word case appears in none of the four.

What that process does require is different work entirely. The final stage has a discussion of about 30 to 40 minutes on a topic prepared in advance, plus a skills and motivation interview of about 50 minutes, and the preparation is compulsory: "Before you attend, we need you to complete some mandatory preparation which involves researching and conducting analysis on a topic." The same page publishes academic requirements: a 2:1 or better for graduate and work experience programmes, and 104 UCAS points across the three best A-Levels for BrightStart. The spellings of programmes and A-Levels are the firm's own.

Meanwhile the same firm publishes ten practice cases for US recruiting. Same employer, two countries, two exercises.

Three checks, in order, before you spend forty hours.

  1. Read the invitation. Every firm in our corpus defers to it. McKinsey states that the recruiter confirms the interview details in advance.
  2. Read the country page for the office you applied to. Not the global careers page, which generalizes.
  3. Read the posting itself. Accenture's company-wide process page never mentions a case interview, and one of its own postings names case interviews twice.

Bain draws the line by role rather than geography, stating that case interviews are used for consulting roles only.

Reading a worked example is not producing one.

Score my answer

Sixteen days covers half the 43 cases five firms publish

Here is the calendar, from a Copenhagen posting read through Accenture's own Workday system at accenture.wd103.myworkdayjobs.com.

Event Published date Days elapsed
Application deadline 18 October 2026 0
Outcome communicated by 10 November 2026 23
Recruitment bootcamp 26 and 27 November 2026 39

Recomputed by hand on 30 September 2026. From 18 October to 10 November is 13 days remaining in October plus 10 in November, so 23 days. From 18 October to 26 November is 13 plus 26, so 39 days. From 10 November to 26 November is 16 days.

Now put those against the work available. Firms publish 43 readable practice cases: Accenture 21, Deloitte 10, Oliver Wyman 6, McKinsey 4, Bain 2. At 45 minutes each that is 43 times 0.75, which is 32.25 hours.

  • The confirmed window of 16 days, at an hour a day, gives 16 hours. That is 16 over 32.25, which is 49.6 percent of the published material.
  • The full window of 39 days, at an hour a day, gives 39 hours, which covers all of it with slack.

So the decision is clear and it is the opposite of what most candidates do. Serious preparation starts when you apply rather than when you are invited, because the invited window buys you half the material. The posting says the same thing in its own words: "We therefore encourage you to familiarize yourself with consulting case interviews ahead of the bootcamp."

Deloitte publishes windows rather than dates, which point the same way: an initial interview "usually within one to two weeks" of selection, then "Expect two to three rounds of interviews", then an offer "in writing within one week of your final interview."

Two models of case interview prep, and where AI is allowed

Two firms publish preparation advice and the two models barely overlap.

L.E.K.'s page at lek.com gives four suggestions: attend campus case preparation workshops, read the business press regularly, practice with people who currently work in strategy consulting and understand the firm's type of work, and "Review the case studies and other content available throughout our website". Not one of the four names a timed exercise. It is a model of exposure.

Simon-Kucher's page at simon-kucher.com gives four of its own: discuss cases out loud, simulate interview conditions including time awareness, reflect on your approach and iterate, and build business acumen through the business press and the firm's publications. Two of those four are about deliberate practice, and one names the clock.

Both models are published by firms that hire. Our reading, and we label it as ours: the exposure model builds the vocabulary and the practice model builds the execution, and a candidate who does only the first meets a timer for the first time on the day.

McKinsey adds a boundary that matters for how you rehearse. Its interviewing page permits using AI to polish a resume, practice questions and explain a concept, and prohibits generating answers in real time during an interview or using AI during an assessment without explicit permission. So rehearsal assistance is explicitly allowed by one firm, and live assistance is explicitly not.

Forty hours of case interview prep, block by block

Forty hours is a realistic total for someone working or studying full time, spread over six weeks. Here is an allocation, and the allocation logic is ours.

Block Hours What it covers
Structure 18 openings on 24 prompts, branch design, justification sentences
Arithmetic 9 chains under time, unit discipline, sanity checks
Exhibits 6 published exhibit questions, share against share reads
Synthesis 4 closes timed at 60 seconds
Fit stories 3 eight stories written and said out loud

Eighteen plus nine plus six plus four plus three is 40, checked by addition.

The structure share is 45 percent of the total, which is 18 over 40. That number is taken from our own structuring rubric, where Coverage carries 45 of 100. It is our weight rather than any firm's, and the allocation follows from it directly: the largest scored line gets the largest block.

Read that against what firms publish and it holds up. Every one of the four published in-case sequences opens on comprehension and structure, and none opens on arithmetic. Deloitte's five steps put "Understand the issue" first. L.E.K.'s seven put Take your time first. Accenture's six put Listen to case first. Oliver Wyman's six put pinpoint the issue first.

The fit block looks small at three hours, and it is enough for the published requirement. McKinsey asks for "two personal examples that demonstrate how you've made an impact in each of the four areas below", which is eight stories. Three hours writes eight stories and says each one out loud twice.

The arithmetic and exhibit blocks are the two that reward raw repetition most, and a subscription aimed at exactly those exists. Our RocketBlocks review reads its published prices and quotes the help center entry where it explains why it drills components and leaves whole cases to people.

Can you hold the basics under a clock?

What to do in the first six hours

Order matters more than volume at the start, because the wrong first move wastes the next ten hours.

  1. Read four firms' published instruction sets. L.E.K.'s seven steps, Deloitte's five, Oliver Wyman's six, Accenture's six. Ninety minutes, and they disagree in four places you should notice.
  2. Read one worked case as a reader. One of McKinsey's four, which the firm says contain "logic, thought processes, and suggested answers". Thirty minutes.
  3. Do one case badly, out loud, timed at 30 minutes. The published lower bound for a case portion. Doing it badly is the point.
  4. Write your first opening and time it. Restate, assumptions, structure, justification. Ninety seconds, twenty attempts. That is ninety minutes and it is the single highest-return block in the forty.
  5. Write the eight fit stories as titles only. Thirty minutes, so the bank exists before you need it.

Six hours, and at the end of them you can open a case competently. That is a better state than having read twenty cases silently.

What to do in the last six hours

The last six hours before an interview should contain no new material.

  1. Four cases in one sitting. Two firms publish four-case rounds, so this is the realistic load, and you will find out what case four feels like.
  2. Twenty calculation chains, timed at three minutes each. One hour, and it resets your error rate.
  3. Say your close ten times. Call, two figures, risk with direction, next step with timeframe. Deloitte's fifth published step is the one most closes omit.
  4. Reread the four contradictions and pick your side. Questions against crutch, framework against no framework, hold against adapt, coachable against firm. Decide in advance rather than under time.
  5. Check the mechanics. One firm's remote instruction is to disable previously installed AI note-taking software, which is the item nobody remembers.
  6. Copy the equations out by hand. Profit, contribution, breakeven, margin, markup, compound growth, payback. Copying is the last rehearsal.

Nothing in that list is new learning, which is deliberate. The last six hours are for retrieval speed.

The weights that decide the allocation

The four numbers below are ours, read from the grading code in src/lib/grading. No consulting firm publishes a marking scheme, which is why an allocation has to be justified rather than asserted.

Structuring, weighted
  • Coverage45
  • Separation25
  • Depth20
  • Concision10
Case mathAnswer 40 · Working 25 · Units 15 · Interpretation 20
Idea generationBreadth 30 · Buckets 30 · Specificity 20 · Prioritisation 20
Market sizingScope 25 · Assumption chain 30 · Arithmetic 25 · Sanity check 20
ExhibitsObservation 25 · Implication 25 · So what 30 · Next test 20
SynthesisAnswer first 35 · Evidence 25 · Risk 20 · Next step 20

Coverage at 45 is why structure gets 18 of the 40 hours. It is the largest single line in any of our rubrics, and it is the only one that cannot be recovered later in a case: a branch you never named does not come back.

Separation at 25 is why branch design gets its own sessions inside that block. Overlapping buckets are the product of a template applied to the wrong business, and the fix is practice on varied prompts rather than more repetitions of one frame.

Depth at 20 is why the drill is 24 prompts rather than 24 repetitions of one. A second level under each branch is what makes a branch usable.

Concision at 10 is why none of the forty hours goes to shortening things. It is the cheapest fault available.

Our math rubric puts Answer 40, Working 25, Units 15 and Interpretation 20, which is the justification for the nine-hour arithmetic block being about chains and units rather than raw speed. The reasoning behind every weight sits on how an answer is assessed.

Our structure drill, math drill and synthesis drill each report their lines separately with the first correction named, and all three are free with no account. None of them talks back, so the conversation has to be practiced with a person.

What this guide will not tell you

Four things, and each is an absence in the public record rather than something we are withholding.

How many rounds you will have. Published in three places only: Deloitte says two to three, Oliver Wyman's Brazil and Mexico page says two rounds of two, BCG Switzerland says one plus three. McKinsey, Bain, Kearney, L.E.K. and Simon-Kucher publish nothing.

How long your interview will be. Two published clocks exist, one from a 2018 document and one from a single country page.

What score you need. Nobody publishes a threshold, a weight or an aggregation rule. Bain refers to "an objective scoring system" and describes none of it.

Whether more cases beats fewer. We have not measured it, and no firm publishes anything on it. What is measurable is the material: 43 published cases, 32.25 hours to work them once.

For the day-of version of this plan, see the case interview checklist. For the cases to work, see case interview examples. For MBB specifics, see MBB case interview prep.

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-interviewpreparationplanning
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.