L.E.K. Case Interview Examples: 9 Practice Cases and Market Sizing Walkthroughs (2026)

L.E.K. Consulting runs case interviews across 27 offices and roughly 2,300 employees, and it publishes exactly three case examples on its own careers site with no answers attached: an aircraft brainteaser, a brewery profit-volatility case, and a pharmaceutical market sizing and pricing case. Two of those three require an estimate, which tells you what the firm weights. L.E.K.'s official application page says the process "differs slightly by region" and includes "a mixture of experiential and case interviews (both quantitative and strategic cases are possible)". Candidate reports describe two 30-minute cases in round one and a written case of about 60 minutes in the final round.

The failure mode specific to this firm is not weak structure. It is treating the opening estimate as a warm-up. At L.E.K. the number you build in the first eight minutes becomes the denominator for everything after it: the pricing decision, the investment recommendation, the go or no-go. Candidates who rush the sizing to "get to the real case" spend the rest of the interview defending a figure they never pressure-tested. This page works nine cases end to end, starting with L.E.K.'s own three, and ends each one in a decision rather than a framework.

L.E.K. practice cases at a glance

How an L.E.K. case works: it usually starts with a number. You size a market or a customer pool, then build the pricing, investment or go/no-go decision on top of that estimate. Candidate reports describe two cases of about 30 minutes each in round one and a written case of about 60 minutes in the final round, and L.E.K. says the process differs slightly by region.

Start free now: iPhone X-Change is an L.E.K.-style market sizing case: a US reseller of used phones, tablets and laptops wants to know how big its market is. It has 6 questions, 2 with math, 1 exhibit, and fewer than 10 completed runs in the last 90 days, so no median time yet. An AI interviewer runs it in voice mode: it asks each question out loud, shows the exhibit when you ask for the data, and scores a debrief on seven categories. Your first full case is free when you create an account; the worked excerpt below runs its sizing. A free account also gets one full run in every simulator inside the app: one complete run of each of our four McKinsey Solve practice formats: Red Rock, Sea Wolf, SFL Classic and SFL Island, a full Casey-style case, a full CCA-style diagnostic and a complete Bain SOVA-style diagnostic.

For Solve, you can start without an account; create a free account to finish and keep your score, points breakdown and worked feedback on submitted decisions, with visual explanations, without paying. Learning mode provides step-by-step coaching as you go, untimed in the clocked games, with the matching game pack, Solve pack or qualifying Annual access. The free Solve plan uses invitation length and test date for dated practice steps over up to 28 days; a free account is needed to view and save it.

iPhone X-Change (adapted from the Duke Fuqua consulting club casebook) is the only L.E.K. case in the library of 71. The other three are practice cases written in another firm's style and chosen for the mix L.E.K. tests, sizing, pricing and private equity growth; the style column says which. Every debrief scores seven categories: structure (25% of the score), quantitative analysis (20%), business judgment (15%), and problem understanding, exhibit reading, communication and synthesis (10% each), with feedback on every question. Median times come from completed runs between June 25 and September 22, 2026, and count runs, not people. SurfacePro and BeanCraft run in guided mode only.

Prep plans for McKinsey, BCG, Bain, Deloitte, Accenture, EY-Parthenon, Kearney, Oliver Wyman, Roland Berger, L.E.K., OC&C, Strategy&, Simon-Kucher, Arthur D. Little, Capital One and your target firm. Choose your firm in the plan builder.

Get your prep plan

A week-by-week schedule built around your target firm and interview date.

Build my prep plan

1

Learn the format

17 steps

• Learn how to caseGuided learning · 35 min
• The Case Interview MethodLesson · 8 min
• Profitability & Turnaround: AfrihealSkill drill · 5 min
• Profitability & Turnaround: Berty BankSkill drill · 5 min

What a good L.E.K. practice example has to show

Most published L.E.K. material stops at the prompt. That is useful for exposure and useless for calibration, because you never learn whether your estimate chain or your recommendation would have survived. A practice case earns its place only if it carries five things.

Use the same loop on every case below: state the decision, build the first tree, ask for the highest-value evidence, then close with a recommendation that has a number attached to it. If you want the format and process detail rather than the material, the L.E.K. case interview guide covers rounds, timings and what the invitation controls.

How L.E.K. cases differ from MBB and other tier-2 firms

L.E.K. is not "BCG with fewer offices". Its case mix reflects its book of business, which is heavily weighted toward commercial due diligence, life sciences and consumer, and that shows up in the questions.

What this means for your practice mix. If you have only run generalist profitability cases, you will feel the gap in two places: sector vocabulary and estimate discipline. Fix the first with the life sciences consulting case interview guide and the private equity due diligence framework. Fix the second by getting an estimate graded rather than self-marked, because the part you cannot assess alone is whether your driver chain would have survived a challenge.

The five question types inside an L.E.K. case

An L.E.K. case is a sequence of separately scored asks, and each one is rehearsable on its own before you ever sit a full case.

L.E.K.'s published in-case guidance maps almost one to one onto that list. Its seven tips tell candidates to take time and restate the facts, gather thoughts before speaking, stick to the approach, ask clarifying questions, formulate hypotheses aloud, "come up with an answer" while showing willingness to rethink, and be yourself. Read together, that is a request for a candidate who commits and then updates, not one who hedges until the interviewer decides for them.

L.E.K.'s three official case examples, decoded

These are the only cases L.E.K. publishes, and it deliberately gives approach questions instead of answers. Work each prompt cold before reading the solution.

Official case 1: the aircraft brainteaser

L.E.K.'s prompt. Airplanes take off from JFK and LAX every hour on the hour, continuously, flying identical New York to Los Angeles routes. The flight takes five hours. You leave New York at 12:00 PM EST. How many airplanes flying the opposite route will you pass on the way?

The trap. Most candidates answer five, counting only the aircraft that depart LAX during their flight. That ignores every plane already airborne when you take off.

Worked approach. Work in one time zone and put every departure on a single clock. You are in the air from 12:00 to 17:00 EST.

1. An eastbound plane that left LAX at 07:00 EST lands at JFK at 12:00 EST, exactly as you take off. That is the earliest one you encounter.
2. An eastbound plane that leaves LAX at 17:00 EST departs exactly as you land. That is the last one you encounter.
3. Every hourly departure between those two endpoints is in the air at some point during your flight: 07:00, 08:00, and so on through 17:00.
4. That is 11 aircraft. If the interviewer wants only planes you pass mid-air, exclude the two endpoint encounters and answer 9.

What is actually being scored. Not the number. It is whether you convert to a single clock before counting, whether you name the endpoints explicitly, and whether you volunteer the 9 versus 11 boundary condition instead of being caught by it. For more of this format, see the consulting brain teasers guide.

The library's airline case extends the transport setting into a commercial decision, rather than repeating L.E.K.'s brainteaser.

Official case 2: Brewery Profits

L.E.K.'s prompt. Two competing brewers, Cheapo Beer Company and Bourgeois Beer, Inc., have reported similar revenues over five years with steady growth. Cheapo is the "All-American Beer" staple at tailgate parties. Bourgeois markets itself as brewed fresh from mountain water. Cheapo's profits have grown at a moderate, consistent rate. Bourgeois's profits have fluctuated significantly. Why the difference, and as CEO of Bourgeois, what would you do about it?

The structure that works. Revenue is similar and steady for both, so the volatility is not a demand story. That collapses the tree fast. Profit volatility with stable revenue has three possible homes: input cost exposure, operating leverage (the fixed versus variable cost split), and discretionary spend that is lumpy year to year, mainly marketing and promotion. Say that out loud, then ask for the cost structure.

Worked math (case assumptions, stated as such). Assume both brewers sit at $500M revenue.

• Cheapo: contribution margin 40%, fixed costs $80M. Operating profit = ($500M x 0.40) − $80M = $120M.
• Bourgeois: contribution margin 55% (premium pricing), fixed costs $170M (own brewery, spring water sourcing, glass packaging). Operating profit = ($500M x 0.55) − $170M = $105M.
• Now flex revenue down 5%. Cheapo loses $25M x 0.40 = $10M, an 8.3% profit hit. Bourgeois loses $25M x 0.55 = $13.75M, a 13.1% profit hit on a smaller base.

Add the input exposure. If Bourgeois buys $120M of ingredients and packaging on the spot market and those prices swing 15% year to year, that alone is a ±$18M profit swing, roughly 17% of operating profit, before volume moves at all.

Recommendation with numbers attached. Two levers, in order. First, hedge or forward-contract about 70% of the input basket: residual exposure falls from ±$18M to roughly ±$5.4M. Second, convert fixed capacity to variable by contract-brewing peak volume instead of building for the peak, which lowers the fixed base and shrinks the operating-leverage amplifier. Smoothing marketing spend across quarters is a third-order fix, worth naming and not leading with. Risk to flag: hedging locks you out of input price declines, so the case for it is variance reduction, not expected profit.

That decomposition is the profitability framework applied to volatility rather than level, and it is the single most reusable move in the L.E.K. set. The part candidates get wrong is the structure, not the arithmetic: they list cost categories instead of naming the three mechanisms that can produce volatility under stable revenue.

Official case 3: Large Pharma Co., the baldness cure

L.E.K.'s prompt. A large pharmaceutical client has developed a pill, IPP2, that regrows hair to age-15 thickness within three months and must be taken daily to maintain it. Estimate the size of the market for this drug and describe how you would price it.

This is the case that tells you the most about L.E.K., because the estimate is not the answer. It is the input to the pricing decision, and a wrong denominator makes every price you propose indefensible.

Worked approach, top down. Every number below is a case assumption you should state aloud, not a researched figure. The interviewer is buying the chain.

1. Start with US adults: assume roughly 260 million, split about evenly, so 130 million men and 130 million women.
2. Prevalence of noticeable pattern hair loss: assume 40% of adult men across all ages, giving 52 million, and 20% of adult women, giving 26 million. Total pool: 78 million.
3. Not everyone with hair loss wants a daily pill. Screen for cosmetic priority, tolerance for a chronic medication, and ability to pay a cash price: assume 25%, giving roughly 19.5 million willing patients.
4. Adoption is gated by awareness, prescriber access and switching from existing routines. Assume the drug captures 20% of the willing pool at steady state: 3.9 million patients.
5. Pricing. This is cosmetic, so assume no insurance reimbursement and a cash-pay decision. Anchor between the two existing options: daily generic therapies run tens of dollars a month, surgical restoration runs several thousand dollars once. A daily pill with a visible three-month result sits closer to the surgical value, so price at $100 per month, or $1,200 a year, and validate with willingness-to-pay research before launch.
6. Revenue: 3.9 million x $1,200 = roughly $4.7 billion a year at steady state.

Sanity check. 3.9 million patients is 5% of the 78 million pool and about 3% of adult men. For a cash-pay cosmetic therapy that is a plausible ceiling, not a conservative floor, so flag it as the assumption most worth testing. If the interviewer pushes the capture rate to 10% of the willing pool, revenue halves to $2.3 billion, and that sensitivity is the answer worth volunteering.

How to close it. Price on value, not cost, because the marginal cost of a pill is irrelevant to what a cosmetic buyer will pay. Recommend a single cash price with an annual prepay discount to lift retention, since revenue depends entirely on daily adherence over years rather than a one-time purchase. Name adherence as the primary risk and the first thing you would measure after launch. If your estimate chain felt loose here, the method is in market sizing step by step, and the practice inventory is in the best market sizing practice resources.

The arithmetic in the pricing half is where candidates slip, because it is easy until the interviewer flexes one assumption and asks for the new number out loud.

Keep the original patient pool fixed, change the adoption share, then recompute revenue before you compare the pricing options.

Quick Math

1. What is 14% of $450M?

   Build the percentage from 10% and 1% blocks instead of multiplying directly.

2. Margin moves from 17% to 26%. What is the change in percentage points?

   Percentage-point change = Ending percentage - Starting percentage, kept in points, not converted to a relative percent.

3. Revenue goes from $140M to $175M. What is the percent growth?

   Percent growth = (Ending value - Starting value) / Starting value.

Carry that sensitivity check into MedSuite's hospital software pricing decision: work through how each option changes revenue, explain which customer assumptions you would test, and use the case's own inputs to defend a pricing model you could recommend to the client.

Worked excerpt: sizing the used-device market in iPhone X-Change

This excerpt runs the sizing step of iPhone X-Change from the table above, with the inputs from its own exhibit. Build your own sizing route before you read the table.

The prompt. Electronics Co. buys used phones, tablets, laptops and other devices from consumers and resells them online. It wants to size the total US market opportunity across used-device categories.

A structure that works. Define the scope first: which categories, which geography, which time basis. Then pick one route and say why. A population funnel works when you have household data; here the exhibit gives unit volumes, so the operational route is volume x share that is used x share that is US x average selling price, category by category, plus a rule for the categories the exhibit does not list.

The math. Exhibit 1 gives the four inputs per category, and the case sizes "Other" at 20% of the named categories.

What a strong answer says. "On these inputs the used-device market is about $2.1B, and phones are $1,250.0M of it, 59.6%, so phones are where the client should compete first. The exhibit does not say what period or starting geography the volumes cover, so before anyone calls this an annual US market I would confirm that basis, then pressure-test the used and US shares, because every later call about where to compete sits on the number."

Rebuild a volume-to-value chain on fresh inputs, keeping the time period and units explicit as you did for the device categories.

Quick Math

1. A region has 8 million households. 25% buy the category and spend $160 a year. What is the annual market size, in dollars? Answer in $M.

   Market = Households x Buying rate x Annual spend per buyer.

2. A service has 5 million users paying $11 a month. What is annual revenue, in dollars? Answer in $M.

   Annual revenue = Users x (Monthly price x 12).

3. A retail format has 240 stores. Each serves 150 customers a day at an $18 average ticket, operating 350 days a year. What is annual market revenue, in dollars? Answer in $M.

   Build one store's annual revenue first: Customers per day x Ticket size x Operating days, then multiply by the store count.

Six more L.E.K.-style practice cases with worked solutions

These are written in the sectors L.E.K. actually staffs. Read the prompt, cover the solution, and give yourself two minutes to structure before you look.

Case 4: Commercial due diligence on a veterinary clinic roll-up

Prompt. A mid-market private equity fund is considering a $240M acquisition of a 60-clinic veterinary group. The platform grew revenue 14% last year. Should the fund proceed, and what would change your answer?

Structure. Three testable branches: is the market growth real and durable, is the platform's growth organic or acquired, and does the roll-up thesis have anything left to buy.

Worked approach.

• Decompose the 14%. If 9 points came from four clinics acquired mid-year and 5 points from same-clinic growth, the platform is buying growth, not producing it. Ask for same-clinic revenue growth as your first exhibit; it is the single number that decides this case.
• Test same-clinic growth quality. Split it into visits and revenue per visit. If visits are flat and revenue per visit is up 5%, growth is pricing, and pricing has a ceiling in a discretionary spend category.
• Test the roll-up runway. At 60 clinics with an average purchase price of 8x EBITDA and a platform valued at 14x, the arbitrage is real but only while targets exist. Ask how many independent clinics remain in the platform's geographies and whether competing consolidators are bidding the multiple up.
• Size the return. If EBITDA is $30M (a 12.5% margin on $240M of revenue at an 8x entry) and the fund can add 20 clinics at 8x with $1.5M EBITDA each, that is $30M of acquired EBITDA for $240M of capital, revalued at the platform multiple.

Recommendation. Proceed only if same-clinic growth is above 4% on visits rather than price, and if at least 40 acquirable clinics remain within the existing footprint. Otherwise the fund is paying a platform multiple for an arbitrage that closes before exit. Principal risk: veterinary labour supply, which caps how fast acquired clinics can be integrated.

This is the archetype L.E.K. sells more of than any other. The transferable skill is running the growth decomposition before touching valuation, and it rehearses cleanly on a real private equity growth case.

Case 5: Life sciences in-licensing decision

Prompt. A mid-cap biopharma company can in-license a Phase 3 asset for $150M upfront plus $300M in milestones. Peak annual sales are projected at $600M. Should they take it?

Structure. Probability of approval, commercial realism of the peak sales figure, and whether the client can actually sell it with the salesforce they have.

Worked approach.

• Risk-adjust first. If Phase 3 assets in this indication approve at roughly 60%, the $600M peak becomes a $360M expected peak before you discount anything.
• Build the revenue shape, not just the peak. Assume a four-year ramp to peak, seven years at peak, then a patent cliff. Ten years of contribution at a 30% net margin on the risk-adjusted number is roughly $1.0B undiscounted, and materially less after discounting at a biotech cost of capital.
• Compare with the cost. $150M upfront is spent regardless of outcome. Milestones are contingent, so weight them by the same 60% and by the sales thresholds attached to them.
• Ask the question that decides it: does the client already call on the prescribers this drug needs? If yes, the incremental commercial cost is close to zero and the deal works at these terms. If no, add the cost of a new specialty salesforce and the deal likely does not.

Recommendation. Take it only if the target prescriber overlap with the existing portfolio exceeds roughly 60%, and negotiate more of the $300M into approval-contingent milestones rather than upfront. The sensitivity worth naming is probability of approval: at 45% instead of 60%, the deal stops clearing its cost of capital. Sector concepts behind this case are in the life sciences consulting case interview guide.

Case 6: MedTech sizing opener into a pricing decision

Prompt. Before we discuss strategy, how many hip replacement procedures are performed in the United States each year? Our client makes hip implants and is considering a premium-priced ceramic bearing.

Worked approach, top down. Again, these are stated assumptions.

• US population 330 million. The at-risk group is adults 65 and older: assume 17%, so 56 million.
• Assume roughly 0.6% of that group receives a hip replacement in a given year: 336,000 procedures. Add younger patients from trauma and early-onset arthritis at about 25% of that figure: roughly 420,000 procedures a year.
• Client share at 20% gives 84,000 implants. If the premium bearing carries a $1,200 price uplift and converts 30% of the client's volume, that is 25,200 units x $1,200 = $30M of incremental revenue.

How the sizing decides the strategy. $30M of uplift against a development and regulatory cost of, say, $45M means the product needs three years to pay back before competitive response. If instead the conversion rate is 60% because surgeons in this segment default to premium bearings, payback halves and the decision flips. So the strategic question is not "is the market big" but "what fraction of surgeons choose the bearing rather than the hospital procurement team", and that is the exhibit to request.

Quick Math

1. A market has a TAM of $8,000M. SAM is 35% of TAM. What is SAM, in dollars?

   SAM = TAM x SAM share.

2. Using the same $8,000M TAM, you can serve 35% of it and expect to capture 6% of that segment in three years. What is SOM, in dollars?

   SOM = TAM x SAM share x Capture rate, applying the capture rate to SAM, not TAM.

3. A three-year revenue target of $126M sits against the $2,800M SAM. What share of SAM does that require, in percent? Round to the nearest 0.1 percentage point.

   Required share of SAM = Revenue target / SAM.

Case 7: Industrials profitability, specialty coatings

Prompt. A specialty coatings manufacturer has grown revenue 9% over two years while operating margin fell from 16% to 11%. The CEO wants to know why.

Structure. Mix, price, and cost-to-serve. Not the generic revenue-minus-cost tree, because the case has already told you revenue is up.

Worked approach.

• Ask for revenue and margin by product line. If aerospace coatings at a 28% margin were flat while general industrial coatings at a 7% margin grew 30%, the margin decline is mix, and the growth is the problem rather than the cure.
• Quantify it. Assume revenue moved from $400M to $436M, with general industrial rising from $120M to $156M. Blended margin at the old mix would have been about 15.7%; at the new mix, roughly 12.9%. Mix explains most of the five-point fall, and the residual is real cost inflation.
• Chase the residual. Raw material index up 8% with pricing up only 3% means the client is absorbing input inflation on annual contracts with no pass-through clause.

Recommendation. Reprice or exit the lowest-margin general industrial accounts, add raw-material pass-through clauses at contract renewal, and reset the sales incentive from revenue to gross margin dollars. Quantify the ask: recovering three of the five lost points on $436M is roughly $13M of operating profit. See the profitability framework for the underlying tree.

Quick Math

1. A client books $180M of revenue against $126M of total cost. What is its profit margin, in percent? Round to the nearest whole percent.

   Margin = (Revenue - Total cost) / Revenue.

2. A division books $240M of revenue at a 15% profit margin. What is profit, in dollars?

   Profit = Revenue x Margin.

3. The same $180M revenue business wants to lift its margin from 30% to 34%. What total cost level does that require, in dollars?

   Cost = Revenue x (1 - Target margin).

Case 8: Consumer growth, subscription pet food

Prompt. A direct-to-consumer fresh pet food brand at 180,000 subscribers wants to reach 500,000 in three years. What is the path, and what does it cost?

Worked approach.

• Net adds needed: 320,000 over 36 months. But churn eats gross adds first. At 3.5% monthly churn on an average base of 300,000, the brand loses roughly 10,500 subscribers a month, or 378,000 over the period.
• So gross acquisitions required are roughly 320,000 + 378,000 = 698,000. At a blended acquisition cost of $70, that is $49M of spend.
• Now flex churn. At 2.5% monthly, losses fall to about 270,000 and gross adds to 590,000, saving roughly $7.6M at the same acquisition cost. Retention is a cheaper lever than acquisition here, and the math proves it rather than asserting it.
• Check unit economics before recommending spend. At $75 monthly revenue and a 35% contribution margin, a subscriber contributes $26 a month, so a $70 acquisition cost pays back in under three months. The spend is justified as long as churn does not rise with channel scale.

Recommendation. Fix retention first, then scale acquisition into the channels that hold payback under four months. Name the risk plainly: blended acquisition cost usually rises as you scale paid channels, so the $49M figure is a floor. The same reasoning pattern appears in the growth strategy cases guide.

Case 9: Market entry, airport foodservice concession

Prompt. A restaurant group is bidding for a five-year foodservice concession at a regional airport with 6 million annual passengers. What should they bid?

Worked approach.

• Convert passengers to customers. 6 million passengers, assume 55% are departing and pass the concession area, and 25% of those buy: 6M x 0.55 x 0.25 = 825,000 transactions.
• Revenue at a $14 average ticket: $11.6M a year.
• Costs: food at 30%, labour at 28%, and airport rent typically charged as a percentage of sales, say 15%. That leaves 27%, or $3.1M, before fixed overhead and the build-out.
• Build-out of $4M amortised over five years is $800K a year, leaving roughly $2.3M of annual operating profit, or $11.5M across the term.
• The bid ceiling is what remains after a required return. If the group needs a 20% return on the $4M invested, the maximum incremental guaranteed payment is well below the $2.3M annual profit line.

Recommendation. Bid with a revenue-share structure rather than a fixed guarantee, because passenger volume is the risk you cannot control and a fixed minimum transfers that risk to you for free. Flag the assumption most likely to be wrong: the 25% purchase rate, which swings the whole case. The underlying tree is the market entry framework, adapted to a bid rather than a build.

The L.E.K. written case

Candidate reports consistently describe a written exercise in the final round: a pack of slides with text, charts and tables, roughly 60 minutes to work through it, then a short deck and a presentation followed by questions. The volume is deliberately more than you can read.

What it actually tests. Triage and headline writing, not exhaustive analysis. You will not read every page. The scored behaviour is deciding in the first five minutes which three or four questions the pack is really asking, then reading only the pages that answer them.

How to prepare for it.

• Write slide headlines as conclusions, not labels. "Enterprise churn drove 70% of the revenue decline" beats "Revenue analysis".
• Build the deck backwards from your recommendation, then cut anything that does not support it.
• Rehearse the 60-minute clock with a real data pack. The written case interview guide and the consulting assessment centre written case playbook both give timed drills for this.
• Practise saying the answer first. In the presentation, the recommendation goes in the first sentence, then the support.

Start with the evidence behind a slide headline: read the exhibit, identify the comparison that matters and explain what it changes in the recommendation.

Self-diagnostic: are you practising like an L.E.K. candidate?

Work through these after the nine cases above. Each maps to something L.E.K. asks for explicitly in its own interview guidance.

1. Did you state your assumptions out loud before calculating? L.E.K.'s first published tip is to restate facts and state assumptions. An estimate delivered without its assumptions cannot be challenged, which means it cannot be scored.
2. Did your estimate carry through to the decision? In Case 3 and Case 6, the sizing sets the denominator for pricing. If your recommendation would have been the same at half the market size, you did not use the number.
3. Did you commit? L.E.K. asks candidates to "come up with an answer" while remaining willing to rethink. Hedging until the interviewer decides reads as the opposite of both halves.
4. Did you name the assumption most likely to be wrong? Volunteering the sensitivity is the single fastest way to look like someone who has done diligence work.
5. Did you speak the sector's language? In a life sciences or MedTech case, terms like probability of approval, prescriber overlap and site of care are table stakes rather than bonus points.

For fit and experiential rounds, which run alongside the cases rather than after them, use the L.E.K. behavioral interview questions guide, and rehearse those answers on the behavioral practice simulator rather than in your head. Before the round opens, run your CV through the consulting resume grader, since L.E.K. screens applications before any of this matters.

L.E.K.'s official practice resources

Work the firm's own material first, because it is the only material that is definitionally on-format.

• The interview preparation page. L.E.K.'s interview preparation page carries all three published cases plus the seven in-case tips. Treat the "Approaching the Case Interview" questions under each case as the actual exercise: they are the questions your interviewer is scoring you against.
• The application page. L.E.K.'s apply page states the process differs by region and level, and that candidates may apply to only one global office per year. Confirm your own sequence from your invitation rather than from any guide, including this one.
• The firm's insights library. L.E.K.'s published insights are organised by the exact sectors its cases come from. Reading two or three in the sector you are interviewing for is the cheapest sector-vocabulary win available.
• Its own numbers. L.E.K.'s about page reports 27 offices, 200 or more partners, 2,300 or more employees and 2,000 or more cases annually, which is useful material for a specific "why L.E.K." answer rather than a generic one.

Once the official material is exhausted, the constraint stops being reading and starts being reps under pressure. Pick candidate-led estimation, profitability and diligence prompts from the case library and run them out loud.

How to use these nine cases

Do not read them a second time. Run the loop below on each one, and convert every failure into a specific rep rather than another pass through the material.

How to run each case

• ✓

  Read only the prompt, then cover the solution. Reading a worked answer feels like learning and produces none of it

• ✓

  Write your structure in two minutes. Your structure is the first scored move, whoever leads the case

• ✓

  Say every assumption out loud before you calculate. L.E.K.'s own guidance asks for stated assumptions and restated facts

• ✓

  Finish with a recommendation that has a number and a risk in it. A framework without a decision is the most common way these cases end badly

• ✓

  Compare against the worked solution and name the one thing that broke. Diagnosis is what turns nine cases into a prep plan

• ✓

  Convert that break into a rep: market sizing, structure or case math. A scored rep on the exact skill beats another passive read

• ✓

  Then run a full live case out loud from the case library. Holding all five moves together under time pressure is the thing the interview actually tests

If you want more worked material across other firm formats, the case interview examples hub collects them, and the L.E.K. case interview guide covers process, rounds and what your invitation controls. For firm background and the recruiting route into it, see L.E.K. Consulting.

Prep plans for McKinsey, BCG, Bain, Deloitte, Accenture, EY-Parthenon, Kearney, Oliver Wyman, Roland Berger, L.E.K., OC&C, Strategy&, Simon-Kucher, Arthur D. Little, Capital One and your target firm. Choose your firm in the plan builder.

Turn this guide into a weekly plan

Choose your firm and interview date and get the weeks laid out, from the first lesson to the last case.

Build my prep plan

1

Learn the format

17 steps

• Learn how to caseGuided learning · 35 min
• The Case Interview MethodLesson · 8 min
• Profitability & Turnaround: AfrihealSkill drill · 5 min
• Profitability & Turnaround: Berty BankSkill drill · 5 min

If your application is not in yet, the L.E.K. resume guide and cover letter guide cover the documents, and the L.E.K. salary breakdown shows what the offer looks like.

Sources and further reading (checked July 31, 2026)

• L.E.K. official interview preparation page, including all three published case examples and the seven in-case tips: lek.com/careers/apply/interview-preparation
• L.E.K. official apply page, on regional variation and the one-office-per-year rule: lek.com/careers/apply
• L.E.K. about page, for office count, partner count, employee count and annual case volume: lek.com/about-lek
• L.E.K. published insights by sector: lek.com/insights
• My Consulting Offer L.E.K. case interview guide, for candidate-reported round structure and written case timings: myconsultingoffer.org/case-study-interview-prep/lek-case-interview
• MConsultingPrep L.E.K. case interview page, for candidate-reported round one and round two formats: mconsultingprep.com/lek-case-interview
• PrepLounge candidate forum thread on L.E.K. first-round cases: preplounge.com/consulting-forum/does-anyone-know-what-lek-first-round-casesinterviews-are-like-16971