BCG Case Interview Examples: 10 Practice Cases (2026)

This page gives you 10 BCG-style practice cases you can run, each opened up to its first exhibit with the real numbers worked, plus how a BCG case interview works and how to practice for one. Every example is a case from the Road to Offer library, so the case you read is the case you start. BCG interviewers usually let you propose the structure and ask for the data you need, and BCG France's case interview page gives 45 to 60 minutes per interview, including 15 to 20 minutes of fit questions.

Your first action: read Case 1 and say your structure out loud before you look at the worked exhibit. At Case 3 you can answer the structure question right in this page, with no account. Your first full case is free when you create an account. For the full hiring process, see the BCG Case Interview Guide.

BCG practice cases at a glance

How a BCG case works: the data comes in volume. You drive the case and work through several exhibits (2 to 4 in each case below), building one hypothesis across them. Many candidates also sit BCG's online Casey chatbot case before the live rounds.

Start free now: UrbanBrew Coffee Margin Case is a BCG-style profitability case, and a good first one: a coffee chain's EBITDA halved while revenue grew. It has 6 questions, 2 with math, 2 exhibits, and a median of 16 minutes across 117 completed voice runs. It is one of 21 BCG-style cases in the library, and an AI interviewer runs it out loud in voice mode or step by step in guided text. For Casey, the first 2 questions of the simulator are free with no account. Your first full case is free when you create an account; Case 3 below works its first exhibit. 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, the two Sustainable Futures Lab formats), a full Casey-style case, a full CCA-style diagnostic and a complete Bain SOVA-style diagnostic.

Road to Offer offers one complete free run in each of four Solve practice formats: Red Rock's full Study and six Cases on a 35-minute advisory clock, all three Sea Wolf sites on a 30-minute Simulation clock, all 13 SFL Classic decisions without a countdown, and SFL Island onboarding plus three project days on a 30-minute Simulation clock. Start without an account; create a free account to finish and keep your score, point breakdown and worked feedback on each submitted decision. Worked visuals include highlighted calculations and charts in Red Rock, constraint and treatment visuals in Sea Wolf, decision and dependency diagrams in Classic, and evidence chains, placement maps and team feedback in Island. No payment is required for those first runs or their worked reviews. Learning mode is paid: the matching game pack, Solve pack or qualifying Annual access includes its step-by-step coaching and feedback, untimed in the clocked games. The free Solve plan asks your invitation length and test date, then gives dated practice steps for up to 28 days; a free account is needed to view and save it.

These are practice cases in BCG's exhibit-heavy format (Electro Chargers is adapted from the Wharton consulting club casebook, the rest were written for Road to Offer), 5 of the 21 BCG cases in the library; none is BCG's own material. 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.

What should a BCG case practice example show?

A useful BCG example shows the prompt, the first structure, the evidence to request, the recommendation, and the next rep. Casey and the CCA are separate online assessments, and a practice case in BCG's style is not official BCG material.

Use the prompt, structure, evidence, synthesis method: 1. state the decision, 2. build the first tree, 3. request the highest-value evidence, 4. close with a recommendation and next rep. A practice case should end in a decision, not an explanation of the framework alone.

How BCG Differs from McKinsey and Bain

Understanding where BCG sits relative to the other MBB firms shapes how you practice. BCG is not simply "McKinsey but candidate-led." According to BCG's case interview preparation resources, case interviews assess "business judgment, logical reasoning, and quantitative aptitude," but the format in which those skills are tested differs meaningfully across firms:

What this means for practice: BCG gives you slightly more room than McKinsey to shape your approach, but probes harder on whether you can extract the "so what" from data. If you have been practicing only McKinsey-style cases (where the interviewer hands you exhibits with clear questions), BCG will feel harder because you have to drive the structure and the synthesis yourself. If you have been practicing only Bain-style candidate-led cases, BCG will feel familiar in format but the exhibit volume will be higher than you expect.

The fastest way to feel that difference is to drive a candidate-led case yourself, proposing the structure and calling for exhibits instead of waiting to be handed questions. Each of the 10 cases below starts that way.

Get your BCG prep plan

Build my prep plan

1

Learn the format

15 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

The Five Question Types Inside a BCG Case

A BCG case is not one continuous conversation. It is a sequence of question types, and each one is scored separately against the "business judgment, logical reasoning, and quantitative aptitude" criteria BCG names in its case interview preparation guidance. Five asks recur, and every one of them maps to a skill you can rehearse on its own before you ever sit a full case.

Practicing these as separate reps is the fastest way to find the one that keeps costing you points. Full cases then tell you whether you can hold all five together under time pressure.

How to Build a Running Hypothesis Across a BCG Case

The single skill that separates strong from exceptional BCG candidates is maintaining and updating a hypothesis as each new exhibit arrives. Here is the framework:

1. After exhibit 1: Form an initial hypothesis. "Revenue is declining: my hypothesis is this is price-driven, not volume-driven."
2. After exhibit 2: Refine or pivot. "The second exhibit shows volume is actually up but average transaction value is down. My hypothesis shifts to a mix problem: we are selling more of the lower-priced products."
3. After exhibit 3: Converge. "This exhibit confirms that the product mix shifted after a competitor entered the premium segment. My hypothesis is now that competitive pressure pushed sales toward our value tier."
4. In your synthesis: Connect all exhibits explicitly. "Exhibits 1, 2, and 3 together tell us that the revenue decline is a mix effect driven by competitive entry in premium, not an overall demand problem."

Practice this on every case below. After each exhibit, pause and state your hypothesis in one sentence before reading the worked update.

The part you cannot grade yourself is whether the structure you would have opened with is genuinely MECE and hypothesis-led rather than a template you recognized.

BCG's Preference for Market Sizing as Openers

One BCG-specific nuance that catches candidates off guard: BCG interviewers frequently open cases with a market sizing question before diving into the strategic problem. This is more common at BCG than at McKinsey or Bain.

For example, a BCG case about whether a retailer should enter a new product category might start with: "Before we get into the strategy, let's size this market. How large is the US market for premium pet food?"

The market sizing isn't disconnected from the case. It establishes the opportunity size that informs the subsequent strategic analysis. If your market sizing estimate suggests a $2B market with limited growth, the strategic case for entry is very different than if the market is $8B and growing at 12% annually.

This means your market sizing needs to be solid. BCG interviewers will test your assumptions and push back on your logic more aggressively than in a standalone sizing exercise. For a structured approach, see Market Sizing Step-by-Step.

Since BCG so often opens on a sizing question, this is the skill to get graded before anything else. Run one timed rep and see where your driver chain or assumptions get flagged.

The BCG Written Case Interview

Some BCG offices use a written case format in addition to, or in place of, one of the standard oral case interviews. This format is used primarily in second-round interviews, and BCG offices in the US, Russia, Sweden, the Netherlands, and South Africa have been known to include it (IGotAnOffer written case guide). You'll know well in advance if your interview includes a written case.

The format:

• You receive approximately 40 PowerPoint slides containing data, charts, tables, and sometimes press clippings
• You're given 3-4 key questions to answer
• You have 2 hours to review the material and prepare 3-5 presentation slides with your recommendations
• You then present and discuss your slides with the interviewer for about 40 minutes, during which they will challenge your analysis and probe your reasoning

Key constraints:

• You can use a calculator
• You cannot write on the slides BCG provides (they reuse them with other candidates)
• Your own slides need to tell a clear story: not just dump data, but synthesize it into a recommendation

How this differs from a standard case: The written case tests your ability to prioritize information under time pressure, self-direct your analysis without interviewer guidance, and communicate findings in slide format. You have 40 slides of data and 2 hours; you cannot read everything deeply. You need to triage: which slides contain data relevant to the 3-4 questions, and which can be skimmed or skipped?

How to prepare: Practice building slide-based recommendations from data packages. BCG's preparation pages do not offer a written case example (checked September 2026), so work the one in the written case interview guide under timed conditions. Then rehearse the same triage-and-synthesize loop on full cases with exhibits in the Road to Offer case library, which is the part candidates actually run out of practice material for. Candidate reports on recent written rounds also circulate in community forums such as PrepLounge's.

Quick Math

1. A product sells for $80 and carries $52 of COGS. What is the gross margin, in percent?

   Gross margin = (Revenue - COGS) / Revenue x 100. Answer the margin, not the dollars of gross profit.

2. A retailer books $80M of services at a 60% margin and $20M of hardware at a 25% margin. What is the blended gross margin, in percent?

   Blended margin = the revenue-weighted average of the segment margins. Averaging the two percentages is the trap.

3. Fixed costs are $5.4M, price is $150 and variable cost is $60 per unit. How many whole units break even?

   Contribution margin = Price - Variable cost. Break-even units = Fixed costs / Contribution margin; dividing by price understates it.

Preparing for the Casey Chatbot Assessment

If you are applying to BCG, you may face the Casey chatbot before or alongside traditional case rounds. According to BCG's interview process page, Casey is a text-based online case assessment that tests structured thinking, math, and data interpretation. The assessment lasts approximately 25-35 minutes and consists of more than 20 questions based on several business scenarios, including a combination of multiple-choice questions, free-text answers, and a video summary. It is a separate preparation track from live cases.

Casey question types to practice:

• Integrated reasoning: Combine information from multiple sources (text, tables, charts) to answer a question. This directly parallels the multi-exhibit synthesis skill used in live BCG cases.
• Multi-source reasoning: Evaluate which data sources are relevant to a specific question. Casey may present 4-5 sources and ask which ones you need, testing your ability to triage information, not just analyze it.
• Table and graphic interpretation: Read specific values from complex tables and charts under time pressure. Practice reading charts and exhibits like waterfall charts, stacked bar charts, and scatter plots quickly, as BCG cases use all three frequently.
• Quantitative reasoning: Multi-step math problems embedded in case context. These are similar to what you will see in live cases but delivered in text form.

Key difference from live cases: In Casey, you cannot ask the interviewer clarifying questions. You must extract all relevant information from the materials provided. This makes triage skill (quickly identifying which data matters) even more important than in live cases.

Practicing with any chatbot-based case tool builds relevant skills for this format. For detailed Casey coverage, see the BCG Case Interview Guide.

A different BCG screen is the one-way video interview on Spark Hire. It tests fit and communication, not cases, and it is not HireVue. If that is the invite you received, start with the BCG one-way video interview guide.

10 BCG Case Study Examples, With the First Exhibits Worked

Each example below is a real case from the Road to Offer case library, written in BCG's candidate-led, exhibit-heavy style. You get the client prompt, a structure a strong candidate would propose, the first exhibit with its real numbers, one or two calculations worked from that exhibit, and the hypothesis it supports. The rest of each case, including the recommendation, is yours to run: the card under each example starts that exact case. Cover the worked lines, say your own answer out loud, then check.

Cases 4 and 9 are adapted from the Wharton consulting club casebook; the other eight were written for Road to Offer. None of them is BCG's own material.

Case 1: FreshCart grocery margin recovery (profitability)

The prompt. FreshCart, a regional grocery chain with 320 stores, added a hot soup bar and a fresh juice station 18 months ago. Revenue grew from $600M to $680M, but gross profit stayed flat at about $135M. What is going on, and what should FreshCart do?

A structure that works. Revenue grew 13% and gross profit did not move, so either margins fell everywhere or one business is eating the growth. Split gross profit by business first (traditional grocery versus prepared foods), the first cut of any profitability framework, then, inside the problem business, look at price, material cost, labor and volume per unit.

Exhibit 1, revenue and gross profit by business ($M).

The math. Traditional grocery added $157M - $146M = $11M of gross profit. Prepared foods lost $11M more, going from -$11M to -$22M. The two cancel: +$11M - $11M = $0, which is why total gross profit sits at $135M. Prepared foods revenue grew 50% ($80M to $120M) while its margin worsened from -$11M / $80M = -13.8% to -$22M / $120M = -18.3%.

The hypothesis update. "The grocery core is healthy at about 28% margin. Prepared foods lose more money the more they sell, but the totals do not say why: it could be pricing, cost per unit, or too little volume to cover fixed staffing. I want the per-unit economics and daily volume of each concept next."

The full case continues with your hypotheses for the losses, the unit economics of the soup bar against the juice station (Exhibit 2), the strategic options and their risks, and your recommendation to the CEO.

Case 2: HarborHealth acquisition of MedLink Virtual Care (M&A)

The prompt. HarborHealth runs 95 outpatient clinics in the US Midwest and is weighing the acquisition of MedLink Virtual Care, a fast-growing telehealth provider. MedLink's shareholders ask for an enterprise value of $420M. Should the deal proceed, and under what conditions does it create value?

A structure that works. Four branches, the core of an M&A case framework: strategic fit (hybrid care, MedLink's strong digital front end against its weak payer contracting), standalone value, synergies net of integration costs, and integration risk. Start with price, because it tells you how much of the case rests on synergies.

Exhibits 1 and 2, MedLink financials and peer multiples. MedLink's revenue went from $120M (2022) to $150M (2023) to $180M (2024), and EBITDA from $12M to $18M to $25M. Public digital-health peers trade at a median of 14.0x EV/EBITDA on 14% growth; the fastest-growing peer grows 19% and trades at 16.0x.

The math.

• Implied multiple: $420M / $25M of 2024 EBITDA = 16.8x.
• Standalone value at the peer median: $25M x 14.0 = $350M. The ask is $420M - $350M = $70M above it, a 20% premium.
• MedLink's 2024 growth: $180M / $150M - 1 = 20%.

The hypothesis update. "MedLink grows one point faster than the fastest peer, yet its ask of 16.8x sits above that peer's 16.0x. Standalone value does not cover the price, so the deal only works if synergies, net of integration costs, are worth more than $70M."

The full case continues with a value bridge to the maximum justified offer (Exhibit 3), an integration risk map (Exhibit 4) and your go or no-go for the CEO.

Case 3: UrbanBrew coffee margin case (profitability)

The prompt. UrbanBrew, a regional coffee chain with 140 US stores, saw EBITDA fall from $24M in 2023 to $12M in 2024 while revenue grew. The CEO wants the main cause and the actions that can improve profitability quickly, within the next two quarters.

A structure that works. Revenue grew and EBITDA halved, so the hypothesis is margin compression. Test category gross margins first (is growth coming from a worse-margin product?), then costs below gross profit (labor, promotions), then pricing.

Exhibit 1, gross profit by category ($M).

The math. Sandwich revenue rose $11M while its COGS rose $15M, so sandwich gross margin fell from 60.0% to 41.1% ($27M / $45M, then $23M / $56M). Across all categories, gross profit fell $5M (from $138M to $133M) while EBITDA fell $12M.

The hypothesis update. "Sandwiches grew into a much worse margin, and bakery also lost $4M. But gross profit explains well under half of the $12M drop, so most of the decline must come from costs below gross profit. I want the EBITDA bridge next."

Exhibit 2, the EBITDA bridge ($M). Volume +4 and pricing +2 against food cost -5, labor -8 and promotions -5: a net -12, which is exactly the fall from $24M to $12M. Labor is the largest single headwind. The case's recovery scenario cuts the labor headwind by 25% ($8M x 25% = $2M) and promotions by 30% ($5M x 30% = $1.5M) and assumes a $4M sandwich cost fix: $2M + $1.5M + $4M = $7.5M, taking EBITDA from $12M to $19.5M.

What a strong answer says. "EBITDA halved because costs grew faster than revenue: labor alone took $8M, and sandwiches grew into a worse margin, from 60.0% to 41.1%. I would start with labor scheduling and promotion efficiency, which recover $3.5M under the case's scenario, and, if the case's assumed $4M sandwich cost fix holds, reach about $19.5M, while checking why bakery also lost $4M, testing each change in a few stores first so service and traffic hold up."

UrbanBrew is the one case on this page worked to the end, so you can check your own answer against it. The live case adds the clarifying questions and scores your structure, math and recommendation step by step.

Practice a BCG profitability case yourself. EBITDA halved from $24M to $12M in 2024 while revenue grew. Answer this BCG profitability case step in the article. Browse more practice cases in the case interview library

Case 4: Electro Chargers city entry (market entry)

Adapted from the Wharton consulting club casebook.

The prompt. Electro Chargers makes EV battery swap stations. It already operates in Boston and wants to enter one new US city from a shortlist of New York, Washington DC, Seattle, Miami and Atlanta. Which city should it enter, and what fixed investment does that require through 2030?

A structure that works. Two questions, as in any market entry framework. Where: the size of the swappable EV base today and its growth to 2030, competition and policy support. How much: stations needed at the target share, then setup cost, rent, the operations team and overheads.

Exhibit A, shortlisted cities.

The math. Swappable EVs today = EVs x swappable share: New York 60,000 x 8% = 4,800; Seattle 30,000 x 12% = 3,600; Miami 30,000 x 5% = 1,500; Washington DC 20,000 x 5% = 1,000; Atlanta 15,000 x 6% = 900.

The hypothesis update. "New York and Seattle lead on today's swappable base: New York is a third larger, Seattle has one fewer competitor and the only high policy support. That ranks demand, not entry economics, so I keep all five cities on the list, project each to 2030 and check competition before I cut any."

The full case continues with the 2030 projection, the station count and fixed investment from Exhibits B and C, the other risks of entry and your recommendation.

Case 5: PeakFit omnichannel growth plan (growth strategy)

The prompt. PeakFit, a premium fitness brand with 85 studios in major US metros, wants to accelerate growth over the next three years while protecting profitability. The CEO asks whether to prioritize new studio openings, hybrid memberships or digital conversion to reach a 2027 EBITDA target of $70M.

A structure that works. Break the target into levers: same-store productivity, membership mix (in-studio, hybrid, digital-only), new studios (which cities, how fast they ramp) and digital retention. Then rank the levers by size, speed and execution risk.

Exhibit 1, membership mix (thousands of members).

The math. Total members grew 12k (173k to 185k). Hybrid added 20k (18k to 38k) and more than doubled; digital-only lost 13k, down 22%; in-studio added only 5k. Hybrid's share of members went from 18 / 173 = 10.4% to 38 / 185 = 20.5%.

The hypothesis update. "Hybrid is where the net growth shows up, but the totals cannot tell me whether digital-only members upgraded or churned and were replaced. My working hypothesis is that hybrid conversion is a cheap lever; I want cohort retention and contribution by membership type to test it, and new studios still have to pass a city screen."

The full case continues with year-one studio economics (Exhibit 3), the city screen (Exhibit 2), the EBITDA bridge to $70M (Exhibit 4) and your growth plan.

Case 6: TravelLine call center optimization (operations)

The prompt. TravelLine, an online travel booking platform, moved customer service from 12 regional offices into one national call center over 50 weeks. The center has 45 trained agents, customers wait 20 minutes or more and NPS has dropped 25 points. The CEO wants the root cause and a fix.

A structure that works. Demand against capacity. Demand is customers on the new center x calls per customer, adjusted for seasonality. Capacity is agents x productive hours x calls per hour. Once the gap is sized, decide whether to close it with hiring, deflection, productivity or a slower migration.

Exhibits 1 and 2, migration and capacity drivers. 300,000 customers had moved to the center by week 20 and 780,000 by week 50. Each call takes 210 seconds plus 90 seconds of wrap-up, 5 minutes in total. Customers call 2.5 times a year. Agents work 40 hours a week, 78% of that is net working time and 55% of net time is spent on calls.

The math.

• Capacity per agent: 40 hours x 78% x 55% = 17.16 hours on calls a week. At 5 minutes a call that is 12 calls an hour, so 17.16 x 12 = about 206 calls a week.
• Week 20 demand: 300,000 customers x 2.5 calls / 52 weeks = about 14,423 calls a week.
• Agents needed at week 20: 14,423 / 206 = about 70, against 45 on staff.

The hypothesis update. "At week 20, with less than half the customers migrated, demand already needed about 70 agents. The center has 45 today, and I do not know the week-20 roster, so my hypothesis is that migration outran capacity planning rather than agent performance. I want the staffing history and the full-migration demand next."

The full case continues with week-50 demand, the hiring and budget constraints, the effect of self-service and better agent tools, and your recommendation to the CEO.

Driver-tree arithmetic like this is where candidates lose time in the room. Rep one timed calculation before the next case.

Quick Math

1. A retailer runs 600 promotional items a week at 82% in-stock compliance. Each in-stock promotional item generates $220 of weekly incremental revenue. What is the weekly revenue lost to out-of-stock promotional items, in dollars?

   Lost revenue = Weekly items x (1 - In-stock rate) x Revenue per in-stock item.

2. A manufacturer spends $140M a year on a key component. Consolidating suppliers wins 9% savings. What is the annual cost saving, in dollars?

   Saving = Annual spend x Savings rate.

3. Closing a warehouse removes $15M of annual fixed cost and $4M of annual variable cost while adding $7M a year of logistics cost. Closure and severance cost $9M once. What is the three-year return on investment, in percent? Round to the nearest whole percent.

   ROI = (Three-year net saving - One-off cost) / One-off cost, where net saving nets fixed and variable savings against the added cost.

Case 7: Nashville Legends live venue profitability (profitability)

The prompt. Meridian Entertainment runs the SoundStage Theater, a 2,000-seat venue in Nashville that has hosted the same resident country music show, The Nashville Legends, for 8 years. The annual contract is about to expire, and a show built around a rising TikTok music act is on offer as a replacement. Should Meridian renew the current show for another year?

A structure that works. Three questions: is the current show still worth keeping on its own (revenue, costs, profit), how does the replacement compare once its one-time switching cost is counted, and could the current show be refreshed instead of replaced.

Exhibit 1, revenue per show by seating category (current show). The show plays 6 nights a week for 40 weeks, so 240 shows a year.

The math. Annual ticket revenue = $137,250 x 240 shows = $32.94M, about $32.9M. Unsold seats cost real money in two tiers: 1,000 - 750 = 250 Premium seats at $85 is $21,250 a show, and 600 - 270 = 330 Standard seats at $50 is $16,500 a show. Premium is the bigger gap.

The hypothesis update. "Stale is not the same as unprofitable: VIP still sells out. The empty seats sit in Premium first and Standard second, so the comparison has to include filling those tiers, not only swapping the act."

The full case continues with the cost base and annual profit (Exhibit 2), a comparison with the replacement show including its switching cost (Exhibit 3), growth ideas and your renewal call.

Case 8: NorthStar Logistics acquisition of RouteWave TMS (M&A)

The prompt. NorthStar Logistics, a US third-party logistics company, is weighing the acquisition of RouteWave, a transportation management software provider. RouteWave's shareholders ask for an enterprise value of $360M. Should NorthStar proceed, and under what conditions?

A structure that works. The same skeleton as Case 2, run faster: strategic fit (route-optimization IP, sticky SaaS contracts), standalone value, synergies net of integration costs, and integration risk across commercial teams and data systems.

Exhibits 1 and 2, RouteWave financials and peer multiples. RouteWave's revenue went from $140M (2022) to $172M (2023) to $210M (2024), and EBITDA from $14M to $21M to $28M. Logistics software peers trade at a median of 11.8x EV/EBITDA on 12% growth; the fastest-growing peer grows 17% and trades at 13.3x.

The math.

• 2024 growth: $210M / $172M - 1 = 22.1%.
• Implied multiple: $360M / $28M = 12.9x.
• Standalone value at the peer median: $28M x 11.8 = about $330M, so the ask is about $30M above it.

The hypothesis update. "This is the opposite of MedLink in Case 2. RouteWave grows faster than every peer, yet asks less than the fastest peer's 13.3x. Growth may justify a premium, but at the peer median RouteWave is worth about $330M standalone against a $360M ask, so the price only holds if the synergies in the value bridge cover the gap. I want that bridge next."

The full case continues with the value bridge and headroom against the ask (Exhibit 3), integration guardrails for the first 100 days (Exhibit 4) and your recommendation.

Case 9: Coolant Co. US market entry (market entry)

Adapted from the Wharton consulting club casebook.

The prompt. Coolant Co. is a Chinese manufacturer of cooling equipment for data centers. The CEO wants to enter the US market because AI is driving data-center growth. Should Coolant enter, and how?

A structure that works. Market attractiveness (size, growth, fragmentation), ability to win (product compatibility, regulation and tariffs on Chinese imports), and entry mode economics (direct entry against a partnership, and how fast each pays back).

Exhibit A, US market attractiveness. Cooling equipment is about 10% of data-center industry revenue. The largest cooling player holds about 10% of the market. News flags that Chinese equipment fits US data-center designs, and that tariffs on Chinese imports could rise within two years. Revenues of the four data-center players shown ($M):

The math. Data-center revenue grew from $500M to $760M to $1,200M; 2024 growth is $1,200M / $760M - 1 = 58%. The US cooling equipment market in 2024 is 10% x $1,200M = $120M.

The hypothesis update. "A fast-growing $120M market where no player holds more than about 10% is open to a new entrant. The risk is tariffs, which favors an entry mode that limits upfront exposure. I want the economics of each mode next."

The full case continues with entry modes, the payback of direct entry against a partnership (Exhibit B) and a short close-out.

Case 10: NimbusCloud capacity expansion (growth investment, hard)

The prompt. NimbusCloud is the largest US cloud infrastructure provider, with 230 petabytes (PB) of capacity at the lowest annual cost of $27M per PB. It can build a data center that raises capacity to 360 PB for $3B. Total market demand is 980 PB. The board requires payback in under 5 years. Should NimbusCloud expand?

A structure that works. First the economics of the new capacity: the market price after expansion, profit before and after, and payback against the 5-year rule. Then the competitive response, because a competitor's move can change the answer. Close on strategic implications.

Exhibit 1, the industry cost curve. Providers are ranked from lowest to highest annual cost per PB, and the market price is set by the most expensive provider needed to meet demand. At 980 PB of demand, that cost is $60M per PB.

The math.

• Today's price: $60M per PB, the cost at the 980 PB point.
• NimbusCloud's margin: $60M - $27M = $33M per PB.
• Today's profit: 230 PB x $33M = $7,590M, about $7.6B a year.

The hypothesis update. "Adding 130 PB of low-cost capacity pushes the most expensive providers off the curve, so the price falls. The expansion only pays if the extra volume outweighs a lower price on the 230 PB already sold."

The full case continues with the new market price, profit after expansion, the payback test, a game against Competitor B (Exhibit 3) and your board recommendation. It is the hardest of the ten.

If you only have time for one practice loop, run Case 3 (profitability), Case 2 (M&A) and Case 9 (a market size that feeds an entry decision) back to back, out loud, from their cards above. For more BCG-style cases, browse the case library.

Self-Diagnostic: Are You Practicing Like a BCG Candidate?

Based on BCG's own interview preparation guidance and consistent feedback from candidates in IGotAnOffer's BCG prep guide and Glassdoor interview reviews, after working through the 10 cases above, ask yourself these questions:

1. After each exhibit, could you state your hypothesis in one sentence? If you were still summarizing individual exhibits without connecting them, you are not synthesizing, which is the skill BCG weights most heavily.
2. Did you connect the market size (Case 9) to the entry decision? BCG uses sizing to frame the opportunity. If your sizing was disconnected from your strategic analysis, practice linking them.
3. Did you quantify your final recommendation? A recommendation with a price, a payback or a headcount attached is BCG-caliber. "Look into it further" is not.
4. Did you lead with the conclusion in every answer? BCG interviewers evaluate top-down communication: "The key insight is X, supported by Y and Z," not "Let me walk through the data and see what we find."

For BCG's full scoring criteria, the Casey chatbot assessment, and a complete prep plan, see the BCG Case Interview Guide. For exhibit reading technique, see Reading Charts and Exhibits in Case Interviews. For quantitative sharpness, see Case Interview Math Practice.

Test your exhibit diagnosis on a fresh chart: choose the evidence that supports your hypothesis before stating the conclusion aloud.

BCG's Official Practice Resources

BCG's own practice material changed during 2026. BCG's global case interview preparation page (checked September 2026) explains what to expect in a case study interview and points to OneDay@BCG and a tips article. Here is what is available and what is not:

• Practice quizzes: a climate target business case, a digital bank strategy refresh and a cloud migration case for a consumer goods company. As of September 2026 they are still listed on BCG's Canadian case preparation page, not the global one, but the quiz.bcg.com links they point to did not load when checked. An archived copy of the global page shows they were there in January 2026.
• OneDay@BCG: BCG's free, self-paced virtual job simulation, linked from the same preparation pages. Its Strategy Consulting module is the closest thing to an official BCG practice case you can work through, but nobody times you or pushes back the way an interviewer does.
• No official casebook or written case PDF: the preparation pages offer no downloadable case set, and the written case PDF address that circulates (media-publications.bcg.com) returned a 404 when checked in September 2026. For a worked written case, use the written case interview guide; for free business school case books, see the case book vault.
• Casey: BCG's online assessment, sent by invitation (not a public practice tool), is a text-based case simulation that tests structured thinking, math, and data interpretation in approximately 25-35 minutes, according to IGotAnOffer's Casey assessment guide. Practicing with any chatbot-based case tool (including Road to Offer AI case practice) builds relevant skills for this format.

BCG case studies on bcg.com are a different thing. The firm's Our Client Impact page (checked September 2026) collects stories of real client engagements. Read two or three before writing your Why BCG answer, because they show the work the firm wants to be known for. They are not interview practice: each one reports a result rather than handing you a problem to solve.

Once you have worked BCG's own materials, the constraint stops being material and starts being reps under pressure. Browse the case library and pick candidate-led profitability, market entry, and pricing prompts to run with the same loop.

Turn this guide into a BCG plan

Build my prep plan

1

Learn the format

15 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

How to Use These BCG Practice Cases

Use a BCG-specific practice loop rather than generic case reps. Split your week across three formats: live case interviews, self-paced cases such as OneDay@BCG, and Casey-style data drills where you cannot ask clarifying questions.

Reading a case on your own does not replace a BCG mock case, where someone else holds the exhibits and hands them over only when you ask the right question. Book a few mock interviews with a partner who plays the interviewer that way, and have them score your running hypothesis after each exhibit, not just your final answer. The mock case interview guide covers how to run and score one.

For each of the 10 cases above:

1. Read only the prompt. Cover the structure and the worked exhibit.
2. Write your own structure for 2 minutes: what exhibits would you expect? What is your initial hypothesis?
3. Read the exhibit, do the math yourself, and write one sentence updating your hypothesis.
4. Compare with the worked lines. Where did your structure or math differ?
5. Run the full case from its card and finish with a recommendation that has a number attached.

Then wire each weakness to a specific rep rather than doing generic cases:

• Structure felt slow or template-shaped: run free structure drills until the tree comes out hypothesis-led.
• The sizing opener wobbled: work the market sizing method, then rep it on the free sizing drills.
• Arithmetic slipped: use case interview math practice and the math drill tool.
• Exhibits took too long to read: work reading charts and exhibits and rep free graph drills.
• Synthesis rambled: tighten it with the synthesis guide.
• The Casey track: use BCG's official practice cases, then the Casey simulator with the same discipline of reading the objective, choosing only the relevant data, calculating cleanly, and giving a short video-style synthesis.
• Your application materials: run the resume through the consulting resume grader before the round starts.

For more worked examples across all firm formats, see Case Interview Examples.

Sources and Further Reading (checked June to September 2026)

• BCG case interview preparation and interactive practice cases: careers.bcg.com/global/en/case-interview-preparation
• BCG practice quizzes and OneDay@BCG: careers.bcg.com/ca/fr/case-interview-preparation
• BCG France, the case interview (length and fit segment): careers.bcg.com/global/en/locations/france/the-case-interview
• IGotAnOffer BCG case interview guide: igotanoffer.com/blogs/mckinsey-case-interview-blog/bcg-case-interview
• IGotAnOffer BCG Casey chatbot guide: igotanoffer.com/en/advice/bcg-online-case-assessment
• IGotAnOffer written case interview guide: igotanoffer.com/blogs/mckinsey-case-interview-blog/written-case-interview
• Glassdoor BCG interview reviews: glassdoor.com/Interview/Boston-Consulting-Group-Interview-Questions-E3879.htm
• Management Consulted BCG case interview: managementconsulted.com/bcg-case-interview
• PrepLounge BCG written case preparation: preplounge.com/consulting-forum/how-to-prepare-for-the-written-case-interview-at-bcg-final-interview-rounds-973