Practice cases: US and Canada
Eight full cases set in North America: a pet-food price rise, software growth quality, a private-equity buy-and-build, urgent-care clinics, a ski resort's short season in Quebec, sorting machines at a Canadian parcel hub, a public surgery centre, and a dental clinic deal in Alberta.
Key takeaways
- Where the structure comes from: it is built from the goal of this exact question (Price rise versus the volume it can afford to lose), not taken from a list.
- Where the structure comes from: it is built from the goal of this exact question (Growth = existing customers (net revenue retention) + new customers), not taken from a list.
- Where the structure comes from: it is built from the goal of this exact question (Entry cost, exit value, and the fund's return), not taken from a list.
Eight full cases set in North America: a pet-food price rise, software growth quality, a private-equity buy-and-build, urgent-care clinics, a ski resort's short season in Quebec, sorting machines at a Canadian parcel hub, a public surgery centre, and a dental clinic deal in Alberta.
Cover the solution and run each case out loud, ideally with a partner playing the interviewer. Ask your own clarifying questions, state a hypothesis, build a structure from the maths of the goal (not from a memorised list), and do the math on paper before you look. Then compare your synthesis with the one given, and read the strong and weak candidate notes. Each case is labeled Starter, Standard, or Stretch.
Case 1: Prairie Pet Foods: should it raise prices 10 percent?
Where the structure comes from: it is built from the goal of this exact question (Price rise versus the volume it can afford to lose), not taken from a list. Each branch is one driver of that goal, and the hypothesis above says which branch to test first.
Worked case
Starter: Prairie Pet Foods: should it raise prices 10 percent?
The prompt
Prairie Pet Foods sells premium dog food in Canada. It is considering a 10 percent price rise. The exhibit shows results from test markets. Should it go ahead?
Difficulty: Starter. Format: interviewer-led, with an exhibit. Industry: Consumer goods. Region: Canada. Interview length: about 25 minutes. The company is fictional and all figures are illustrative.
Clarifying questions, with the interviewer's answers
- What are the price, cost, and volume?Answer: CAD 50 per bag, variable cost CAD 30, about 2 million bags a year.
- What is proposed?Answer: A 10 percent price rise, to CAD 55.
- Do we have evidence on customer response?Answer: Yes, two test markets and a control market.
A hypothesis to say out loud: Premium pet food buyers are loyal, and the product has a healthy contribution. My hypothesis is that the price rise pays unless volume falls by a lot.
The structure
- Price rise versus the volume it can afford to lose
- Contribution per bag before and after
- Key: Break-even volume loss
- Evidence from test markets
- Retailer response
The exhibit
| Market | Price change (%) | Volume change (%) |
|---|---|---|
| Test market A | 10 | -5 |
| Test market B | 10 | -7 |
| Control market | 0 | 0 |
Working it through
1. Contribution now
Price minus variable cost.
Contribution now (CAD per bag):50 - 30 = 202. Contribution after the rise
At CAD 55.
Contribution after (CAD per bag):55 - 30 = 253. Break-even volume loss
Price rise divided by (contribution plus the rise).
Break-even volume loss (%):5 ÷ (20 + 5) × 100 = 204. Average test result
The two test markets lost 5 and 7 percent of volume.
Average volume loss (%):(5 + 7) ÷ 2 = 65. Yearly contribution now
2 million bags at CAD 20.
Contribution now (CAD a year):2,000,000 × 20 = 40,000,0006. Yearly contribution after
6 percent fewer bags at CAD 25.
Contribution after (CAD a year):2,000,000 × 0.94 × 25 = 47,000,0007. Curveball: the largest retailer wants a share
Interviewer: "Our largest retailer says it will ask for an extra 3 percent discount if we raise prices." Contribution after the rise and the discount (conservatively applied to all volume):
Contribution with the discount (CAD a year):2,000,000 × 0.94 × (55 × 0.97 - 30) = 43,898,000
What the exhibit shows
Volume fell 5 to 7 percent after a 10 percent price rise, far less than the 20 percent the company could afford to lose.
The recommendation
Raise the price. First, the company could lose up to 20 percent of volume before profit falls, and tests show only about 6 percent. Second, contribution rises from about CAD 40 million to about CAD 47 million a year. Third, even if the largest retailer takes an extra 3 percent discount, contribution is still about CAD 43.9 million, above today. Roll out region by region, watch volume monthly, and offer the retailer joint promotions instead of a permanent discount.
Risks: Long-term volume loss may exceed a three-month test; Rivals may hold prices to win share.
Next steps: Roll out in two regions first; Prepare the retailer negotiation with the test data.
A strong candidate
Found the break-even volume loss, compared it with the tests, and tested the retailer curveball.
A weak candidate
Worried that customers would leave and rejected the rise without checking how many could leave.
Case 2: Bluegrass HR: how safe is its growth?
Where the structure comes from: it is built from the goal of this exact question (Growth = existing customers (net revenue retention) + new customers), not taken from a list. Each branch is one driver of that goal, and the hypothesis above says which branch to test first.
Worked case
Standard: Bluegrass HR: how safe is its growth?
The prompt
Bluegrass HR sells payroll and HR software to mid-sized US companies, priced per employee (per seat). Its board asks how healthy its growth is. What would you tell them?
Difficulty: Standard. Format: candidate-led, with interviewer dialogue. Industry: Technology (software). Region: US. Interview length: about 30 minutes. The company is fictional and all figures are illustrative.
Clarifying questions, with the interviewer's answers
- How big is the business?Answer: USD 50 million of annual recurring revenue (ARR) at the start of the year.
- What happened to existing customers over the year?Answer: 8 percent of ARR churned, 2 percent downgraded, and existing customers added 15 percent through more seats.
- How much came from new customers?Answer: About USD 12 million of new ARR.
A hypothesis to say out loud: Software growth depends a lot on existing customers growing. My hypothesis is that growth relies on customers adding seats, which makes it exposed to a hiring slowdown.
The structure
- Growth = existing customers (net revenue retention) + new customers
- Churn and downgrades
- Key: Expansion from more seats
- New customers
Working it through
1. ARR lost to churn
Candidate: "8 percent of USD 50 million."
Churned ARR (USD):50,000,000 × 0.08 = 4,000,0002. Net revenue retention
Candidate: "Start, minus churn and downgrades, plus expansion, divided by start."
Net revenue retention (%):(50,000,000 - 50,000,000 × 0.08 - 50,000,000 × 0.02 + 50,000,000 × 0.15) ÷ 50,000,000 × 100 = 1053. ARR at year end
Existing customers at 105 percent, plus USD 12 million of new ARR.
Ending ARR (USD):50,000,000 × 1.05 + 12,000,000 = 64,500,0004. Growth
Change over the starting ARR.
Growth (%):(64,500,000 - 50,000,000) ÷ 50,000,000 × 100 = 295. Curveball: hiring slows
Interviewer: "Our customers are slowing hiring. Expansion could halve." Candidate: "Net revenue retention would be:"
Net revenue retention (%):(1 - 0.08 - 0.02 + 0.075) × 100 = 97.56. Growth in that case
Same new ARR.
Growth (%):(50,000,000 × 0.975 + 12,000,000 - 50,000,000) ÷ 50,000,000 × 100 = 21.57. A lever not tied to seats
Interviewer: "We could sell an analytics module." Candidate: "If 20 percent of customers buy it at 10 percent of their current spend:"
Extra ARR (USD):50,000,000 × 0.2 × 0.1 = 1,000,000
The recommendation
Growth is healthy today but depends too much on customers hiring. First, the business grew 29 percent, with existing customers growing 5 percent net (105 percent retention). Second, most of that expansion comes from added seats, so a hiring slowdown that halves expansion drops retention below 100 percent and growth to about 21.5 percent. Third, 8 percent churn is the largest single loss and worth reducing. Reduce churn with better onboarding, and add revenue not tied to seats, such as an analytics module worth about USD 1 million of ARR at 20 percent take-up.
Risks: A deeper hiring slowdown; Module take-up may be lower than 20 percent.
Next steps: Find which customer groups churn most, and why; Test the analytics module with 50 customers.
A strong candidate
Split growth into retention and new sales, calculated net revenue retention, and tested its dependence on hiring.
A weak candidate
Reported 29 percent growth as strong and stopped, without asking where it came from.
Case 3: Corvane Home Comfort: a buy-and-build in home services
Where the structure comes from: it is built from the goal of this exact question (Entry cost, exit value, and the fund's return), not taken from a list. Each branch is one driver of that goal, and the hypothesis above says which branch to test first.
Worked case
Stretch: Corvane Home Comfort: a buy-and-build in home services
The prompt
A US private-equity fund plans to build Corvane Home Comfort, a heating and air-conditioning services group in the Midwest, by buying a platform company and five add-ons. The exhibit shows the deal pieces. What return could the fund earn, and what could go wrong?
Difficulty: Stretch. Format: interviewer-led, with an exhibit. Industry: Private equity and home services. Region: US. Interview length: about 40 minutes. The company is fictional and all figures are illustrative.
Clarifying questions, with the interviewer's answers
- What is the strategy?Answer: Buy one larger heating and air-conditioning services company (the platform), then buy five small local companies (add-ons) and combine them.
- How is it financed?Answer: 60 percent debt; debt falls to USD 134 million by exit.
- What growth and exit?Answer: About 5 percent a year EBITDA growth for five years; the fund expects to sell at 10 times EBITDA.
A hypothesis to say out loud: Small companies sell for lower multiples than large ones, so combining them can create value. My hypothesis is that the return depends heavily on the exit multiple.
The structure
- Entry cost, exit value, and the fund's return
- Total price and blended entry multiple
- EBITDA growth to exit
- Key: Exit value, MOIC, and IRR
- Risk: a lower exit multiple
The exhibit
| Item | Platform | Each add-on |
|---|---|---|
| EBITDA (USD million a year) | 20 | 3 |
| Purchase multiple (EV/EBITDA) | 10 | 6 |
| Number of companies | 1 | 5 |
Working it through
1. Total price
Platform at 10 times USD 20 million, plus five add-ons at 6 times USD 3 million each (USD million).
Total price (USD million):20 × 10 + 5 × 3 × 6 = 2902. Combined EBITDA
Platform plus add-ons.
Combined EBITDA (USD million):20 + 5 × 3 = 353. Blended entry multiple
Below the 10 times paid for the platform alone.
Blended EV/EBITDA:290 ÷ 35 = 8.294. Equity invested
40 percent of the total price.
Equity (USD million):290 × 0.4 = 1165. Exit value
EBITDA grows 5 percent a year for five years and sells at 10 times.
Exit EV (USD million):35 × 1.05 × 1.05 × 1.05 × 1.05 × 1.05 × 10 = 4476. MOIC
Exit value minus USD 134 million of remaining debt, divided by equity.
MOIC:(35 × 1.05 × 1.05 × 1.05 × 1.05 × 1.05 × 10 - 134) ÷ 116 = 2.77. IRR check
About 2.7 times in five years is roughly 22 percent a year.
1.22 to the power of 5:1.22 × 1.22 × 1.22 × 1.22 × 1.22 = 2.78. Curveball: exit multiples fall
Interviewer: "If buyers pay only 8 times at exit, what happens?"
MOIC at an 8 times exit:(35 × 1.05 × 1.05 × 1.05 × 1.05 × 1.05 × 8 - 134) ÷ 116 = 1.93
What the exhibit shows
Add-ons cost 6 times EBITDA against 10 for the platform, so combining them lowers the average price paid.
The recommendation
The plan could return about 2.7 times the fund's money, roughly 22 percent a year, but much of that depends on the exit multiple. First, buying add-ons at 6 times lowers the blended entry multiple to about 8.3 times. Second, selling the combined group at 10 times turns that gap into value, alongside 5 percent yearly EBITDA growth and debt paydown. Third, if the exit multiple is only 8 times, MOIC falls to about 1.9 times, below a typical target. So only proceed if the group becomes a real, integrated business that deserves a higher multiple: shared booking, purchasing, and technician training, not just a collection of local firms.
Risks: Add-on owners and key technicians may leave after the sale; Integration may fail, so the group does not earn a higher multiple; Higher interest rates increase debt costs.
Next steps: Build an integration plan for booking, purchasing, and training; Agree retention terms with add-on owners.
A strong candidate
Calculated the blended entry multiple, exit value, MOIC, and IRR, and identified the exit multiple as the main risk.
A weak candidate
Added up EBITDA and said the deal is good because the market is fragmented, without calculating the return.
Case 4: A health system considers urgent-care clinics
Where the structure comes from: it is built from the goal of this exact question (Size the visits, then test one clinic's economics), not taken from a list. Each branch is one driver of that goal, and the hypothesis above says which branch to test first.
Worked case
Standard: A health system considers urgent-care clinics
The prompt
First, estimate how many urgent-care visits a year happen in a US metro area of 3 million people. Then: a local health system wants to open urgent-care clinics. How many, and will they make money?
Difficulty: Standard. Format: market-sizing opener, then a business question. Industry: Healthcare. Region: US. Interview length: about 30 minutes. The company is fictional and all figures are illustrative.
Clarifying questions, with the interviewer's answers
- What area?Answer: A US metro area of about 3 million people.
- How often do people use urgent care?Answer: About 0.4 visits per person a year.
- What share could our system win?Answer: About 10 percent; a clinic handles about 15,000 visits a year.
- For the business question, what does a clinic cost and earn?Answer: About USD 1.2 million of fixed costs a year per clinic. A visit earns about USD 150 from commercial insurers and costs about USD 60 in variable costs.
A hypothesis to say out loud: Urgent care is a volume business with high fixed costs per clinic. My hypothesis is that the market supports several clinics, but payer mix decides whether each clinic makes money.
The structure
- Size the visits, then test one clinic's economics
- People x visits per person x our share
- Clinics = our visits / visits per clinic
- Key: Break-even visits per clinic
- Payer mix
Working it through
1. Visits in the metro area
3 million people at 0.4 visits a year.
Visits a year:3,000,000 × 0.4 = 1,200,0002. Our visits
A 10 percent share.
Our visits a year:3,000,000 × 0.4 × 0.1 = 120,0003. Number of clinics
At 15,000 visits each.
Clinics:120,000 ÷ 15,000 = 84. Break-even visits
Each clinic has about USD 1.2 million of fixed costs a year; a visit earns USD 150 and costs USD 60.
Break-even visits a year:1,200,000 ÷ (150 - 60) = 13,3335. Curveball: payer mix
Interviewer: "About 30 percent of visits would be covered by Medicaid, which pays about USD 80." Blended revenue per visit:
Blended revenue per visit (USD):0.7 × 150 + 0.3 × 80 = 1296. New break-even
Fixed cost divided by the new contribution per visit.
Break-even visits a year:1,200,000 ÷ (0.7 × 150 + 0.3 × 80 - 60) = 17,3917. Profit per clinic at 15,000 visits
With the blended revenue.
Profit per clinic (USD a year):15,000 × (129 - 60) - 1,200,000 = -165,000
The recommendation
The metro area has about 1.2 million urgent-care visits a year, enough for about eight clinics at a 10 percent share, but not all of them would make money. First, at USD 150 per visit a clinic breaks even at about 13,300 visits. Second, with 30 percent Medicaid patients paying USD 80, break-even rises to about 17,400 visits, so a clinic at 15,000 visits loses about USD 165,000 a year. Third, the fix is better economics, not avoiding patients: choose sites on unmet demand and access, accept all payers, and close the gap through better Medicaid managed-care contracts, lower-cost staffing (nurse practitioners and physician assistants), evening hours to raise visits, and telehealth for simple cases.
Risks: Payer mix may differ by site; Retail clinics and telehealth rivals may take share.
Next steps: Map payer mix and competition by neighborhood; Model three clinics first, with evening hours.
A strong candidate
Sized the market cleanly, then showed how payer mix moves break-even and shaped the plan around it.
A weak candidate
Opened eight clinics because the market supports eight, without checking whether each one makes money.
Case 5: Mont Albret: a ski resort's profit fell by more than half
Where the structure comes from: it is built from the goal of this exact question (Profit = skier visits x contribution per visit - snowmaking - other fixed costs), not taken from a list. Each branch is one driver of that goal, and the hypothesis above says which branch to test first.
Worked case
Starter: Mont Albret: a ski resort's profit fell by more than half
The prompt
Mont Albret, a ski resort in Quebec, saw its profit fall by more than half this season. The exhibit shows the last two seasons. Why did profit fall, and what should the resort do?
Difficulty: Starter. Format: interviewer-led, with an exhibit. Industry: Travel and leisure. Region: Canada. Interview length: about 25 minutes. The company is fictional and all figures are illustrative.
Clarifying questions, with the interviewer's answers
- Where is the resort, and what happened to the season?Answer: It is in Quebec. Last season it opened for 120 days; this season warm spells cut it to 100 days (illustrative).
- Did prices change?Answer: Lift revenue per visit rose from CAD 60 to CAD 62. Spending on food and rentals stayed at CAD 25 a visit.
- Which costs are fixed?Answer: Lift staff, maintenance and insurance cost about CAD 16 million a year. Snowmaking is shown on its own line, and it rose this season.
A hypothesis to say out loud: Prices went up, so my hypothesis is that the fall comes from fewer skier visits in a short season, plus the cost of making more snow.
The structure
- Profit = skier visits x contribution per visit - snowmaking - other fixed costs
- Key: Skier visits: days open x visits per day
- Spending per visit: lift tickets, food and rentals
- Snowmaking and other fixed costs
- Weather risk: season passes and summer activities
The exhibit
| Measure | Last season | This season |
|---|---|---|
| Days open | 120 | 100 |
| Skier visits | 400,000 | 340,000 |
| Lift revenue per visit (CAD) | 60 | 62 |
| Food and rental spending per visit (CAD) | 25 | 25 |
| Variable cost per visit (CAD) | 15 | 15 |
| Snowmaking cost (CAD millions) | 3 | 4.5 |
| Other fixed costs (CAD millions) | 16 | 16 |
Working it through
1. Contribution per visit, last season
Lift revenue plus food and rentals, minus CAD 15 of variable cost per visit.
Contribution per visit last season (CAD):60 + 25 - 15 = 702. Contribution per visit, this season
The same, with lift revenue at CAD 62.
Contribution per visit this season (CAD):62 + 25 - 15 = 723. Profit last season
400,000 visits, minus CAD 3 million of snowmaking and CAD 16 million of other fixed costs.
Profit last season (CAD):400,000 × (60 + 25 - 15) - 3,000,000 - 16,000,000 = 9,000,0004. Profit this season
340,000 visits, with snowmaking up to CAD 4.5 million.
Profit this season (CAD):340,000 × (62 + 25 - 15) - 4,500,000 - 16,000,000 = 3,980,0005. Effect of fewer visits
60,000 fewer visits at last season's CAD 70 each. Extra snowmaking cost another CAD 1.5 million, and the higher lift revenue added back CAD 680,000.
Effect of fewer visits (CAD):(340,000 - 400,000) × (60 + 25 - 15) = -4,200,0006. Visits per open day
Last season it was 400,000 over 120 days, about 3,333 a day. This season:
Visits per open day this season:340,000 ÷ 100 = 3,4007. Curveball: another short winter
Interviewer: "Forecasters warn that next winter may be short too. How many visits does the resort need to cover this season's costs?"
Break-even visits a season:(4,500,000 + 16,000,000) ÷ (62 + 25 - 15) = 284,7228. Cash before the snow
Interviewer: "Our research says 8,000 people would buy a season pass at CAD 600 if it went on sale in spring." Candidate: "That money arrives whatever the weather:"
Season pass sales (CAD):8,000 × 600 = 4,800,000
What the exhibit shows
Skiers came as often per open day as before, so the short season and the extra snowmaking explain the fall, not weaker demand.
The recommendation
Mont Albret should protect itself against short seasons rather than cut prices, because the fall came from the weather, not from weaker demand. First, profit fell from CAD 9 million to CAD 3.98 million: fewer visits cost CAD 4.2 million and extra snowmaking CAD 1.5 million. Second, skiers still came at about 3,400 a day, so each open day was as busy as before. Third, the resort now needs about 284,722 visits to break even, which leaves little room in another short winter. Sell season passes in spring: 8,000 at CAD 600 bring CAD 4.8 million before the first snow. Also spend on snowmaking where it adds the most open days, and test summer activities.
Risks: Warm winters may become more common, so short seasons repeat; Pass holders may ski less often than expected, so food and rental spending falls.
Next steps: Launch early season passes in spring for next winter; Rank snowmaking spending by the open days each project adds.
A strong candidate
Split profit into visits, spending per visit and fixed costs, and saw that visits per open day held up. Then answered the weather risk with cash raised before the season.
A weak candidate
Blamed the higher lift price for the fall and suggested cutting it, which would lower profit without bringing back the lost days.
Case 6: Westbound Courier: cut parcel sorting costs with a machine?
Where the structure comes from: it is built from the goal of this exact question (Value of the machine = hours saved x cost per hour - upkeep, set against the upfront cost), not taken from a list. Each branch is one driver of that goal, and the hypothesis above says which branch to test first.
Worked case
Standard: Westbound Courier: cut parcel sorting costs with a machine?
The prompt
Westbound Courier runs a parcel sorting hub in Canada. Sorting by hand is its largest cost at the hub, and wages rise every year. Should it install an automated sorter?
Difficulty: Standard. Format: candidate-led, with interviewer dialogue. Industry: Logistics. Region: Canada. Interview length: about 30 minutes. The company is fictional. All figures are illustrative, except the facts that name their source.
Clarifying questions, with the interviewer's answers
- How much does the hub sort?Answer: About 60,000 parcels a day, 300 days a year. One person sorts about 150 parcels an hour by hand (illustrative).
- What do sorters cost?Answer: Westbound carries parcels between provinces, and courier services are federally regulated in Canada, so the federal minimum wage applies. It rose to CAD 18.15 an hour on 1 April 2026 and is adjusted every year with inflation (Government of Canada, checked 2026-10-02). Westbound pays sorters that minimum plus CAD 2.85, and benefits and payroll taxes add 25 percent (illustrative).
- What would a machine do?Answer: An automated sorter costs CAD 9 million installed, cuts sorting hours by 70 percent and costs CAD 400,000 a year to maintain. Westbound wants new equipment to pay back within five years (illustrative).
A hypothesis to say out loud: Hand sorting is a large cost that rises every year with the minimum wage. My hypothesis is that the machine pays back within five years, and faster if volume grows.
The structure
- Value of the machine = hours saved x cost per hour - upkeep, set against the upfront cost
- Sorting hours: parcels a year / parcels sorted per person per hour
- Cost per hour: wage, premium, benefits and payroll taxes
- Key: Yearly saving and payback against the five-year target
- Volume growth, breakdowns and staff plans
Working it through
1. Parcels a year
Candidate: "60,000 parcels a day for 300 days."
Parcels a year:60,000 × 300 = 18,000,0002. Sorting hours
Candidate: "At 150 parcels an hour per person."
Sorting hours a year:60,000 × 300 ÷ 150 = 120,0003. Cost per sorting hour
Candidate: "The federal minimum of CAD 18.15 plus CAD 2.85, with 25 percent on top."
Cost per sorting hour (CAD):(18.15 + 2.85) × 1.25 = 26.254. Yearly sorting cost
Candidate: "Hours times cost per hour."
Sorting labour cost (CAD a year):60,000 × 300 ÷ 150 × 26.25 = 3,150,0005. Net yearly saving
Candidate: "70 percent of the hours saved, minus CAD 400,000 of upkeep."
Net yearly saving (CAD):60,000 × 300 ÷ 150 × 0.7 × 26.25 - 400,000 = 1,805,0006. Payback
Candidate: "CAD 9 million divided by the net yearly saving."
Payback (years):9,000,000 ÷ (60,000 × 300 ÷ 150 × 0.7 × 26.25 - 400,000) = 4.997. Curveball: a new customer
Interviewer: "A large online retailer will send its parcels through us from next year, so volume rises 20 percent. The machine can handle up to 90,000 parcels a day." Candidate: "The net saving becomes:"
Net yearly saving with 20 percent more parcels (CAD):60,000 × 1.2 × 300 ÷ 150 × 0.7 × 26.25 - 400,000 = 2,246,0008. Payback with the new volume
Candidate: "The same CAD 9 million, divided by the larger saving."
Payback with 20 percent more parcels (years):9,000,000 ÷ (60,000 × 1.2 × 300 ÷ 150 × 0.7 × 26.25 - 400,000) = 4.01
The recommendation
Westbound should install the automated sorter, timed to arrive with the new retailer's parcels. First, hand sorting takes 120,000 hours a year at CAD 26.25 an hour, about CAD 3.15 million. Second, the machine saves a net CAD 1.805 million a year, so it pays back in about 4.986 years, just inside the five-year target. Third, with 20 percent more parcels the net saving rises to CAD 2.246 million and payback falls to about 4.007 years. Because the federal minimum wage is adjusted every year with inflation, the saving should also grow each year. Keep enough trained sorters to cover breakdowns, and agree an uptime guarantee with the supplier.
Risks: A breakdown at peak season could stop the hub if too few trained sorters remain; The new retailer's parcels may arrive later or in smaller numbers than promised.
Next steps: Ask two suppliers for fixed quotes with uptime guarantees; Plan which sorters move to loading and checking roles.
A strong candidate
Turned parcels into hours and hours into dollars, compared payback with the target, and saw that volume growth and yearly wage rises both strengthen the case.
A weak candidate
Compared the CAD 9 million price with one year of wages and rejected the machine as too expensive.
Case 7: Should a provincial health ministry fund a hip and knee surgery centre?
Where the structure comes from: it is built from the goal of this exact question (Change in the waiting list = new referrals - operations done, then the cost of adding operations), not taken from a list. Each branch is one driver of that goal, and the hypothesis above says which branch to test first.
Worked case
Standard: Should a provincial health ministry fund a hip and knee surgery centre?
The prompt
A provincial health ministry in Canada has a growing waiting list for hip and knee replacements. It is considering a public day surgery centre that would do only these operations. The exhibit shows the numbers. Should it fund the centre, and what must be true for it to work?
Difficulty: Standard. Format: interviewer-led, with an exhibit. Industry: Public sector and healthcare. Region: Canada. Interview length: about 30 minutes. The company is fictional and all figures are illustrative.
Clarifying questions, with the interviewer's answers
- Would the centre be public or private?Answer: Public. A regional health authority would run it, the province would pay for it, and patients would pay nothing.
- What target matters?Answer: The ministry wants the list down to about 3,000 people, which it expects would mean most patients wait under six months (illustrative).
- What limits how many operations the centre can do?Answer: Mainly staff: surgeons, anaesthesia staff and nurses. I will tell you more later.
A hypothesis to say out loud: The list grows because referrals outpace operations, so my hypothesis is that a dedicated centre can shrink it at a lower cost per operation, if it can be staffed.
The structure
- Change in the waiting list = new referrals - operations done, then the cost of adding operations
- Key: Waiting list flow: referrals in, operations out
- Extra operations from the centre, and years to reach the target
- Cost per operation in the centre and in main hospitals
- Staffing: surgeons, anaesthesia staff and nurses
The exhibit
| Measure | Value |
|---|---|
| People waiting today | 9,000 |
| New referrals a year | 13,000 |
| Operations in hospitals a year | 12,000 |
| Operations the centre could do a year | 2,500 |
| Cost per operation in a main hospital (CAD) | 18,000 |
| Cost per operation in the centre (CAD) | 14,000 |
| One-time setup cost of the centre (CAD millions) | 20 |
Working it through
1. How fast the list grows
New referrals minus operations done in hospitals each year.
Growth of the waiting list (people a year):13,000 - 12,000 = 1,0002. With the centre
Hospitals plus 2,500 centre operations, minus new referrals.
Fall in the waiting list with the centre (people a year):12,000 + 2,500 - 13,000 = 1,5003. Years to reach the target
From 9,000 people waiting down to 3,000.
Years to reach the target:(9,000 - 3,000) ÷ (12,000 + 2,500 - 13,000) = 44. Yearly cost of the centre's operations
2,500 operations at CAD 14,000.
Cost of centre operations (CAD a year):2,500 × 14,000 = 35,000,0005. Saving against buying them from hospitals
If the ministry paid main hospitals for the same 2,500 extra operations, each would cost CAD 18,000.
Yearly saving against hospitals (CAD):2,500 × (18,000 - 14,000) = 10,000,0006. Setup cost recovered
The CAD 20 million setup cost divided by the yearly saving.
Years to recover the setup cost:20,000,000 ÷ (2,500 × (18,000 - 14,000)) = 27. Curveball: not enough anaesthesia staff
Interviewer: "The health authority can only hire enough anaesthesia staff to run the centre at 80 percent of its capacity." The list then falls by:
Fall in the waiting list at 80 percent (people a year):12,000 + 2,500 × 0.8 - 13,000 = 1,0008. Years to the target at 80 percent
The same 6,000 people to clear, at the slower rate.
Years to reach the target at 80 percent:(9,000 - 3,000) ÷ (12,000 + 2,500 × 0.8 - 13,000) = 6
What the exhibit shows
Referrals exceed operations by 1,000 a year, so the list keeps growing until capacity is added.
The recommendation
The ministry should fund the centre, and secure its staff before it opens. First, the list grows by 1,000 people a year today, so waits will keep getting longer without new capacity. Second, 2,500 extra operations a year make the list fall by 1,500 a year, reaching the 3,000 target in 4 years. Third, each operation costs CAD 4,000 less than in a main hospital. So the centre saves CAD 10 million a year against buying the same operations from hospitals, and recovers its setup cost in 2 years. Staffing is the main risk: at 80 percent capacity the target takes 6 years. Recruit anaesthesia staff now, without drawing them from hospital operating rooms.
Risks: Hiring staff away from hospitals would cut hospital operations and cancel out the gain; Referrals may grow faster as the population ages.
Next steps: Agree a staffing plan with the health authority before approving the build; Track the list size and waiting times every month after opening.
A strong candidate
Treated the waiting list as a flow, found the yearly gap, timed the target, and tested the plan against the real limit, which is staff.
A weak candidate
Said the centre is cheaper per operation and stopped, without checking whether it shrinks the list or can be staffed.
Case 8: Should Cedarline Dental acquire a 12-clinic group in Alberta?
Where the structure comes from: it is built from the goal of this exact question (Deal value = EBITDA with savings x what the market pays for it - price paid, and what could take the EBITDA away), not taken from a list. Each branch is one driver of that goal, and the hypothesis above says which branch to test first.
Worked case
Stretch: Should Cedarline Dental acquire a 12-clinic group in Alberta?
The prompt
Cedarline Dental runs 40 dental clinics in Western Canada. It has been offered a group of 12 clinics in Alberta for CAD 36 million. The data pack shows the numbers. Should Cedarline buy, and on what terms? Write a one-page recommendation.
Difficulty: Stretch. Format: written case, with a data pack. Industry: Healthcare services. Region: Canada. Interview length: about 40 minutes. The company is fictional and all figures are illustrative.
Clarifying questions, with the interviewer's answers
- How is the price set?Answer: The seller asks 8 times this year's EBITDA (earnings before interest, tax, depreciation and amortisation), which is CAD 4.5 million (illustrative).
- What can Cedarline bring?Answer: Its buying terms would cut the group's supply costs by about 2 percent of revenue, and a shared back office would save CAD 0.9 million a year (illustrative).
- How are dental groups valued?Answer: For this case, assume groups like Cedarline are valued at about 9 times EBITDA (illustrative).
A hypothesis to say out loud: Cedarline can run the clinics more cheaply than the seller. My hypothesis is that the price is fair once the savings are counted, but that the deal depends on keeping the dentists.
The structure
- Deal value = EBITDA with savings x what the market pays for it - price paid, and what could take the EBITDA away
- Price against today's EBITDA and margin
- Key: Savings: supply buying terms and a shared back office
- Value created at the multiple dental groups trade at
- Dentist retention, and paying part of the price later
The exhibit
| Item | Value |
|---|---|
| Clinics | 12 |
| Revenue (CAD millions a year) | 30 |
| EBITDA (CAD millions a year) | 4.5 |
| Asking price (CAD millions) | 36 |
| Supply savings Cedarline can bring (% of revenue) | 2 |
| Back-office savings (CAD millions a year) | 0.9 |
| EBITDA of the three clinics run by owner-dentists (CAD millions a year) | 1.5 |
Working it through
1. Margin today
EBITDA of CAD 4.5 million on CAD 30 million of revenue.
EBITDA margin (%):4.5 ÷ 30 × 100 = 152. Price multiple
The asking price over today's EBITDA.
Asking price as a multiple of EBITDA:36 ÷ 4.5 = 83. Savings
2 percent of revenue from supply terms, plus CAD 0.9 million from a shared back office, in CAD millions a year.
Yearly savings (CAD millions):30 × 0.02 + 0.9 = 1.54. EBITDA with savings
Today's EBITDA plus the savings.
EBITDA with savings (CAD millions):4.5 + 30 × 0.02 + 0.9 = 65. Effective multiple
The price over EBITDA with savings.
Effective multiple with savings:36 ÷ (4.5 + 30 × 0.02 + 0.9) = 66. Value created
EBITDA with savings valued at 9 times, minus the price, in CAD millions.
Value created (CAD millions):(4.5 + 30 × 0.02 + 0.9) × 9 - 36 = 187. Curveball: the owner-dentists retire
Interviewer: "The three owner-dentists plan to retire within a year of the sale. If no one replaces them, their clinics' CAD 1.5 million of EBITDA is lost. For simplicity, keep the savings the same."
EBITDA with savings, without the three clinics (CAD millions):4.5 + 30 × 0.02 + 0.9 - 1.5 = 4.58. Multiple in that case
The same CAD 36 million over the lower EBITDA.
Effective multiple without the three clinics:36 ÷ (4.5 + 30 × 0.02 + 0.9 - 1.5) = 89. Price for the other nine clinics
At the seller's own 8 times, on the CAD 3 million EBITDA of the nine clinics not run by owner-dentists, in CAD millions.
Upfront price for nine clinics (CAD millions):(4.5 - 1.5) × 8 = 24
What the exhibit shows
A third of the EBITDA sits in three clinics whose owners plan to leave, so the terms of payment matter as much as the price.
The recommendation
Cedarline should buy the group, but pay CAD 24 million at closing and hold back the remaining CAD 12 million until the owner-dentists' clinics keep their earnings. First, the asking price is 8 times EBITDA of CAD 4.5 million, on a 15 percent margin. Second, Cedarline's buying terms and shared back office add CAD 1.5 million a year, lifting EBITDA to CAD 6 million, so the effective price is only 6 times. At the 9 times that groups like Cedarline are valued at, that creates about CAD 18 million of value. Third, three clinics depend on owner-dentists who plan to retire. If those clinics lose their CAD 1.5 million of EBITDA, the effective price goes back to 8 times and most of the value disappears. So pay 8 times the CAD 3 million EBITDA of the other nine clinics now. Pay the rest over three years, only if the three clinics keep their earnings, for example after new dentists join. The main risk is that patients follow their dentists elsewhere. Next, meet the three owner-dentists and agree how they will hand over their patients.
Risks: Patients may follow retiring dentists to other clinics; The supply savings may take longer than a year to arrive.
Next steps: Meet the three owner-dentists and agree a handover plan; Check each clinic's patient numbers for the last three years in due diligence.
A strong candidate
Valued the deal on EBITDA with savings, spotted that a third of the earnings sit with dentists who plan to leave, and turned that risk into the payment terms.
A weak candidate
Called 8 times EBITDA expensive and walked away, missing both the savings and the option of paying part of the price later.
Sources for this lesson (3)
- Recognized public explanations of case-interview concepts and terms
- Government of Canada, Employment and Social Development Canada: "Government of Canada raises the federal minimum wage" (news release, March 2026; CAD 18.15 an hour from 1 April 2026, adjusted each year with inflation), checked 2026-10-02
- Government of Canada: list of federally regulated industries and workplaces (includes postal and courier services), checked 2026-10-02
My notes on this lesson
0 of 5,000 characters. Saves automatically.
Completing lessons builds your skill levels and your readiness.
Next: put this module into practice
You have read every lesson in this module. A full case puts them to work, with a model answer to compare against.
Spotted something wrong or out of date? Report a mistake. We check every report and correct the page.