Public, social, and non-profit cases
Government, charity, and non-profit cases: agree the real objective and success metric, then compare options on outcome per unit of money and commit to an answer.
Key takeaways
- Agree the objective and the success metric first, then compare options on outcome per unit of money, state the trade-off, and commit to a recommendation.
- Many firms have dedicated public and social-sector practices, and public-sector cases are frequent in some regions, notably the Gulf.
- The strong answer optimizes for the agreed objective and commits. The weak one applies the wrong logic or avoids a decision.
- Why a generic structure scores lower: Treating a public case like a business and picking whatever is cheapest misses outcome per unit of money and who is left out, which is what the interviewer scores.
What this case type is and when it shows up
A public or social-sector case involves a government, a charity, or a non-profit, where the goal is impact (health, education, jobs, access) rather than profit. Government cases are common in the Gulf, India, Singapore, and Europe, and several firms have large public-sector practices. The same structuring and math apply, but you must first agree what success means.
The underlying theory, in plain language
Start by naming the objective and the unit of success: students helped, lives improved, cost per outcome, jobs created. Applying pure profit logic to a mission organization misreads the problem.
Then bring the same rigor: structure the problem, size the options, and weigh cost against outcome. Budgets are usually fixed, so cost-effectiveness (outcome per unit of money) often becomes the key measure. When one option costs more per person, it wins only if its effect per person is larger by a bigger factor than its cost.
Typical government prompts: how to reach a national target (for example more tourists, more citizens in private-sector jobs, faster permit approvals), how to deliver a public service at lower cost, or how to choose among programs with a fixed budget. The objective often mixes several goals (impact, cost, fairness, speed), so agree on the main metric with the interviewer before analyzing.
What the prompts sound like, from simple to hard
- Simple: how can a charity in India reach more children on a fixed budget.
- Medium: a Gulf city wants to reduce traffic jams; evaluate the options.
- Hard: a ministry must reach a national target for citizens working in the private sector within five years.
Build the structure from the goal
Three moves that give you the structure
- 1Start from the decision. What outcome does the ministry or charity want, how will success be measured, and which option delivers the most of it for the money?
- 2Write the maths of the goal. Total outcome = people reached x effect per person. With a fixed budget, people reached = budget / cost per person, so compare effect per person against cost per person across options, and check who is left out.
- 3Let the business pick the branches. Knowing the sector tells you what limits reach and what counts as an effect. In schools, teacher time and attendance limit what works. In health, the people who most need a service are often the hardest to reach. In transport, fairness to people without cars matters alongside cost.
- Minutes saved per unit of money
- Riders on the routes affected
- Key: Minutes saved per trip under each option
- Cost: lanes once, buses every year
- Who gains: riders without cars; who loses: drivers who lose a lane
This comes from total outcome = people reached x effect per person, per unit of money. Bus lanes cost little but change minutes per trip, so the effect branch leads.
- Families fed per unit of money
- Key: Families in need who do not come today, and why
- Cost per family: food, storage, volunteers
- Reach: a mobile pantry or partner schools
- Food wasted before it is used
Same maths, a different sector. A food bank is limited by reaching families, not by food, so the branch on why families do not come leads.
Treating a public case like a business and picking whatever is cheapest misses outcome per unit of money and who is left out, which is what the interviewer scores.
Crack any case in five movesFinding and narrowing the real problem
Key idea
Agree the objective and the success metric first, then compare options on outcome per unit of money, state the trade-off, and commit to a recommendation.
Names you may hear, kept as questions that fall out of the goal maths
- Objective and metric first: What outcome counts, and how will we measure it? The goal maths cannot start without it. Where it stops helping: Stakeholders may disagree on the goal; raise that early.
- Cost-effectiveness: How much outcome does each option give per unit of money? Where it stops helping: Some outcomes are hard to measure and should not be ignored because of that.
- Cost ratio versus effect ratio: Does the dearer option's effect per person beat the cheaper one by more than its extra cost per person? Where it stops helping: Needs effects measured the same way for each option.
Methods for solving this type
- Agree the objective and the success metric
- Map the options to reach it
- Estimate outcome per unit of money for each
- Recommend, stating who is served and who is left out
The math patterns it relies on
- People reached = budget / cost per person
- Total outcome = people reached x effect per person
- Cost ratio versus effect ratio
Worked cases
Worked case
Choosing a tutoring program with a fixed budget
The prompt
A Gulf ministry of education has AED 10 million to raise math results for grade 8 students. Program A, a tutoring app, costs AED 200 per student and raises test scores by 5 points on average. Program B, small-group tutoring in person, costs AED 500 per student and raises scores by 15 points. The table below summarizes the two programs. Which should it fund?
Interviewer-led: the interviewer shows the program table and asks for reach, total improvement, and a recommendation that names the trade-off.
Clarifying questions, with the interviewer's answers
- What is the main goal: the most total improvement, or getting as many students as possible above a pass mark?Answer: The most total improvement in scores.
- Are the effects measured the same way?Answer: Yes, both come from trials using the same national test.
- How many grade 8 students need help?Answer: About 60,000, more than either program can reach.
A hypothesis to say out loud: Program B costs 2.5 times as much per student, so my hypothesis is that B wins only if its effect is more than 2.5 times A's. The trials suggest 3 times, so B is likely better for total improvement.
The structure
- Total improvement per AED 10 millionThis comes from total improvement = students reached x effect per student, where students reached = budget / cost per student.
- Students reached by each program
- Key: Cost ratio versus effect ratio
- Total improvement, and who is left out
The exhibit
| Program | Cost per student (AED) | Average score gain (points) | Evidence |
|---|---|---|---|
| A: tutoring app | 200 | 5 | Trial in 40 schools |
| B: small-group tutoring in person | 500 | 15 | Trial in 12 schools |
Working it through
1. Reach of A
AED 10 million at AED 200 per student.
Students reached by A:10,000,000 ÷ 200 = 50,0002. Reach of B
AED 10 million at AED 500 per student.
Students reached by B:10,000,000 ÷ 500 = 20,0003. Cost ratio
How much more B costs per student.
Cost ratio B to A:500 ÷ 200 = 2.54. Effect ratio
How much more B raises scores per student. 3 is more than 2.5, so B gives more improvement per AED.
Effect ratio B to A:15 ÷ 5 = 35. Total improvement from A
Students reached times points gained.
Total points, A:50,000 × 5 = 250,0006. Total improvement from B
Students reached times points gained.
Total points, B:20,000 × 15 = 300,0007. Compare
B gives about 20 percent more total improvement.
B versus A:300,000 ÷ 250,000 = 1.2
What the exhibit shows
B costs 2.5 times as much per student but raises scores 3 times as much, so it gives more improvement per AED, though its evidence comes from fewer schools.
The recommendation
Fund Program B. First, it costs 2.5 times as much per student but raises scores 3 times as much, so it gives more improvement per AED. Second, the same AED 10 million produces about 300,000 points of improvement against 250,000 for A, about 20 percent more. Third, the trade-off is reach: B serves 20,000 students instead of 50,000, leaving most of the 60,000 who need help without support, so give B to the students furthest behind, where in-person help matters most, and propose funding the cheaper app for the rest in the next budget. Recheck the effects after the first term, because trial results often shrink when a program grows.
Risks: Trial effects often shrink at larger scale; Finding enough qualified tutors for B may take time.
Next steps: Identify the 20,000 students furthest behind; Set up measurement after the first term.
A strong candidate
Agreed the metric, compared the cost ratio with the effect ratio, committed to B, and addressed who is left out.
A weak candidate
Picked the cheaper program because it reaches more students, or ended with "it depends on the objective."
Worked case
More health screenings for older adults in Singapore
The prompt
A Singapore government agency wants more adults aged 50 to 69 to take a free health screening. It has two main tools: text-message reminders and community outreach by volunteers and nurses. How should it spend its budget?
Candidate-led: you agree the metric, ask for costs and effects, and build the plan; the interviewer answers and challenges. The agency and its figures are illustrative.
Clarifying questions, with the interviewer's answers
- What is the goal, and how is success measured?Answer: More people aged 50 to 69 taking the free health screening. Success is the number of extra screenings this year; the target is 15,000.
- What is the budget?Answer: SGD 3 million for this year.
- How many people in the group are not screened today?Answer: About 400,000.
A hypothesis to say out loud: Cheap reminders reach many people but move few of them, while personal outreach moves more people at a higher cost. My hypothesis is that the best plan uses reminders for everyone and spends the rest of the budget on outreach, so I will compare the cost of each extra screening.
The structure
- Extra screenings per SGD, within the budget and the targetThis comes from extra screenings = money spent on each tool / cost per extra screening, up to the number of people each tool can reach.
- Key: Cost per extra screening for each tool
- Limits: how many people each tool can reach
- Best mix within SGD 3 million, against the 15,000 target
- Who is reached and who is left out
Working it through
1. Reminders: cost per extra screening
Candidate: "What does a reminder cost, and how many people does it move?" Interviewer: "About SGD 1 per person, and 2 in every 100 people reminded then get screened who otherwise would not."
Cost per extra screening, reminders (SGD):1 ÷ 0.02 = 502. Outreach: cost per extra screening
Interviewer: "Outreach costs about SGD 40 per person visited, and 15 in every 100 visited then get screened who otherwise would not, on top of any reminder."
Cost per extra screening, outreach (SGD):40 ÷ 0.15 = 2673. The limit on reminders
Interviewer: "Then why not spend it all on reminders?" Candidate: "Because there are only 400,000 people to remind. Can we remind them several times?" Interviewer: "A second reminder adds very little." Candidate: "Then reminders reach everyone for SGD 400,000 and stop there."
Extra screenings from reminders:400,000 × 0.02 = 8,0004. Outreach with the rest of the budget
The remaining SGD 2.6 million pays for this many visits.
People visited:(3,000,000 - 400,000) ÷ 40 = 65,0005. Extra screenings from outreach
15 in every 100 of the people visited.
Extra screenings from outreach:(3,000,000 - 400,000) ÷ 40 × 0.15 = 9,7506. Total against the target
Reminders plus outreach, against the target of 15,000.
Total extra screenings:400,000 × 0.02 + (3,000,000 - 400,000) ÷ 40 × 0.15 = 17,750
The recommendation
I recommend sending reminders to all 400,000 people and spending the other SGD 2.6 million on community outreach, for about 17,750 extra screenings against the 15,000 target. First, reminders are the cheapest tool at about SGD 50 per extra screening, giving about 8,000. Second, outreach costs about SGD 267 per extra screening but is the only way to use the rest of the budget, adding about 9,750. Third, outreach reaches people that text messages miss, so aim it at the neighborhoods with the lowest screening rates. Track screening rates by neighborhood each quarter.
Risks: The 2 percent reminder effect comes from past campaigns and may be lower now; Clinics may not have enough appointments if demand rises quickly; Outreach quality depends on training volunteers well.
Next steps: Confirm clinic capacity for about 18,000 extra screenings; Pick the 20 neighborhoods with the lowest screening rates for outreach; Test two reminder messages to see which works better.
A strong candidate
Agreed the metric, compared cost per extra screening, noticed the limit on reminders, and built a mix that meets the target and reaches the people most often missed.
A weak candidate
Put the whole budget into reminders because they are cheapest per person, without noticing that each person can only be reminded once.
Prompt: "Which program should the ministry fund?"
Weaker answer
Treats it like a business and picks whichever is cheapest, or lists the trade-offs and never recommends.
Stronger answer
Agrees the success metric, shows B's effect ratio (3) beats its cost ratio (2.5), recommends B for the students furthest behind, and names the reach trade-off.
Why the stronger answer wins: The strong answer optimizes for the agreed objective and commits. The weak one applies the wrong logic or avoids a decision.
Common mistakes, traps, and curveballs
- Optimizing profit where impact is the goal
- Ending with "it depends" instead of a recommendation
- Ignoring who is served and who is left out
- Choosing the eye-catching option over the cost-effective one
- Skipping measurement because outcomes seem hard to measure
Many firms have dedicated public and social-sector practices, and public-sector cases are frequent in some regions, notably the Gulf. These cases test whether you keep the same rigor while respecting that the goal is not money. Formats differ by office and change over time, so check the current process for your target office.
Practice
A job-training program in India costs INR 300 million (INR 30 crore) and helps 12,000 people find jobs. What is the cost per job, in INR?
A provincial vaccination drive in South Africa costs ZAR 18 million and vaccinates 120,000 children. What is the cost per child vaccinated, in ZAR?
A job program in Spain costs EUR 900 per person, and 12 percent of participants find a job they would not have found otherwise. With a budget of EUR 3 million, how many extra people find jobs?
Program X costs 3 times as much per person as Program Y. When is X the better use of a fixed budget?
Tool A costs SGD 1 per person reached and Tool B costs SGD 40. Why might Tool B still be part of the best plan?
A ministry wants both the largest total improvement and fairness to the students furthest behind. What should you do first?
Agree the metric, compare the cost ratio with the effect ratio, commit to an answer, and say who is left out.
Sources for this lesson (1)
- Recognized public explanations of case-interview concepts and terms
My notes on this lesson
0 of 5,000 characters. Saves automatically.
Try the 6 remaining checks and drills above to complete this lesson (0 of 6 done).
Spotted something wrong or out of date? Report a mistake. We check every report and correct the page.