A non-standard case prompt is one that arrives without a recognisable case type attached — a perception problem, a fixed budget, a soft objective — and it is solved by decomposing the objective until it becomes measurable rather than by selecting a framework. This page is a full practice case built on exactly that kind of prompt, laid out the way an interview runs: the prompt, the clarifying questions worth asking, the data an interviewer would release when asked, the arithmetic, and a model recommendation.
Work it under time before reading past the prompt. Thirty minutes, paper, no calculator. The figures are fictional and constructed for practice.
The prompt
“There is a widespread national perception that air travel is unpleasant. Our client is a consortium of airlines operating in a single country. They have engaged us to change that perception within one year, with a budget of ten million dollars. What would you do?”
Stop here and work it.
Step 1: Make the objective measurable
The prompt gives an objective that cannot be tested as stated. “Change the perception” has no measure, no target and no defined population, so nothing in the analysis could be sized against it. The first move is to fix that, and it is the move most candidates skip.
Questions worth asking, and the answers this interviewer gives:
Whose perception? — Focus on people who could realistically fly and do not, or who fly less than they otherwise would. Existing frequent flyers are not the target.
What do they believe, specifically? — The consortium has survey data. It is available if you ask.
How is success measured? — The consortium wants to see it in demand, not only in sentiment. Treat a measurable increase in leisure bookings from infrequent flyers as the goal.
Is the ten million a one-year budget or total? — One year, all in.
Can the consortium change how airlines actually operate, or only how they communicate? — This is the best question in the case. The consortium can coordinate members and fund shared initiatives, but it cannot compel an individual airline to change its own pricing or service.
That last answer reshapes everything. It means the client can fund and coordinate, but not mandate — so any recommendation requiring one airline to act alone against its own interest will not happen.
Restated objective: increase leisure bookings from infrequent flyers within twelve months, by addressing the specific beliefs that keep them from flying, with ten million dollars, using interventions the consortium can actually deliver.
That is now a case that can be structured.
Step 2: Structure
The useful structure separates the belief from its cause, because the two need different responses.
- What do people actually believe, and which beliefs suppress demand? Not every complaint changes behaviour.
- For each belief, is it accurate? A belief that reflects real experience needs the experience fixed. A belief that has outlived the experience needs communication.
- What can the consortium do about each, within ten million and twelve months?
- Which options give the largest demand response per dollar?
The screen at step 3 is what a fixed budget is for. Ten million dollars is a large marketing budget and a trivial infrastructure budget, which means the answer will not be capital investment — and knowing that early stops you analysing options the client cannot buy.
Step 3: The data
Requested at step 1, the consortium’s survey of infrequent flyers gives the reasons cited for avoiding air travel:
| Cited reason | Share citing it | Accurate today? | Consortium can influence? |
|---|---|---|---|
| Security queues and airport time | 61% | Partly — varies sharply by airport and time | Partly, via airport coordination |
| Total cost once fees are added | 54% | Yes | No — pricing is each airline’s own |
| Fear of delays and disruption | 47% | Overstated relative to actual performance | Yes |
| Cramped seating | 43% | Yes | No |
| Booking process complexity | 29% | Partly | Yes |
| Environmental concern | 18% | — | Limited |
And the operating data:
- Roughly 70 per cent of flights arrive within fifteen minutes of schedule.
- Median security wait is 18 minutes, but the 90th percentile is over an hour.
- Infrequent flyers take about 0.4 leisure trips a year; the consortium’s target is a 10 per cent increase.
- Average leisure fare is $220, with roughly $60 of contribution per passenger.
Step 4: Read the data before acting on it
Three findings, and the second is the one the case is built around.
The largest cited reason is not the largest addressable one. Cost is cited by 54 per cent, and the consortium cannot touch it. Analysing pricing here would be analysing something the client cannot do.
The delay perception is a genuine gap between belief and reality. 47 per cent fear delays; 70 per cent of flights arrive within fifteen minutes. This is the one belief that is measurably worse than the experience, which makes it the one where communication is the correct instrument rather than a substitute for fixing something. Everywhere else, communication would be advertising against the customer’s own experience — which does not work and damages credibility.
The security complaint is really a variance complaint. An 18-minute median with a 90th percentile over an hour means the problem is unpredictability rather than duration. People do not plan around a median; they plan around the worst case, and then arrive two hours early, which is itself the unpleasant experience. That reframing matters because reducing variance is much cheaper than reducing average wait.
Step 5: Size the prize
Before choosing interventions, establish what a 10 per cent increase is worth, so the ten million can be judged against it.
The interviewer supplies the population: roughly 12 million infrequent flyers in the country.
- Current trips: 12m × 0.4 = 4.8 million trips a year
- A 10 per cent increase: 480,000 additional trips
- Contribution at $60: $28.8 million a year
Against a ten million dollar budget, that is a strong return if achievable — but the increase is spread across the consortium’s members, and the budget is spent by the consortium. Worth naming as a governance point rather than an economic one.
Sanity check. 480,000 additional trips is about 4 per cent of the infrequent-flyer population taking one extra trip. That is a meaningful behavioural shift but not an implausible one for a category where the barrier is reluctance rather than affordability. It is achievable, and it is not easy.
Step 6: The recommendation
Allocate against the two things the consortium can actually change, and be explicit about what is being left alone.
1. Close the delay perception gap — $4 million.
Publish real, verified punctuality data per route and per airline, in the booking flow at the point of choice rather than in a campaign. The gap between 47 per cent fear and 70 per cent on-time performance is the single largest belief-versus-reality gap in the data, and it is the only one where telling people the truth improves their experience of the decision. The consortium is well placed to do this because it is credible in a way an individual airline advertising its own punctuality is not.
2. Attack security variance, not security duration — $4 million.
Fund shared queue-time measurement and publication across member airports, and a coordinated slot-booking system for security. Reducing the 90th percentile matters more than reducing the median, because the worst case is what people plan around. This is operational rather than communicative, which is why it is durable, and it is within reach at this budget because measurement and coordination are cheap relative to infrastructure.
3. Simplify booking — $1.5 million.
The 29 per cent citing complexity are describing something specific and fixable: fee structures that make total cost unclear until late in the flow. A consortium-wide standard for displaying all-in prices early addresses part of the cost perception indirectly, without requiring any airline to change its actual prices.
4. Hold $500,000 for measurement.
Baseline the sentiment and booking data now, and measure at six and twelve months. Without this the consortium cannot tell whether it worked, and the twelve-month horizon leaves no room to find out late.
What is deliberately not funded: a general brand campaign. Advertising that air travel is pleasant, to people whose direct experience says otherwise, spends the budget on the one intervention the data argues against.
Step 7: State the risks
The delay data may not survive publication. Publishing per-airline punctuality will show some members performing badly, and those members may block it. This is the main implementation risk and it is political rather than analytical — worth raising with the consortium’s governance before committing the four million.
Twelve months is short for a behavioural change. Sentiment can move in a year; booking behaviour in a category people engage with twice a year may lag. Recommend the consortium agree in advance that the six-month read is on sentiment and leading indicators rather than on bookings.
What would change the recommendation: if the punctuality figure were closer to the perception — say 50 per cent on time rather than 70 — then the belief would be accurate, publication would confirm rather than correct it, and the entire budget should shift to operational improvement. The first thing to verify is that the 70 per cent figure is a true fifteen-minute arrival measure and not a departure or schedule-padded one.
What this case is testing
Whether you decompose an objective that arrives without a framework attached. Whether you treat the budget as a constraint that shapes the structure rather than a detail. Whether you separate a belief from its cause, and notice the one place where they diverge. And whether you can recommend against the obvious intervention — the advertising campaign — with a reason.
The habit that generalises: when a prompt feels unstructurable, the problem is almost always that the objective has not been made measurable yet. The case interview opening guide covers that move in detail.
More worked cases in standard formats are in the case study library, and the case interview hub sets out a full preparation sequence.
