How do you structure a profitability case?
What this question is really asking
Profitability is the most common case archetype in consulting interviews, and the one where candidates most often reach for a memorised framework before they've understood the problem.
The underlying structure is almost embarrassingly simple: profit equals revenue minus cost, so something has happened to one or both. Everything else is disciplined narrowing. The skill being assessed isn't knowing the framework — everyone knows the framework. It's whether you can clarify the objective, split the problem cleanly, work out which branch matters, and get to a recommendation rather than a description.
The worked example here is a regional airline losing money.
How to structure your answer
Clarify before you structure. Two or three questions, no more. What does the client mean by profitability — absolute profit, or margin? Over what period has this been happening? Is this the whole business or a particular route network? Is the industry facing the same pressure, or is this specific to them? That last question saves candidates a great deal of wasted analysis.
Then state your structure out loud before using it. Profit is revenue minus cost, so you want to establish whether revenue has fallen, costs have risen, or both — then drill into whichever branch is driving it.
Revenue, for an airline: passengers multiplied by average fare. Passengers break into routes flown, capacity per flight, and load factor. Fare breaks into ticket mix, class mix, and ancillary revenue. Revenue per available seat mile is the industry framing.
Cost, for an airline: fuel, labour, aircraft ownership or lease, maintenance, airport and landing fees, distribution. Split fixed from variable — an airline is heavily fixed-cost, which is why load factor drives the economics so hard. Cost per available seat mile is the counterpart metric.
Ask for the data and isolate. Which side moved? Then which line within it? Keep narrowing until you find something specific enough to act on.
Look outward as well as inward. If fuel rose for everyone, it isn't the client's problem to solve — the question becomes why they're worse affected. Hedging policy, fleet age and fuel efficiency, route mix. Competitive comparison usually determines whether the answer is operational or strategic.
Close with a recommendation, quantified. Name the driver, propose two or three actions, size the impact roughly, and state the main risk. A case that ends with a diagnosis and no recommendation is an incomplete answer, however good the analysis was.
What the interviewer is testing
Whether you clarify first. Diving into structure without understanding the objective is the most common early error, and interviewers are watching for it specifically.
Whether your structure is MECE. Mutually exclusive, collectively exhaustive. Overlapping buckets suggest you haven't thought it through; gaps suggest you're reciting.
Whether you're hypothesis-driven. Strong candidates form an early view and test it. Weaker ones examine every branch equally and run out of time.
Whether you can do arithmetic out loud. Cases involve numbers, delivered verbally, under time pressure. Comfort here is visible immediately.
Whether you segment. Aggregate numbers hide the answer. Routes, customer types, time periods — the driver usually lives in a segment rather than the average.
Whether you synthesise. The last sixty seconds matter disproportionately. Interviewers remember whether you landed the plane.
Frequently asked
Is profit equals revenue minus cost too simple a structure?
No — it is the correct starting point, and interviewers expect it. What distinguishes candidates is the level below: how cleanly you break down each side, and which branch you choose to pursue based on early data.
How many clarifying questions should I ask?
Two or three, focused on the objective, the timeframe, and whether the problem is industry-wide or company-specific. More than that starts to look like stalling rather than structuring.
Should I name a framework?
Describe your structure rather than labelling it. Saying you want to look at revenue and cost separately, then explaining how you would break each down, is stronger than naming a framework and letting the interviewer infer the content.
What if the data does not point clearly to one branch?
Then both moved, and you size which contributed more. Saying so explicitly — that revenue explains roughly two thirds of the gap and costs the rest — is a stronger answer than forcing a single cause.
How do I decide which branch to go down?
Ask for high-level data on both sides before drilling. One question about the direction of revenue and cost over the period usually tells you where the problem is, and choosing based on data rather than instinct is exactly the behaviour being assessed.
Common mistakes
Reciting a framework before understanding the problem. Interviewers can tell within seconds, and it signals memorisation rather than thinking.
Skipping clarification. Two questions at the start prevent ten minutes of analysing the wrong thing.
Jumping straight to cost cutting. It's the reflex, and it's frequently wrong. Establish which side moved before proposing remedies.
A structure that isn't MECE. Overlapping or incomplete buckets undermine everything built on top.
Not segmenting. "Revenue is down 8%" is where the analysis starts, not where it ends. Down where, for whom, on which routes?
Never forming a hypothesis. Examining every branch with equal weight is how candidates run out of time with no answer.
Getting lost in a branch that doesn't matter. If costs are flat, do not spend eight minutes on cost structure. Say what you found and move.
Ignoring the competitive lens. Industry-wide problems and company-specific problems have different answers.
No synthesis. Ending on "so those are the factors" wastes the case. Name the driver, give the recommendation, state the risk.
Practice this question with an AI interviewer that pushes back, asks follow-ups, and gives instant feedback.
Start your free mock →