All chapters
How to think in a case
Day 2 of 3 · Start here
Start here · Your first 3 days - Day 2
02

How to think in a case

Learn the basic case-thinking loop: start with the goal, build the path, test the path.

Read time
~7 min
Day
02
Level
Start here
What you'll take away
  • Understand why case thinking is structured movement through uncertainty.
  • Use a simple goal, structure, prioritize, test, update loop.
  • Practice a first hypothesis without pretending to know the answer.

Yesterday, you learned what a case interview is: a short working session where you and the interviewer move through a messy business problem together. Today is about the kind of thinking underneath that session.

The beginning of case confidence
I may not know the answer yet, but I know how to start finding it.

The intuition

Case thinking is structured movement through uncertainty. You start with a goal. You break the problem into parts. You choose a sensible first place to look. Then you let the evidence update your view.

That is it. Not because cases are easy. They are not. But because the interview is not asking you to be magically right from the first second. It is asking whether you can stay calm, create a path, and move the problem forward.

Weak opening

I would use a profitability framework and look at revenue, costs, customers, competitors, and operations.

Stronger opening

Since profit fell, I would first separate the problem into revenue and cost drivers, then compare revenue and costs versus last quarter to see which side drove the decline.

The second answer is not stronger because it is trying to impress. It is stronger because it shows a path.

Frameworks are tools, not scripts

Frameworks are useful because they give your brain a starting shape. Profit can be broken into revenue and costs. Market entry can be broken into market attractiveness, competitive position, economics, and risks. Growth can be broken into more customers, more spend per customer, higher retention, or new channels.

But memorizing frameworks can hurt when the framework starts driving the case instead of the client question. The interviewer is not hoping you recite a famous template. They are watching whether your structure fits the problem in front of you.

Use frameworks like training wheels
They help you feel the shape of the skill. The real skill is learning to build the right structure for the actual case.

The basic case-thinking loop

LoopStart with the goal. Build the path. Test the path.
  1. 1
    GoalWhat are we trying to solve?
  2. 2
    StructureWhat are the main parts of the problem?
  3. 3
    PrioritizeWhere should we look first?
  4. 4
    TestWhat data would confirm or change our view?
  5. 5
    UpdateWhat does the evidence tell us now?
That is case thinking in miniature.

What this sounds like

Same prompt · calmer thinking
AI
Prompt
A coffee chain's profits dropped 15% last quarter. What happened, and what should they do?
YOU
Calm opening
I would first clarify that the goal is to understand why profit fell and then recommend actions to recover it. To diagnose the issue, I would split profit into revenue and costs. On revenue, I would look at customer traffic, average order value, product mix, and channel mix. On costs, I would look at ingredients, labor, rent, delivery fees, and waste. I would start by comparing revenue and costs versus last quarter to see which side explains most of the decline, then go deeper into the biggest driver.
Goal, structure, business logic, first analysis - without pretending to know the answer yet.
Why it works
  • It confirms the goal.
  • It proposes a clear, multi-layer framework instead of a flat list.
  • It chooses a first analysis.
  • It does not pretend to know the answer yet.

Good case thinking is not fake certainty. It is organized curiosity.

A first hypothesis, lightly held

You may hear people say consulting is hypothesis-driven. That can sound intimidating, but the basic idea is simple: based on what I know so far, what do I think is most likely, and how would I test it?

A hypothesis is a testable starting point
It is not your final answer. It is your best current view of what might be driving the problem, based on the facts you have so far. The point is to give your analysis direction, then test it with data and update quickly if the evidence points somewhere else.
Weak

The issue is definitely competition.

Strong

Competition may be a driver, so I would test whether traffic fell in the affected stores.

For the coffee chain, if you hear that a competitor opened nearby last quarter, you might say that the profit decline may be revenue-driven, possibly from lower customer traffic. Then you would test that by looking at traffic, average order value, and revenue by region before assuming that is the answer.

Common beginner traps

Watch for these
  • Trying to memorize too many frameworks.
  • Starting analysis before clarifying the goal.
  • Creating buckets that are too broad to be useful.
  • Treating the first hypothesis like a final answer.
  • Panicking when the case does not match a template you have seen before.

The deeper trap is thinking strong candidates already know the answer. They usually do not. They just know how to begin.

Next, Day 3 walks through a simplified mini case so you can feel how the pieces move together before you try to solve one on your own.

Up next
Day 3 · Walk through your first mini case
Follow a shorter, guided coffee-chain case from the case question to recommendation before moving into the Core Toolkit.
Continue to Day 3
Finished Day 2?Mark it complete to track your progress through the handbook.
← PreviousWhat a case interview really isUp next →Walk through your first mini case