What the case method expects
The case method trains judgment under incomplete information. Cases are written so that no answer is obviously right, and instructors grade the quality of your reasoning, not whether you matched a hidden solution. Three habits separate strong responses from average ones: you commit to a decision, you support it with numbers drawn from the case, and you say what could make you wrong.
Start by finding out which of three tasks you have been given. Many written cases come with questions at the end, and those questions are the rubric.
| Task | Typical wording | What a strong answer does |
|---|---|---|
| Decide | What should the CEO do? | States the decision in the first paragraph and defends it |
| Diagnose | Why is performance declining? | Separates symptoms from causes before proposing fixes |
| Evaluate | Was the acquisition a good move? | Sets criteria first, then judges against them using evidence |
Read the case in three passes
Do not read a case once from beginning to end with a highlighter. Read it three times with a different purpose each time.
- Skim (10 minutes). Read the opening and closing paragraphs, headings and exhibits. Write one sentence: who must decide what, and by when?
- Read for facts (30 to 40 minutes). Mark numbers, dates, quotations from managers and any statement of strategy. Note what the case does not tell you.
- Read for conflict (20 minutes). Underline disagreements between people and the tensions between goals, such as growth against margin or speed against quality.
Cases often hide the real issue in an exhibit or a throwaway comment. A price list, a footnote on customer churn or an offhand remark by a middle manager is frequently where the diagnosis lives.
Frame the problem before you analyze it
The most common failure is solving the wrong problem. Write the problem as a decision with a deadline and a constraint: Should Harbor Coffee open ten company-owned stores in the Southwest next year while keeping debt below 2.5 times earnings? A good problem statement is short, specific and decision-shaped.
| Weak problem statement | Strong problem statement |
|---|---|
| The company has declining sales. | Should the company cut prices, add a premium line or exit the low-margin segment to stop a 6 percent annual decline in operating profit? |
| Leadership is poor. | Should the board replace the division head now, or fund a six-month turnaround plan with clear milestones? |
Next, pick frameworks that fit the question. Do not apply all of them. Strategy questions suit industry and capability analysis (see our strategic management guide); entry or investment questions need financials; organizational questions need stakeholder and change analysis. State why each framework is used in one sentence and then use it to produce a finding, not a diagram.
Generate options and score them honestly
Offer at least three realistic options, including a do-nothing or minimal option as a baseline. Then compare them against criteria drawn from the company's own goals. A weighted scorecard makes your reasoning visible and is easy for a reader to challenge, which is the point.
Weighted scorecard (hypothetical)
Weights are chosen to reflect what the board cares about: strategic fit 30 percent, financial return 30 percent, risk 20 percent, feasibility 20 percent. Scores run from 1 (poor) to 5 (strong); for risk, 5 means low risk.
| Option | Fit (30%) | Return (30%) | Risk (20%) | Feasibility (20%) | Weighted score |
|---|---|---|---|---|---|
| A: Expand into new region | 4 | 3 | 2 | 3 | 1.2 + 0.9 + 0.4 + 0.6 = 3.1 |
| B: Deepen existing markets | 3 | 4 | 4 | 4 | 0.9 + 1.2 + 0.8 + 0.8 = 3.7 |
| C: Acquire a competitor | 5 | 2 | 1 | 2 | 1.5 + 0.6 + 0.2 + 0.4 = 2.7 |
Option B scores highest. The ranking is sensitive to the weights, so test it: if return is weighted at 50 percent and the others cut proportionally, check whether the winner changes and say so.
A scorecard is a thinking tool, not a substitute for argument. Explain two or three of the scores in words, especially any that are close.
Put numbers behind the recommendation
Every case has some quantitative evidence, and using it separates an opinion from an analysis. Choose the few calculations that decide the question and show your working.
| Question | Useful calculation |
|---|---|
| Is the business healthy? | Margins, growth, return on invested capital, trend over three years |
| Can we afford the move? | Payback, NPV, funding need, debt to earnings after the move |
| Is the new line worth it? | Contribution margin, break-even volume, cannibalization of existing sales |
| Is the market attractive? | Size, growth rate, concentration and customer switching costs |
State every assumption you make that the case does not give you, and test the ones that matter with a quick sensitivity check: what if volume is 20 percent lower, or the cost is 15 percent higher? Put detailed calculations in an appendix and the result in the text. See our guide to cost of capital for the discount rate step.
Working on this assignment now? Get a price for help with your paper.
Get an instant quoteWrite the response
Open with the recommendation, then the reasons, then the evidence. Readers should know your answer within the first 100 words.
| Section | Contents | Approximate share |
|---|---|---|
| Recommendation | The decision, the main reason and the key number | 10 percent |
| Situation and problem | Only the facts needed to understand the problem | 15 percent |
| Analysis | Framework findings and calculations, each ending with so what | 35 percent |
| Options and comparison | The scorecard and the argument for the winner | 20 percent |
| Implementation | First 90 days, owners, milestones, resources | 10 percent |
| Risks and contingencies | What could go wrong, early warnings, a fallback | 10 percent |
- Use the case, not the internet Unless told otherwise, rely on information available to the decision-maker at the time.
- Write so what after each finding Say what the evidence means for the decision.
- Be specific in implementation Name who does what by when, not "improve communication".
- Acknowledge the strongest objection Then explain why your recommendation still holds.
A schedule for one evening
The case method rewards discipline with time. If you have three to four hours, spend them deliberately instead of reading until you run out of energy.
| Block | Minutes | What you produce |
|---|---|---|
| Skim and question the task | 15 | One-sentence decision, list of questions the case must answer |
| Read for facts and exhibits | 50 | Fact sheet with numbers, dates, quotations, gaps |
| Frame and choose tools | 20 | Problem statement and two or three frameworks, each with a purpose |
| Analysis and calculations | 60 | Findings with so what, a small financial model, sensitivity check |
| Options and scorecard | 25 | Three options scored against weighted criteria |
| Draft the answer | 50 | Recommendation first, then reasons, implementation, risks |
| Cut and check | 20 | Remove retelling; confirm every number matches the exhibits |
Total: about four hours. If you only have two, shorten reading and drafting, but protect the framing and options blocks, because that is where the quality of the answer is decided.
A worked mini-case: opening and numbers
Imagine a case about Harbor Coffee, a regional chain deciding whether to open ten company-owned stores. A strong opening names the decision and the main reason in two sentences.
Model opening paragraph (hypothetical)
Harbor Coffee should open the ten stores, but in two waves of five, starting in the two metro areas where it already has supply routes. The first wave pays back in about three years on current economics, and the second wave should proceed only if the first meets 90 percent of its sales target after twelve months.
Then the numbers. Suppose each store costs $450,000 to open and the case implies an annual store contribution of $140,000.
| Scenario | Annual contribution per store | Payback (450,000 / contribution) |
|---|---|---|
| Base case | $140,000 | 3.2 years |
| Sales 20 percent lower, contribution $105,000 | $105,000 | 4.3 years |
| Contribution $175,000 | $175,000 | 2.6 years |
The sensitivity row matters most. If a quarter of the contribution disappears, payback stretches by more than a year, which is why the recommendation is staged. This is the pattern graders reward: a decision, a number, a test of the number and a design choice that follows from it.
Pressure-test your answer before you submit
- Could a reader disagree with your first paragraph? If not, it is a description, not a recommendation.
- Does every framework change the answer? Delete any that only decorates.
- Is there a number a manager could check? Cite the exhibit or state the assumption.
- Have you named the strongest objection? Then answered it in one or two sentences.
- Is the first action clear? Who does what in the first 30 days.
- What would make you change your mind? A trigger, such as a sales threshold or a competitor move.
Classroom cases are often discussed aloud before they are submitted. Come to class with your recommendation, two supporting numbers and one risk. Instructors notice who can defend a choice under challenge, and the written work then benefits from the discussion.
Common mistakes
- Retelling the case The reader has read it. Use only the facts that support your argument.
- Framework dumping A SWOT list with no conclusion earns little. Every tool must produce a finding that feeds the decision.
- Hedging the recommendation Do A if X, otherwise B is rarely a decision. Commit, then give a trigger for changing course.
- Ignoring implementation A good decision poorly executed fails. Cover people, timing and resources.
- Inventing facts If the case is silent, state an assumption and label it.
If you want help with a case response, you can order MBA case study help and upload the case and your instructor's questions.