01
Enter before the outcome
Work only with the facts available when the company faced the decision.
How to play a Case.
How to win?
Make a decision before the outcome is revealed. Defend it with evidence, compare credible alternatives, and show what would change your mind.
Outcome hidden until you commitFeedback is advice, not a gate
Sign the rookie or spread the risk?
Case 43 · Nike · decision point
Back five proven athletes
Make one concentrated bet
Delay until the next season
Your memo
Make the bet only if the downside is capped and the next campaign can test whether attention converts into repeat demand.
Five-dimension reviewSee the gaps →
Outcome still hidden
3moves in every case
5rating dimensions
20points per dimension
0answer gates
A case is a decision, not a reading test.
You are not trying to guess the historical answer. You are building a call that could survive a serious challenge with the information available at the time.
02
Make the call
Play the decision board or write the memo. A blank answer never blocks the case.
03
Review the reasoning
Compare with history and use the five-dimension review to improve the next draft.
Five dimensions. Twenty points each.
The reviewer sees the case, the question, your memo, and your latest board run. It scores the reasoning visible there, not the probability that the decision succeeds in the real world.
Framing
Frame the decision
Name the real decision, the objective, the boundary, and the time horizon.
0–20Evidence
Use the evidence
Use decision-relevant facts from the case and separate them from assumptions.
0–20Alternatives
Compare alternatives
Test genuinely different options and name the trade-off, stakeholders, and opportunity cost.
0–20Uncertainty
Price uncertainty
Name what could break, what is reversible, and the evidence that would change your mind.
0–20Commitment
Commit and learn
Choose, answer the strongest opposition, and name the next move or learning loop.
0–20Six moves make a memo harder to break.
Use them as a pre-flight check, not a formula. A phrase on the page earns nothing unless the reasoning behind it is real.
20 words is the review floor · length itself earns nothing
- 01Framing
Make the question decidable
Name the call, the objective, the horizon, and what success means.
- 02Evidence
Use a fact that matters
Cite the number, date, or event that actually changes the recommendation.
Put a real alternative on the table
A different date is not a different strategy. Compare another credible future.
Name the price
Say what you give up, who carries the cost, and what you get in exchange.
Expose the failure condition
Name the assumption, downside, or signal that would make you change course.
Commit to the next move
Choose, answer the strongest objection, and specify the next test or action.
Every dimension uses the same scale.
A score is a position on a published scale, not a reaction to effort or writing style. Seventeen to twenty means the reasoning is unusually resilient to a strong challenge.
0–4absentAbsent, off-topic, or contradicted by the case.
5–8assertedAsserted but largely unsupported.
9–12partialUseful partial reasoning with a material gap.
13–16groundedClear, specific, and grounded in the case.
17–20resilientUnusually rigorous and resilient to a strong challenge.
What is never scored
The case is evidence, not an answer key.
- Matching the choice the company eventually made
- Writing a longer or more polished answer
- Naming a framework without using it
- Agreeing with the case author or AI reviewer
- The board result on its own
Make the Nike supplier decision.
Choose before you see Phil Knight’s call, defend the recommendation, and read the five-dimension review.
Play the Nike case