How to write a case study analysis

Find the decision before you find the frameworks

Every well-written case has a protagonist and a decision point. Before analysing anything, write one sentence: who has to decide what, by when.

Students who skip this analyse the company instead of the decision, and produce a competent strategic profile that answers no question. Marks in case assessment attach to judgement, and judgement requires a decision.

Read it three times, differently

First pass: fast, no notes. Get the shape of the situation and find the decision.

Second pass: with the exhibits. The financial appendices are not decoration — in most cases they contain the fact that settles the argument, and most students never open them.

Third pass: hunting for the constraint. Cases are written so that the obvious answer is blocked by something — a covenant, a capacity limit, a contract, a regulator. Find it; it is what the case is testing.

Two frameworks, applied properly

Two applied well beat five applied shallowly, every time. Pick for the question, not for what you remember.

• PESTEL — when the decision turns on external forces the company cannot control.
• Porter's Five Forces — when the question is whether this industry is worth being in.
• VRIO — when the question is whether an advantage will last.
• SWOT — as a synthesis at the end, never as the analysis itself. A SWOT that opens an essay is a list; one that closes it is a conclusion.
• Financial ratio analysis — almost always, because almost every case decision has a number attached and most students avoid it.

The structure markers expect

Roughly, for a 2,500-word analysis.

• Executive summary (250w) — states the recommendation. Not a preview of the report; the answer.
• Problem statement (150w) — the decision, the protagonist, the deadline.
• Analysis (1,000w) — the frameworks, evidenced with figures from the case.
• Options (400w) — two or three genuine alternatives, each with its trade-off.
• Recommendation (450w) — one option, with cost, owner, timeline, and the risk that could sink it.
• Implementation and metrics (250w) — what happens in the first ninety days and how you would know it worked.

Evidence every claim from the case

"The company has cash flow problems" is an assertion. "Operating cash flow fell from £4.2m to £1.1m while inventory rose 38% (Exhibit 3), so the company is funding working capital from a shrinking cash base" is analysis.

Every substantive claim should be traceable to a figure, a quotation or an exhibit. Markers scan for this; it is the fastest way they distinguish a student who read the case from one who skimmed it.

Recommendations that survive 'so what?'

A recommendation needs four things or it is an opinion: what specifically to do, what it costs, who owns it, and what could go wrong.

Weak: "The company should invest in digital transformation."

Strong: "Migrate order management to a cloud platform over nine months at an estimated £680k capital cost, owned by the COO, funded from the deferred warehouse expansion. The main risk is the twelve-week parallel-running period over the Christmas peak, which is why the migration should start in February rather than September."

Notice that the strong version names a trade-off — the deferred warehouse — and a risk with a mitigation. Recommendations with no downside read as sales pitches.

Case method by country

Harvard-style cases centre a named protagonist facing a specific decision, and are dominant in the US and widely used elsewhere.

Ivey cases, common in Canada, are similar but often place heavier weight on implementation detail.

Australian and UK business schools frequently use live or local cases and mark regulatory feasibility more heavily — a recommendation that breaches local competition or consumer law will be penalised even if it is commercially sound.