Overview
1. Choose one firm
Not every firm on this list is dead. MGM restructured and survived. Sweetgreen is still trading. The question does not change: name the advantage, and name what broke first. A firm that is still open can have lost its advantage years ago. |
Sweetgreen carries a useful trap. A mid-2026 food-safety scare hit demand, and it is tempting to stop there. Same-store sales had already fallen for five straight quarters before it. A shock that lands on an advantage already gone is a symptom, not a driver.
2. The tools, and what each one answers
Tool | What it answers in your memo |
Competitive Advantage Timeline | When, and in what order, the advantage changed. Turning points, not a company history. If an entry did not change the advantage, cut it. |
VRIO (Barney) | What the advantage actually was, and which of the four conditions stopped being met first. |
Barriers to Imitation | Why competitors could not copy it, and what changed so that they could. |
Causal Chain | How that first break became the final outcome. This is the spine of the memo. |
Strategy (what, where, how, when), Strategic Fit and Value Creation are the vocabulary for naming which test broke. They are not separate sections.
3. What the memo contains
Short prose with three exhibits set into it. The exhibit carries the sequence, so the paragraph beside it does not repeat it. That is how the whole argument fits on one page. References go on page two and do not count against it.
Part | Length | What goes in it |
Executive Summary | 3 sentences | The firm, the advantage it had, which condition broke first and the decision behind it, plus a one-line preview of the alternative. Not an introduction to the company: your reader already knows it. The first sentence is the answer, not background. |
Exhibit A. Advantage Timeline | 5 boxes | Five turning points across the page, each tagged builds, holds, breaks, symptom or outcome. Without the tags it is a company history. |
The advantage, and why rivals never took it | 2 to 3 sentences | Name the barrier. Naming a weaker candidate you rejected is worth more than another adjective. |
Exhibit B. VRIO | 4 lines | V, R, I, O with a yes or no and a few words of evidence. Highlight only the condition that broke. Add the barrier underneath in one sentence. |
What broke first | 3 to 4 sentences | The decision, sourced. Name the driver and separate it from the symptom everyone else names. Do not narrate the chain: Exhibit C does that. |
Exhibit C. Causal Chain | 5 rows | Decision, capability, customer value or cost, competitor response, performance. Each row carries a reason, not a label. |
What they should have done, and what it would have cost | 3 to 4 sentences | Your alternative at a named point on the timeline. See the three rules below. |
What would make me wrong | 2 sentences | The evidence that would change your conclusion, and the competing hypothesis it would support. |
References (page two) | At least two credible sources: filings, industry reports, trade press, or peer-reviewed research. APA. |
The rule that keeps the exhibits from becoming templates: Nothing goes in an exhibit that the memo does not use, and nothing is said twice. A turning point you never argue from should be cut, and a paragraph that walks through its own exhibit row by row should be deleted and rewritten. Tidy exhibits under an argument that does not follow from them score near zero on the first criterion below. |
4. The three rules for “what should they have done”
Not an afterthought. It is how I check whether your diagnosis is real: if your chain says one thing and your fix addresses another, one of the two is wrong.
1. It must attack the link you named as the driver.
If your chain says the lease structure was the driver and your fix is a menu refresh, the two do not connect. Fix what you said was broken.
2. It must have been available at the time.
Judge the decision on what was knowable then, not on what we know now. A 2014 firm cannot be faulted for not using tools that did not exist.
3. It must have a price.
Money, growth given up, a customer segment abandoned, a partner angered, a slower rollout. Name it. A fix with no cost is not a strategy, it is a wish, and it is the most common way this section fails.
5. How it is graded
Criterion | Weight | What I am looking for |
Analysis using the assigned tools | 35% | Whether the argument was actually built with the four tools, which the exhibits let me check at a glance |
Judgment and recommendation | 30% | A defended claim about what broke first, a costed alternative, and what would make you wrong |
Quality and use of evidence | 20% | Sources you actually read, cited where the claim needs them |
Clarity, structure and professional format | 15% | One page, the answer at the top, exhibits legible without explanation |
This memo is not: A company history. Causes with no ranking among them. Exhibits the prose repeats in full. A VRIO table submitted as the answer. Single-spaced, 10-point, 0.7-inch margins, APA. Every claim must come from a document you actually read. |