The answer
One sentence: what the client should do and the size of the prize.
- Comes
- First, before any reason
- Test
- Could the CEO act on it alone?
The pyramid principle is Barbara Minto's method for structuring an argument: state your main point first, then the two to four reasons that support it, then the data under each reason. You work the thinking out from the bottom up but present it from the top down. Minto built it while editing reports at McKinsey, where she worked from 1963 to 1973, and Minto's own site calls it the de facto standard at major consulting firms. Soreno, an AI voice interview practice app for consulting, investment banking and other finance candidates, runs a free profitability case with no sign-up that ends on your recommendation to the CEO, and full mock cases with scored feedback.
Updated
Minto's own site puts it in one line: your thinking is easy to grasp if you present the ideas organized as a pyramid under a single point. The single point sits on top. Under it sits a small group of ideas that together prove it, and under each of those, the facts. A reader, or an interviewer, who stops after the first sentence still has your answer.
One sentence: what the client should do and the size of the prize.
Two to four grouped ideas that together prove the answer.
The numbers and facts under each reason.
Alder & Finch, an invented chain of 120 home goods stores, saw operating profit fall this year. You have done the analysis and the CEO asks for your answer. Alder & Finch and its numbers are made up by us for practice. Watch the thinking go up the pyramid, then the answer come down it.
Collect the facts
Profit fell from $60 million to $36 million, a drop of $24 million or 40%. 18 of the 120 stores lose about $800,000 each a year. Shipping on the 1,800,000 online orders that ship free now costs $9 an order, $16.2 million a year. Online rivals price about 3% below the stores.
Group and size
Each finding becomes one reason, and each gets a number. Closing the losing stores saves 18 times $800,000, $14.4 million. A $7 fee on the 33% of free orders under a $50 threshold brings in 594,000 times $7, $4.2 million. Matching rivals' prices would cost 3% of $900 million in store sales, $27 million.
Draw the point
Add the two moves: $14.4 million plus $4.2 million is $18.6 million, 77% of the drop, taking profit to about $54.6 million. Matching prices would cost 1.1 times the whole drop, so it is a reason not to act. Now the top of the pyramid writes itself.
Open with SCQA
Situation: "Alder & Finch made $60 million last year." Complication: "This year profit fell to $36 million." Question: "How do we win it back?" Answer: the top line. In a case the interviewer knows the situation, so keep the first three to one sentence.
Weak close (ours)
"So first we looked at revenue, and store sales were down, and then we looked at costs and shipping had gone up a lot, and some stores are losing money, and the online players are cheaper, so I think there are a few things they could do, maybe close some stores, and look at shipping, and pricing is a question too." Every fact is right. The CEO still has to dig for the answer.
Strong close (ours)
"Close the 18 stores that lose money and charge a shipping fee on small online orders. Together that wins back $18.6 million of the $24 million drop, without starting a price war. Three reasons. First, 18 stores lose about $800,000 each, so closing them saves $14.4 million. Second, shipping a free order now costs $9, and a $7 fee under $50 brings in $4.2 million. Third, matching online prices would cost $27 million, more than the whole drop, so we hold price. The risk is losing some shoppers from closed stores to rivals, so next I would check how many of their customers live near another of our stores."
Minto told McKinsey's alumni team that the pyramid meant there were only three logical rules to obey.
Rule 1
Minto's words: the point above has to be a summary of those below, because it is derived from them. If your top line says something the reasons don't prove, one of them is missing.
Rule 2
You can't derive an idea from a grouping unless its ideas are logically the same. Three reasons are fine. Two reasons and a risk in the same list are not.
Rule 3
In our experience three orders work out loud: by size (biggest first), by time (first, next, then) and by structure (the branches of your tree, in the order you drew them).
MECE
McKinsey's profile says the concept required groups of ideas to be MECE, mutually exclusive and collectively exhaustive. Minto told the firm the idea goes back to Aristotle, but she was the first to abbreviate it and apply it to groups of ideas.
MECE, with examplesFrom McKinsey's alumni profile of Barbara Minto and her own site, barbaraminto.com.
1963
McKinsey calls Minto the first female MBA professional hire the firm ever made. She graduated from Harvard Business School in 1963 and was with the firm from 1963 to 1973, first in Cleveland, then in London.
The idea
Editing the reports that crossed her desk, Minto noticed she kept reorganizing ideas into a pyramid shape. She saw the problem as the thinking, not the language: people were starting to write before they had worked out their thinking.
The manual
She turned the idea into a course taught at McKinsey offices around the world. Her first manual had that title, and colleagues in every office taught it to newcomers.
1985 and 1996
After leaving McKinsey in 1973 she taught the course independently, with Accenture and Bain among the hosts. McKinsey says she published The Pyramid Principle: Logic in Writing and Thinking in 1985 and that Pearson mass-produced it from 1987. That version is the one most sold online, but McKinsey says it is not the authoritative edition: her own 1996 revamp is The Minto Pyramid Principle: Logic in Writing, Thinking and Problem Solving.
Today
McKinsey's profile says its lessons are still taught at the firm today. Minto's site calls the pyramid principle the de facto standard for all major consulting firms.
Minto's site names the Situation, Complication, Question framework, SCQ, as the way to find the question in your reader's mind. Monash University's writing guide lists it with the answer added, as SCQA.
Situation
Monash's guide calls it your foundation: start with facts your audience already knows and accepts.
Complication
The problem that makes the situation worth talking about. In a case, it is usually the prompt itself.
Question
Often unspoken. If you can't name it, your answer may answer a different question.
Answer
Your main point. Everything after it is a reason or a fact supporting it.
From McKinsey alumni: Barbara Minto, MECE: I invented it, Barbara Minto: the pyramid principle concept, Barbara Minto: course, textbook and SCQ framework, Monash University: business paper using the Minto approach and Monash University: structuring with SCQA.
In our experience it matters most at three moments: the end of the case, the end of each analysis and every time the interviewer asks so what. Here is what we listen for.
Do this, worth that. If the recommendation arrives after a minute of recap, it was too late.
"Three reasons" tells the listener how long to keep listening and makes a missing reason easy to spot.
Each reason carries the figure that proves it. A reason with no number reads as an opinion.
The same rule works mid-case: say what the exhibit means, then the numbers that show it.
After the reasons, one risk and what you would check next. That shows the answer is a decision, not a guess.
Each one ends on a spoken answer that should come out top down. Under each is what a strong answer does.
The answer, the reasons, the risk.
You have finished the analysis on a market entry case and the CEO walks in. What is your recommendation?
Opens with go or no go and the size of the prize, gives two or three numbered reasons, then one risk and one next step. Practice it, see a model answer
A grocery chain's profits fell 20% in a year. What happened, and what should it do?
After the profit tree, closes with the cause and the fix in one sentence, then the numbers behind each. Practice it, see a model answer
A retailer needs to cut costs by 10%. Where would you look?
Groups the cuts by cost bucket, sums them to show they reach the target and leads with the total. Practice it, see a model answer
An airline is flying more passengers but making less money. Why?
States the cause first, such as cheaper fares or a costlier mix of routes, then the data that rules out the alternatives. Practice it, see a model answer
Small pyramids after each piece of analysis.
A streaming service's churn has doubled. What would you look at?
After each piece of data, says what it means for churn before reading out the numbers. Practice it, see a model answer
A software company is growing fast but losing money. Should it worry?
Answers yes or no up front, then groups the reasons into unit economics and fixed costs. Practice it, see a model answer
The same rule works outside the case.
Why consulting?
Gives the reason in one line, then two or three short proofs from the candidate's own experience. Practice it, see a model answer
All of these come from practice cases we have heard.
"We looked at revenue and costs" is a table of contents, not a point. The top line must say what to do.
Two reasons and a risk in the same list of three break Minto's second rule. Keep reasons with reasons, and say the risk after.
Six reasons is a list, not a group. Merge them under two to four headings.
If the top line promises $18 million and the reasons sum to $12 million, the interviewer will notice. Check the sum before you speak.
Early in a case you rarely have the answer. Lead with your structure or your hypothesis instead, and save the full pyramid for the close.
Soreno's AI interviewer runs a full case, asks for your recommendation at the end and scores whether you led with it. Start your free week.
A way to structure writing and speaking with the main point first, a small group of supporting reasons under it, and the evidence under each reason. Barbara Minto developed it at McKinsey and published it as a book in 1985.
A Harvard Business School graduate who joined McKinsey in 1963 as its first female MBA professional hire. She developed the pyramid principle while editing consultants' reports, left in 1973 and taught it independently to consulting firms and companies.
Minto's site describes building your thinking by grouping similar facts, drawing an insight from each group and grouping the insights, until one point sits on top. Presenting is the reverse: you guide the reader down from that point.
Situation, Complication, Question, Answer: a short opening that sets up the main point. Minto's site calls the first three the SCQ framework, and many writing guides, Monash University's among them, add the answer as a fourth part.
No. The pyramid is the shape of the whole argument. MECE is the test each group of ideas inside it must pass: no overlap and nothing missing. McKinsey's profile says Minto was the first to abbreviate MECE and apply it to groups of ideas.
Monash University's guide says two to four grouped arguments under the main recommendation. In our experience three is the most common in a case close, since it is easy to say and easy to remember.
McKinsey's profile says the widely sold Pearson version is not the authoritative edition, and Minto's own site says the 1996 edition, The Minto Pyramid Principle: Logic in Writing, Thinking and Problem Solving, supersedes it. In our view, for a case interview the early chapters on building the pyramid are enough.
Yes. Answer "why consulting" with your reason in one line, then the two or three experiences that prove it. It reads as clear thinking, which is what the fit round is testing too.