Pyramid Principle

The problem it fixes

Most unclear documents are badly ordered, not badly worded. Minto's argument is that a reader can only take in one sentence at a time, so if your ideas arrive in the order you thought of them, the reader has to guess how they connect. Anyone who runs a business writes constantly: proposals, investor updates, briefs for a team, a note explaining why a price is changing. Each one fails in the same way when the point sits at the bottom. The book gives you a method for finding the point first and a set of rules for testing it.

What changes after you read it

You stop opening with background. You decide who the reader is and what question they have, write the answer at the top, and put the reasons directly underneath. The tests are concrete. Each idea must summarise the ideas below it, the ideas in a group must be the same kind of thing, and they must follow a logical order. When a list fails these tests, the book says your thinking is unfinished, which is better to learn before you send a document than after. You will also start to spot other people's lists that announce a number of problems and never say what the problems add up to.

The introduction is the part to copy

The book's most reusable idea is the opening story: a Situation the reader knows, a Complication that changed it, and the Question that follows, then your Answer. It reminds the reader of what they already accept and tells them what you are about to settle. It takes a few minutes to draft and I would use it for anything that asks someone to approve, choose or act. Minto also notes that most documents answer one of a few questions: what should we do, how should we do it, should we do it, or why did it happen. Naming yours first keeps the document short.

It teaches thinking as much as writing

Parts Two and Three go past formatting. Part Two shows how to word a group of actions as an end product you could hold in your hand, which exposes steps you left out. Part Three defines a problem as a gap between the result you have and the result you want, and shows how to plan the analysis before you collect data. If you hire consultants, brief an agency or hand a task to a colleague, this is a clear way to describe the work so the person doing it knows what a finished answer looks like.

Where it fits with decisions and repeatable work

A decision worth keeping has the same shape as the book's introduction: the situation, what changed, the question, the answer, then the reasons. I think a decision written that way can be understood months later by someone who was not in the room. A structure with rules is also something you can turn into a template, which is how a piece of work that one person does well starts to become a playbook other people, or software, can repeat.

When to read it, and when to skip parts

Read Part One and chapters 6 and 7 first. Part Three is written mainly for people who analyse complex problems for others, and the author says so, so skip it if that is not your job. Part Four on slides will matter only if you present often, and its examples of overhead projectors and 35 mm slides show their age. The business memos used as examples are dated and the text is dense. Do not read it for tips on sentence style: Minto says style is a lesser cause of unclear writing.

Minto argues that unclear writing is usually a thinking problem. Readers take in ideas one at a time, so they understand most easily when the summarising idea comes before the detail it covers. The ideas in any document therefore form a pyramid under a single main point. The book shows how to find that point, how to test the groups beneath it, how to use the same discipline to analyse a problem, and how to show the structure on a page or a slide.

Part One: Logic in Writing

The first part claims that sentence style is the lesser cause of unclear writing. The bigger and easier one to fix is structure, meaning the order in which ideas appear. It names the two relationships every idea has: vertical, where a point summarises the ideas under it, and horizontal, where the ideas in a group make a logical argument together.

Chapter 1: Why a Pyramid Structure?

The mind sorts information into groups to hold it, and it copes with only a handful of items at once, which Minto ties to George Miller's work on the magical number seven. A reader therefore needs the grouping stated, and stated first. As Minto puts it, "Controlling the sequence in which you present your ideas is the single most important act necessary to clear writing." You think from the bottom up but present from the top down. Three rules govern the pyramid: each idea summarises those below, ideas in a group are the same kind, and ideas in a group follow a logical order.

Chapter 2: The Substructures Within the Pyramid

An idea, in Minto's sense, is a statement that raises a question in the reader's mind, which you answer on the line below. That gives a question and answer dialogue down each branch. Horizontally, each group is either deductive or inductive, not both at once. The introduction tells a story with a Situation, a Complication and the Question it raises, and your document supplies the Answer.

Chapter 3: How to Build a Pyramid Structure

You can build from the top down by naming the subject, the reader's question and your answer, then checking them against the Situation and Complication. If you cannot do that yet, build from the bottom up: list the points, work out how they relate, and draw conclusions. The test of a finished introduction is that the reader has your whole thinking in the first thirty seconds. The chapter ends with advice for beginners: try top down first, think through the introduction before the body, and keep history in the introduction.

Chapter 4: Fine Points of Introductions

An introduction should remind the reader, not inform them, so it holds only what they would accept as true. Its length depends on what the reader needs to stand where you stand. Reordering the parts changes the tone without changing the content. Minto shows how the Key Line points, the reasons or steps under your answer, can be set out near the top, and warns against listing topics. In her words, "Never write about categories, only about ideas." She then walks through common patterns: giving direction, seeking approval to spend money, explaining how to do something, choosing between alternatives, and in consulting, letters of proposal and progress reviews.

Chapter 5: Deduction and Induction: The Difference

A deductive group is a chain in which the second point comments on the first and the third draws a conclusion. An inductive group collects ideas that are alike and draws an inference from the likeness, and you can test it by giving the group one plural noun. Minto advises inductive reasoning at the Key Line level because it is easier on the reader, and keeping deduction lower in the pyramid. She also separates thinking from news: facts that share nothing do not belong together.

Part Two: Logic in Thinking

This part says to look hard at each group once the pyramid exists. Two common faults are lists of loosely related points and a blank assertion on top of them. Minto calls the discipline Hard-Headed Thinking, and an eight-item list of complaints about a report becomes three useful points once it is applied.

Chapter 6: Imposing Logical Order

Every group comes from one of three mental acts, so its order must be time order, structural order or degree order. Time order covers steps in a process, and the chapter shows how to separate causes from effects so that the steps sit at the right level. Structural order follows a whole divided into parts that are mutually exclusive and collectively exhaustive, the MECE test. Degree order ranks items you have classified, usually strongest first. If you cannot find an order, the grouping is flawed or incomplete.

Chapter 7: Summarizing Grouped Ideas

Avoid intellectually blank assertions such as announcing three reasons without saying what they show. For a group of actions, state the effect of carrying them out, worded as an end product you can picture, and separate steps taken before from steps taken so that. For a group of situation statements, find what is similar in their subjects, verbs or objects, and make the inductive leap to the insight they imply. Doing this often shows what you have missed.

Part Three: Logic in Problem Solving

This part is for people who write reports and proposals that give the solution to a problem, such as consultants and analysts. Its aim is to define the problem and structure the analysis so the findings fall straight into a pyramid.

Chapter 8: Defining the Problem

A problem is a gap between an undesired result, R1, and a desired result, R2. The framework adds a Starting Point or Opening Scene, which is the structure or process involved, and a Disturbing Event that upset it. Where the reader already stands, whether with no solution, a proposed one or one that failed, decides the question. Read left to right, the definition converts into the introduction. It sits inside a sequence of five questions, from whether there is a problem to what should be done.

Chapter 9: Structuring the Analysis of the Problem

Minto criticises going out to collect data first, noting that one consulting firm estimated about 60 per cent of its fact-finding was wasted. She prefers diagnostic frameworks that show the physical structure, trace cause and effect, or classify possible causes, each producing yes or no questions that tell you what data you need. Logic trees then generate possible solutions and expose flaws in a list of issues.

Part Four: Logic in Presentation

Once the logic is settled, the reader or audience must be able to see it. The form depends on the length of the message and how many people will receive it.

Chapter 10: Reflecting the Pyramid on the Page

Headings, underlining, decimal numbering, indentation and dot-dash outlines can all mirror the levels of the pyramid. Minto favours hierarchical headings worded as ideas, and she says headings are for the eye and should not carry the message. She adds advice on transitions between groups, summarising long sections and ending with a conclusion that points to action.

Chapter 11: Reflecting the Pyramid on a Screen

A presentation needs more than a report cut into slides. Text slides should be short and direct, exhibit slides should state their message across the top, and a storyboard lays out the structure before you make a single slide. Her closing advice is to rehearse.

Chapter 12: Reflecting the Pyramid in Prose

Murky sentences usually come from having no clear picture of what you want to say. Minto's remedy is to visualise the nouns and how they relate, then copy that image in words.

Appendices A, B and C

Appendix A covers problem solving where no structure exists to analyse, using hypotheses and experiments. Appendix B gives worked introductions for common situations. Appendix C is the author's own outline of the key points of every chapter, useful as a checklist.

What to do with it

• Before your next important document, write down the subject, the reader's question and your answer, then check them against a Situation and Complication.
• Put the answer at the top and the reasons under it, and ask for each group whether it is one kind of idea in a logical order.
• Rewrite any heading that names a category so that it states the insight the group shows.
• Word every step in a plan as an end product, and look for the step you forgot.
• For a new problem, write R1 and R2 before you collect any data.