Barbara Minto’s The Pyramid Principle: Logic in Writing and Thinking is the book behind a way of structuring documents that consulting firms have taught for decades: the answer first, the supporting points grouped beneath it, and every level of the argument summarising the level below. Minto joined McKinsey in 1963 as its first female MBA professional hire, spent much of her time there editing other people’s reports, and built a method out of what she found herself doing to their drafts. Its central claim is that most unclear business writing is a failure of thinking rather than of style: the writer has not decided what the document says, or has decided but presents the ideas in the order they occurred to the writer rather than the order in which a reader can absorb them. For an internal auditor the book is directly useful. An audit report is an argument with a conclusion; an audit committee paper is read by people with little time; and the Global Internal Audit Standards give communication a principle of its own and require an engagement conclusion that judges the findings taken together. This guide summarises the book without reproducing it, explains its rules and its pattern for introductions, maps both onto audit reports and audit committee communication, and ends with a worked example of a UK food manufacturer’s audit function rewriting its annual report to the audit committee ahead of the board’s first Provision 29 declaration.
In this guide
- The book in one page
- Who Barbara Minto is, and how the book came about
- The structure: two parts, ten chapters and an appendix
- The pyramid and its three rules
- Introductions: situation, complication, question, answer
- Questioning the thinking: order, summaries and the problem
- The nine ideas worth keeping
- The vocabulary: fourteen terms defined for practitioners
- What it means for internal audit: reports and the audit committee
- The criticisms, and what the book leaves out
- Editions, reading time and how to read it
- A one-hour discussion guide for an audit team
- Worked example: Pennine Foods rewrites its annual report to the audit committee
- Frequently asked questions
- Related guides
The book in one page
| Question | The answer, in our words |
|---|---|
| What is the book about? | How to decide what a document says and arrange it so that a reader grasps it at once: the main point at the top, the supporting ideas grouped beneath it, and each group summarised by the idea above it. The publisher describes its starting point as the observation that the mind sorts information into pyramid-shaped groupings of its own accord. |
| What is the central claim? | That unclear writing is mostly unclear thinking. Sentence style matters, but structure is the more widespread problem and the easier one to fix, and it has to be fixed before drafting rather than after. |
| What are the rules? | Three, in Minto’s own summary: an idea at any level must summarise the ideas grouped below it; the ideas in a group must be the same kind of idea; and they must be in a logical order. |
| How should a document open? | With a short story the reader already accepts (the situation), the change that creates the need to write (the complication), the question that change raises, and the answer, which is the document’s main point. |
| How is the thinking tested? | By questioning each grouping: are its ideas related deductively or inductively, is it in time, structural or ranking order, and does the statement above it say what the group implies rather than naming what kind of thing it contains? |
| Where does it come from? | Minto’s years editing McKinsey reports from 1963, first in Cleveland and then in London, and the courses she built from that work after she left the firm in 1973. |
| Why read it as an auditor? | Because an audit report is an argument with a conclusion, because audit committees read under time pressure, and because the book gives a reviewer objective tests for structure (does the summary say what the findings imply, do the groups overlap?) in place of matters of taste. |
| Reading time | A short book that repays slow reading; allow two or three sittings, and test each part against a report the team has written. |
Who Barbara Minto is, and how the book came about
Minto graduated from Harvard Business School in 1963, the year the school had only just begun granting MBAs to women, and joined McKinsey the same year in its Cleveland office, the firm’s first female MBA professional hire. A McKinsey alumni profile records that the analytical side of consulting did not come to her easily but her writing did: she took over editing the office’s reports, and the firm began using her across its offices to improve associates’ report writing. In 1966 she moved to London. Editing reports, and finding the same problems in offices in several countries, she noticed that she kept rearranging the ideas into the same shape, a single point at the top supported by groups of points below, and she set out to understand why, reading widely, from Piaget’s psychology to the Bourbaki mathematicians. The result became a course taught in McKinsey offices, with a first manual titled Skillful Writing through Structured Thinking. She also claims a coinage now common in consulting: she says she was the first to abbreviate the idea of mutually exclusive, collectively exhaustive groups as MECE and apply it to grouping ideas, though she credits Aristotle with the idea itself. After the 1973 oil crisis brought staff reductions in London she was among those let go, and she set up her own business, teaching the method to consulting firms and corporate clients in small groups. Her company’s website still offers a two-day course conducted by her, alongside self-study and online versions.
The book’s publishing history is tangled. Versions circulated under Minto’s own imprint before a trade publisher took it on: the Internet Archive holds a 1981 Minto International edition. The text that Pearson sells today under its Financial Times imprint has the same ten chapters and appendix as that 1981 edition, which library catalogues already describe as a revised third edition; Minto’s site dates the Pearson text to 1987, Financial Times Prentice Hall reissued it as the third edition in 2002, and Pearson reissued it in 2021 and again as a paperback in 2026. Minto herself regards that text as superseded. Her site states that the Pearson book is the 1987 text reissued unchanged in 2002, and that her 1996 book, The Minto Pyramid Principle: Logic in Writing, Thinking and Problem Solving, published by her own company with twelve chapters and three appendices in four parts, replaces it. The McKinsey profile adds that Pearson never acquired the rights to the 1996 revision, which is why the version most readers buy is not the one its author considers authoritative. This guide summarises the Pearson edition, the one in bookshops and libraries, and says so where the 1996 book differs in scope.
The structure: two parts, ten chapters and an appendix
The Pearson edition has two parts and a short appendix. The first part, on logic in writing, explains the pyramid, how to build one, how to open a document and how to make the structure visible; the second, on logic in thinking, turns the pyramid into a set of questions for testing whether the ideas are right before they are written up. The table gives the chapter titles as they appear in the edition’s contents, what each title promises, and the guide’s view of why each matters to an auditor. The sections that follow arrange the book’s main ideas by theme rather than chapter by chapter.
| Part and chapter, as listed | What the title promises | Why an auditor should read it (our view) |
|---|---|---|
| Part 1, Logic in writing. 1. Why a pyramid structure | The case for presenting ideas from the top down | The argument to put to a team that writes reports in the order the fieldwork was done |
| 2. The substructures within the pyramid | The components inside the pyramid and how they fit together | The vocabulary a reviewer needs to say precisely what is wrong with a draft |
| 3. How to build a pyramid structure | A method for constructing a pyramid from a mass of points | The step audit teams skip: deciding the point before drafting |
| 4. Fine points of introductions | How to open a document | Where most audit reports lose the reader: the first paragraphs |
| 5. Deduction and induction: the difference | The two ways ideas in a group can relate to one another | Why findings strung together as one long argument read worse than findings grouped under a conclusion |
| 6. How to highlight the structure | Making the pyramid visible on the page | Headings, numbering and summaries in reports and committee papers |
| Part 2, Logic in thinking. 7. Questioning the order of a grouping | Testing whether the ideas in a group are in a defensible order | Ordering findings and recommendations |
| 8. Questioning the problem-solving process | Testing whether the analysis behind a document addresses the right problem | Engagement objectives and scope: whether the audit answered the question it was set |
| 9. Questioning the summary statement | Testing whether the point at the top of a group says what the group implies | The engagement conclusion and the executive summary |
| 10. Putting it into readable words | Turning the tested structure into prose | Drafting comes last, not first |
| Appendix: Problem solving in structureless situations | Approaching problems that have no obvious framework | Advisory work and investigations where the question is not yet clear |
The pyramid and its three rules
The book’s starting point, as its publisher puts it, is that the mind sorts information into pyramid-shaped groupings whether or not the writer helps it to. The practical consequence, in our reading, is that a reader given a list will try to work out what the items have in common, while a reader given the point first reads the support knowing what it is for. Minto’s answer is a hierarchy in which a single idea at the top is supported by a small group of ideas below it, each of which is supported in turn by a group below that, and in which three rules hold at every level: an idea must summarise the ideas grouped beneath it, the ideas in any group must be the same kind of idea, and they must be in a logical order. A writer thinks from the bottom up, gathering facts into groups and asking what each group means, but presents from the top down, so that the reader meets the conclusion first and the evidence in the order in which it supports the conclusion. The book’s first chapter draws on the ‘magical number seven’ to explain why ideas have to be grouped; for an audit report, in our view, three or four ideas to a group is usually enough.
The relationship between the levels is a dialogue. Each statement raises a question in the reader’s mind (why, how, which ones, so what), and the level below answers it. The writer’s discipline is not to raise a question the document does not answer and not to answer a question the reader has not asked, which rules out much of what audit reports contain: methodology paragraphs nobody requested, sample sizes in the executive summary, background that the reader already knows. The relationship within a level is either deductive or inductive, and the book gives a chapter to telling the two apart. The table turns the rules into tests a report reviewer can apply, with the related test of overlap that Minto calls MECE. The guide to reports people actually read describes the same failures from the reader’s side.
| Rule | What it means | The audit report failure it prevents | The reviewer’s test |
|---|---|---|---|
| The idea above summarises the ideas below | Every heading or lead sentence states what the points beneath it add up to | An executive summary that lists findings without saying what they mean together | Cover the detail and read only the summary: does it state a conclusion that the detail supports? |
| Ideas in a group are the same kind | A group holds causes, or actions, or findings about one control, not a mixture | A section that puts a control failure, a process inefficiency and a data quality note under one heading | Name the group with a plural noun (three control failures, four causes); anything that does not fit belongs elsewhere |
| Ideas in a group are in a logical order | The order is by time, by structure or by rank, and the writer can say which | Findings ordered as the fieldwork happened rather than by severity or by process step | Ask the author why the second point is second; if the answer is that it was found second, reorder |
| Groups neither overlap nor leave gaps (MECE) | Mutually exclusive, collectively exhaustive groupings, an abbreviation Minto says she was first to apply to grouping ideas | Two findings that describe one root cause from different angles; a scope area with no conclusion | Map each piece of evidence to one point only; check that every scope area has a conclusion |
Introductions: situation, complication, question, answer
The book’s best-known device is its pattern for an introduction, which Minto’s course materials call the SCQ framework. Before the reader is given the answer, the writer reminds them of a situation they already know and accept. The writer then introduces the complication: the change in that situation that creates the reason for writing, which need not be a problem: it is the change that gives the reader a reason to read. The complication raises a question in the reader’s mind, and the document’s main point is the answer to it. The question is often left implicit, but the writer should be able to state it. Two further disciplines follow. The introduction should contain only what the reader will accept without argument; anything that needs proof belongs in the body. And once the answer has been given, the key line, the row of ideas directly beneath it, sets out the plan of the document as a set of statements rather than a list of topics.
The pattern is short to describe and hard to apply, and in our view it changes an audit report more than anything else in the book. Most reports open with the situation and stay there: the purpose of the engagement, the standards followed, the scope, the methodology, the period, the sample. The complication, which is what the audit found, appears pages later, and the answer, which is the engagement conclusion, sometimes not at all. Rewriting the first paragraph as situation, complication, question and answer usually shortens the report and moves the conclusion to where a busy reader will see it. The executive summary guide and the anatomy of an audit report give the site’s templates; the table shows the pattern applied to an engagement report.
| Element | What it holds | In an engagement report | The common failure |
|---|---|---|---|
| Situation | What the reader already knows and accepts | The engagement’s scope and why it was in the plan, in a sentence or two | Three paragraphs of background and methodology before any conclusion |
| Complication | The change that creates the reason for writing | What the audit found that the reader did not know: the controls work, or they do not, or the risk has moved | Burying the complication in the findings section |
| Question | What the complication makes the reader ask | Usually some version of: can we rely on this process, and if not, what must change? | Answering a question nobody asked, such as how many samples were tested |
| Answer | The document’s main point | The engagement conclusion, stated as a judgment | An answer that is a category (“four findings were identified”) rather than a judgment |
| Key line | The ideas that support the answer, stated as ideas | Three or four statements explaining the conclusion, each heading a group of findings | Headings such as “Findings”, “Observations” and “Other matters”, which say nothing |
The fit with the Global Internal Audit Standards is close. Standard 14.5 (Engagement Conclusions) requires a conclusion, and the glossary defines it as the auditors’ judgment about the findings viewed collectively, indicating satisfactory or unsatisfactory performance: exactly what Minto requires of the idea at the top of a pyramid, a statement of what the group implies rather than a list of its members. Standard 15.1 (Final Engagement Communication) sets what the final communication must contain, including the people responsible for addressing the findings and the planned completion dates, and it requires the chief audit executive to review and approve it. The Standards say what must be in the report; the pyramid decides where each element goes, under the point it supports. The Domain V guide lists the engagement requirements in full, and the map of the Standards places them in the whole framework.
Questioning the thinking: order, summaries and the problem
The book’s second part, with the chapter on deduction and induction that precedes it, is for an auditor the more useful half, because it treats the pyramid as a test of the thinking rather than a template for the page. The first test is order. The ideas in a group can be related deductively, as an argument in which each point follows from the one before, or inductively, as a set of like ideas from which a conclusion is drawn. Deduction works at the lowest levels of a pyramid, where the steps are short; at the top, where the reader would have to hold the whole chain before reaching the point, it becomes laborious, and one detailed review records Minto’s advice to avoid it at the key line. Induction is harder to do well, because the writer must define precisely what kind of idea the group contains and spot the item that does not belong. An inductive group has one of three natural orders: time, meaning a sequence of steps or of cause and effect; structure, meaning the parts of a whole, which must be mutually exclusive and collectively exhaustive; or ranking, meaning degree of importance. The practical test, in our reading, is whether the writer can say which order a group is in; a writer who cannot has usually not finished thinking about it.
| Order | What it reflects | Example in an audit report | Typical misuse |
|---|---|---|---|
| Deductive | An argument in which each point follows from the one before | A short chain inside one finding: the policy requires a second approval; the payments were released without it; the control is not operating | A whole report written as one argument, with the conclusion on page twelve |
| Time | A sequence of steps, or of causes and effects | Findings on a process ordered from initiation to settlement; recommendations ordered by when they can be done | Ordering findings by the date they were found |
| Structure | The parts of a whole, mutually exclusive and collectively exhaustive | Findings grouped by site, by legal entity, or by the components of a control framework | Overlapping groups, such as “controls” and “IT”, that split one problem in two |
| Ranking | Degree of importance | Findings in descending order of severity under the function’s rating scale | A ranking with no stated basis, or one that the ratings contradict |
The second test is the summary statement. The idea at the top of a group must say what the group implies, not what kind of thing it contains: a heading announcing that there are three problems fails, because it names a category, while a statement of what the three problems mean together passes. For a group of actions, the summary states the effect the actions will achieve, as specifically and measurably as possible so that success can be judged; actions are grouped by the effect they produce, never merely because they look alike. For a group of findings or facts, the summary states the conclusion they support. Audit reports fail this test constantly, at the top (the engagement conclusion that reports the number of findings) and in the middle (the heading “Access management” over four findings that together show the joiner-mover-leaver process does not work). The finding makeovers show the repair on real-shaped findings, and the recommendations guide applies the effect rule to recommendations.
The third test is the problem itself. Before structuring the answer, the writer must define the reader’s problem: what the situation is, how the difficulty arose and what would resolve it. In our words, a well-structured answer to the wrong question is worse than a poorly structured answer to the right one. One review notes the book’s suggestion of framing the key issues as questions that can be answered yes or no, which suits engagement objectives well: does the reconciliation control detect a misstatement of the size that matters, yes or no? The book also recommends conceptual images, which compress a complex set of ideas into one the reader can see. For an audit function the upstream version of this discipline sits in planning, where the engagement scoping memo and the planning memo template fix the question the engagement will answer before the fieldwork starts.
The nine ideas worth keeping
- Most unclear writing is unclear thinking. Fix the structure before the sentences; the sentences are the smaller problem.
- Decide the point before drafting. A writer who cannot state the main point of a report in one sentence is not ready to write it.
- Think from the bottom up, present from the top down. Gather and group the evidence from the bottom; give the reader the conclusion first.
- Every summary must say what its group implies. “Four findings were identified” is a category, not a conclusion; the engagement conclusion is a judgment.
- Groups must be of one kind and in a stated order. Time, structure or ranking; if the writer cannot name the order, the reader cannot follow it.
- MECE is a test, not a slogan. Groups that overlap count a problem twice; groups that leave gaps hide one.
- Open with what the reader knows, then what has changed. Situation, complication, question, answer: the reader should know the point by the end of the first paragraph.
- Headings should carry ideas. A heading that says “Findings” tells the reader nothing that a heading stating the finding would not tell better.
- Define the problem before structuring the answer. A well-built answer to the wrong question is a failure of scoping, not of writing.
The vocabulary: fourteen terms defined for practitioners
| Term | What Minto means | What it looks like in an audit |
|---|---|---|
| Pyramid | A hierarchy of ideas with one point at the top and groups of supporting points below | The report as an argument: conclusion, supporting statements, findings, evidence |
| Top-down presentation | Giving the reader the main point first | The engagement conclusion in the first paragraph |
| Bottom-up thinking | Grouping facts and asking what each group means | Moving from test results to findings to a conclusion |
| Vertical logic | Each idea raises a question that the level below answers | Every statement in the summary answered by the section beneath it |
| Horizontal logic | How the ideas within one group relate: deductively or inductively | How findings sit beside one another under a heading |
| Deductive grouping | A chain in which each point follows from the one before | Criteria, condition and conclusion inside one finding |
| Inductive grouping | Like ideas from which a conclusion is drawn | Several exceptions that together show a control does not operate |
| Situation | What the reader already knows and accepts | Scope and context, in a sentence or two |
| Complication | The change that creates the reason to write | What the audit found that the reader did not know |
| Key line | The ideas directly beneath the main point, stated as ideas | The three or four statements that explain the conclusion |
| MECE | Mutually exclusive, collectively exhaustive groupings | Findings that neither overlap nor leave a scope area unaddressed |
| Logical order | Time, structure or ranking within an inductive group | Process sequence, organisational structure or severity |
| Summary statement | The idea that states what a group implies | The heading over each group of findings; the engagement conclusion |
| Plural-noun check (our label) | Naming a group with a plural noun to check that its members are alike | “Three control design gaps”: anything that is not a design gap moves out |
What it means for internal audit: reports and the audit committee
An audit function produces three kinds of document that the book improves: engagement reports, read by management and summarised for the committee; papers for the audit committee, read by non-executives with a stack of other papers; and the chief audit executive’s periodic and annual reporting, which turns a year of engagements into a view of the organisation. The map below takes each audit activity in turn. The step-by-step report guide gives the end-to-end process, the guide to writing for the board covers altitude and length, and the guide to reporting to the audit committee covers the quarterly pack.
| Audit activity | What the book implies | Procedure it changes |
|---|---|---|
| Drafting the engagement report | Decide the conclusion and the key line before drafting | Require a one-page pyramid (conclusion, key line, groups of findings) at the end of fieldwork, reviewed by the manager before anyone drafts |
| Writing the executive summary | The answer first, in the reader’s terms | The opening paragraph states the engagement conclusion and what it means for the reader’s decision, before any scope or method |
| Writing findings | Each finding is a small pyramid: the point, then the support | Lead with the significance of the condition, then criteria, cause and effect; see the guides on conditions, cause statements and impact |
| Writing recommendations | Actions grouped by the effect they achieve, stated measurably | Group recommendations by the risk they reduce and state the result each will achieve |
| Quality review of reports | Structure can be tested objectively | Add three questions to the review checklist: does each heading state an idea, is each group one kind in a stated order, does the summary say what the findings imply? |
| Audit committee papers | Readers have little time; the key line is the agenda | Open each paper with the question the committee must answer and the chief audit executive’s answer, then the key line as the paper’s agenda |
| Periodic and annual reporting | A conclusion about the organisation is the top of a pyramid built from engagements | Build the annual conclusion from themes across engagements, grouped and ranked, rather than a list of audits in date order |
| Scoping and planning | A well-structured answer cannot rescue the wrong question | State the question the engagement will answer in the planning memo, and check the report answers that question |
Two practical points sit above the table. The first is that the pyramid helps the reviewer as much as the writer. Review notes on reports are often matters of taste, and writers resent them; the book’s tests are not matters of taste. A reviewer who writes that a heading names a category rather than stating an idea, or that a group mixes causes with effects, has made an observation the writer can check, and the guide to review notes describes how to give them. The second is that ratings and the pyramid have to agree. A report whose ranking order puts a moderate finding above a high one, or whose conclusion is satisfactory while a high-rated finding sits beneath it, fails Minto’s rules and the function’s own severity rating scale at the same time. Ratings lie outside the book’s subject, so an audit function has to make the two consistent itself.
The book also bears on a newer problem. Drafting tools that generate fluent prose from fieldwork notes can produce text with no pyramid at all, in roughly the order the notes were written, and the fluency hides the absence. The tests in this guide apply to machine drafts exactly as they apply to human ones; the guide to writing reports with AI assistance covers the workflow and its guardrails.
The criticisms, and what the book leaves out
Three criticisms are worth knowing. The first is about the reading experience: the book is dense and its prose is dry; one consulting-frameworks site that recommends the method calls the book itself a bit of a boring read. The second is about rigidity. One practitioner site that teaches the method warns against forcing every document into a pyramid: a session that shares interim facts before the team has formed a view, or a set of slides meant to be read without a presenter, may need a different shape, though even these benefit from a one-page summary at the front. The third is about editions, and it is the author’s own: Minto regards the Pearson text as superseded by her 1996 book, which adds problem solving to the title and is available mainly from her own company, so the reader of the bookshop edition is reading the version its author would replace.
For an auditor the larger limits come from the book’s subject, and we state them as our reading rather than as anyone’s published criticism. Its subject is the structure of an argument, so three things an auditor needs lie outside it. The first is evidence: a well-structured report can rest on a conclusion the testing does not support, and the pyramid will make the unsupported conclusion more persuasive, not less. The others are the rating scale, which is how audit functions communicate severity, and the independence that requires an auditor to report an unwelcome conclusion as plainly as a welcome one. And its answer-first discipline assumes the writer has an answer. An audit that has not reached a conclusion on part of its scope should say so, and the temptation the method creates, to manufacture a crisp conclusion for the top of the pyramid, is one a reviewer should watch for. The method presents a judgment; it does not supply one. The audit evidence guide and the professional skepticism guide cover the part the book leaves out.
Editions, reading time and how to read it
| What | Details | Best for |
|---|---|---|
| First publication | Minto International, her own imprint; a 1981 edition, catalogued as a revised third edition, already has the ten chapters and appendix of the Pearson text, which Minto’s site dates to 1987 | Historical interest |
| 2002 reissue (third edition) | Financial Times Prentice Hall, 2002; Minto’s site describes it as the 1987 text reissued unchanged | Libraries |
| 2021 reissue | Pearson, Financial Times Series, ISBN 9781292372266 | Recent library copies |
| 2026 paperback | Pearson, FT Publishing International imprint, ISBN 9781292763255, still described as the third edition | The edition in print now |
| The Minto Pyramid Principle (1996) | Minto International; twelve chapters and three appendices in four parts; sold through the author’s website | The version the author considers authoritative |
| Translations | Of the 1996 book: Chinese (two versions), German, Japanese, Korean, Norwegian, Portuguese, Spanish and Vietnamese, according to the author’s site | Teams working in those languages |
| Reading plan | Part 1 straight through; then Part 2 slowly, with a recent audit report open beside it to test against each chapter | An audit manager with two evenings and a draft |
| Companion reading | The library of model audit reports for audit practice; the Influence summary for the psychology of persuasion | Applying the method |
Read it with a report the team has issued in the last quarter and one question: could a committee member who reads only the first paragraph and the headings say what the audit concluded and why? If not, the book explains what went wrong in terms precise enough to fix. Anyone who runs advisory engagements should not skip the appendix, whose subject, problem solving where no structure is given, is the usual condition of advisory work.
A one-hour discussion guide for an audit team
- Ten minutes: the one-sentence test. Each person writes, in one sentence, the conclusion of the last report they drafted. Compare each sentence with that report’s first paragraph.
- Fifteen minutes: rebuild a pyramid. Take a recent report, cover everything except the headings and the executive summary, and draw the pyramid they imply. Mark every heading that names a category rather than stating an idea.
- Ten minutes: the introduction. Rewrite the report’s opening as situation, complication, question and answer, in no more than five sentences, and compare it with the original.
- Fifteen minutes: the order of the findings. For each group of findings, name its order: time, structure or ranking. Reorder any group whose real order is the order of fieldwork, and check any ranking against the function’s severity ratings.
- Ten minutes: the committee paper. Take the most recent audit committee paper and state the question the committee had to answer. If the paper does not answer it on the first page, agree how the next one will.
Worked example: Pennine Foods rewrites its annual report to the audit committee
Pennine Foods plc is a fictional UK listed food manufacturer (commercial companies category) with £900 million of revenue, five UK sites, a December year-end and an internal audit function of four. Its board will make its first declaration under Provision 29 of the UK Corporate Governance Code in the 2026 annual report: a description of how it monitored and reviewed the risk management and internal control framework, a declaration of the effectiveness of the material controls as at the balance sheet date, 31 December 2026, and a description of any material control that was not effective at that date with the action taken or proposed. The chief audit executive’s annual report to the audit committee, due at the February 2027 meeting alongside the 2027 plan, had in earlier years been a chronological account: the year’s engagements with their ratings, a table of open actions, a statement on resources and independence, and an overall conclusion near the end. With the board’s declaration depending on it, the function rewrote it as a pyramid. The Provision 29 guide sets out the requirement and the guide to the UK Internal Audit Code of Practice covers the Chartered IIA’s expectations of the function.
The rewrite started from the question. The situation, which every committee member knew, was that in 2025 the board had identified 38 material controls and asked for assurance over them through 2026. The complication was that three of the 38 had failed testing during the year: two had been remediated and retested before the year-end, and one had not. The question the committee would bring to the meeting was whether the board could make its declaration and what it would have to disclose. The chief audit executive’s answer went into the first paragraph: the material financial reporting and compliance controls were effective as at 31 December 2026, one of them only after remediation in the final quarter; one operational control, the allergen label verification check at one site, was not effective at that date, its remediation would be retested in March 2027, and the board could make its declaration with the description of that control and the action taken that the provision requires.
| Element | The previous annual report | The pyramid version |
|---|---|---|
| Opening | The purpose of the report, the standards followed and the year’s plan | The answer: what the board can and cannot rely on as at 31 December 2026, and what it must disclose |
| Key line | Section headings: plan delivery, findings, actions, resources | Three statements: financial reporting controls effective; compliance controls effective, one after remediation; one operational control not effective and to be disclosed |
| Findings | Fourteen engagement summaries in date order | Grouped under the three statements and ranked by severity, with the three material control failures first in their groups |
| Evidence base | A coverage table at the back | Under each statement: how many of the 38 material controls internal audit tested itself (26) and how many it relied on second-line testing for (12), and how that reliance was checked |
| Open actions | Forty-one actions listed by number and owner | The actions bearing on the declaration first; the rest summarised by theme, with their age |
| The conclusion | Page 19 of 22 | Paragraph one |
| Length | Twenty-two pages | Six pages, with the detail moved to appendices |
The structure changed the meeting. Because the key line separated what was effective from what was not, the committee spent its time on the allergen control: why it had failed, why its remediation could not be retested before the balance sheet date, and how the description in the annual report should be worded. It asked the function, as an advisory engagement, to check the draft description against the evidence before the annual report was signed, which fits the role the Chartered IIA’s guidance on Provision 29 describes for internal audit in establishing the evidence the board’s declaration needs. The reliance on second-line testing for twelve of the controls, which the old report had mentioned in a footnote, became a discussion in its own right once it sat beside the conclusion it supported; the committee asked for the function’s method of checking that testing to be described in the 2027 plan. The guide to presenting to the audit committee covers the meeting itself.
Two cautions came out of the exercise. The first was that the pyramid made the conclusion more persuasive, so the function’s reviewer checked that each statement in the key line was supported by the testing beneath it, and changed one draft statement from “compliance controls are effective” to “effective, one after remediation”, because one control had operated effectively only in the last quarter. The second was that the detail did not disappear: it moved to appendices, where the committee could find it if it wanted to and where the external auditor did. The 2027 plan, presented at the same meeting, was written the same way, with the question it answers first (which risks to the next declaration the function will cover, and how) and the list of engagements after it. A four-person function cannot test all 38 material controls itself every year and still run a full engagement programme, and the plan said so in its first paragraph rather than its last.
Frequently asked questions
Which edition should an auditor buy?
The Pearson edition, the third edition reissued as a paperback in 2026, is the one in bookshops and libraries and the one this guide summarises. Minto’s own 1996 book, The Minto Pyramid Principle, is the version she considers authoritative; it is organised in four parts and sold mainly through her company’s website. For report writing, the Pearson edition covers the rules and the introductions an audit team needs.
Is the pyramid principle the same as putting the bottom line up front?
Answer-first is one part of it. The book adds rules for what goes beneath the answer: groups of one kind, in a stated order, each summarised by a statement of what it implies, and an introduction that tells the reader why the document exists. A report can put a conclusion first and still be badly structured below it.
How does the situation-complication-question-answer pattern fit an audit finding?
At the level of a single finding, the criteria and context are the situation, the condition is the complication, and the conclusion about it is the answer. The 5 C’s guide gives the elements of a finding; the pyramid gives their order, with the significance first and the support after it.
Does a pyramid-shaped report conform to the Global Internal Audit Standards?
The Standards set what a final engagement communication must contain (Standard 15.1) and require an engagement conclusion that judges the findings collectively (Standard 14.5); they do not prescribe the order in which it is presented. The pyramid arranges the required content so that the conclusion leads. The Domain IV guide covers the communication principle at the level of the function.
Does the method work for audit committee slides?
In our view yes, with the key line as the slide titles: each title states an idea, and the slides beneath support it. Slides meant to be read without a presenter need fuller statements than slides that are spoken to. The audit committee deck template is a starting point for applying it.
Where does the book sit in the reading list?
With the communication and report-writing titles; the internal audit reading list arranges the wider library by theme and career stage, and this book underpins the site’s report-writing guides.
Related guides
- How to Write an Internal Audit Report: A Step-by-Step Guide With a Worked Example — the report process the pyramid fits into
- Anatomy of an Audit Report: Structures That Serve Every Reader — report structures by reader
- The Executive Summary: One Page That Decides Your Report’s Fate — the top of the pyramid
- The 5 C’s of Audit Findings: A Masterclass With Five Dissected Examples and a Checklist — the elements beneath each point
- Writing for the Board: Brevity, Altitude, and the Courage to Simplify — writing for directors
- Reporting to the Audit Committee: The Quarterly Pack That Earns Its Airtime — committee papers in pyramid form
- The Best Books for Internal Auditors — where this book sits among the others
- All book summaries — the full set of summaries for auditors
- On Writing Well by William Zinsser: Summary and Lessons for Audit Reports — clear, simple prose for audit reports.
Leave a Reply