EN
Language · same page NLNederlands/ai-en-automatisatie/ai-pyramid-principle-schrijver-skill-voor-claude/ ENEnglish (UK)/en/ai-pyramid-principle-writer-skill-for-claude/ ESEspañolnot translated yet We do not remember your choice and never redirect you automatically.
~/skills/ai-pyramid-principle-writer[ok] loaded
INS.SKILL · FREE FOR CLAUDE, AND FOR CODEX AND CURSOR TOO

The AI Pyramid Principle Writer skill for Claude

You paste a messy text, email or pitch into Claude and you get it back as a pyramid: the core message at the top, below it 3 to 5 arguments that do not overlap, and under every argument the evidence. The method comes from Barbara Minto, who developed it at McKinsey, and the skill enforces her strictest rule: start with the answer, not with the context. What you get back is not a loose block of text but a diagram, an executive summary of at most 250 words and an opening of three sentences, ready to paste into an email, a report or a slide.

Download the skill See all skills (in Dutch) 2.7 KB · zip containing SKILL.md · MIT licence · no account needed
$ installation, line by lineFIG.00
~/.claude/skills
theseo ~ %
  inflating: ai-pyramid-principle-writer/SKILL.md
theseo ~ %
  ---
  name: ai-pyramid-principle-writer
  description: "Applies Barbara Minto's Pyramid Principle to any text, pitch, email, re…
theseo ~ %
[ok] skill active · 883 words of instruction loaded

the lines above come from the zip on this page · SKILL.md is 5,325 bytes

~/skills/ai-pyramid-principle-writer/01-demonstration[run]
Section 01 · The demonstration

Mess in, pyramid out

reg. P.001

The governing thought in this example is taken word for word from the SKILL.md: it is the sentence the file itself puts forward as a good example of a core message. The arguments and figures around it are filled in schematically to show how the method works. Watch the order of the six steps: the answer is sharpened first, and only then may anything be written.

FIG.01: From a messy note to a pyramid[#----] sheet 1/5 · schematic example
~/texts · claude + ai-pyramid-principle-writer
you > this has to become a memo for the board, but the core keeps getting lost. structure this
note.txtschematic example of messy input For a few months now we have seen that enquiries through the site convert well, better than cold outreach really. All sorts of things are going on: the sales team has its hands full, the cost per lead is not bad, and perhaps we should do something with the budget. There is also a discussion about Q3. In short, a choice has to be made.
claude · building the pyramid along the six steps...
[1/6]SCQA framing · situation, complication, question and answer asked for: which question does this piece have to answer?
[2/6]governing thought · the answer in one sentence, a recommendation and measurable, not a description
[3/6]MECE arguments · three arguments proposed, overlap merged, the redundant one dropped
[4/6]form of reasoning · inductive chosen: three parallel points, usually the strongest for executive communication
[5/6]evidence per argument · concrete figures demanded, missing data marked as [open point]
[6/6]final text · seven fixed parts, from governing thought to a closing line that repeats the core
PYRAMIDthe fixed output format of the skill
Raise the Google Ads budget by 40 per cent in Q3 ├── Argument 1: enquiries through the site convert better than cold outreach │ └── Evidence: conversion per channel over the last few months, figure with a source ├── Argument 2: the cost per lead sits below the internal benchmark │ └── Evidence: cost per lead against that benchmark, figure with a source └── Argument 3: the sales team can demonstrably handle the current volume └── Evidence: [open point] capacity figure still to be supplied
text.mdexecutive summary, shown shortened First sentence: the governing thought itself, so the recommendation is there before a single line of context has been read. Then: the SCQA opening in at most three sentences, a signpost (there are three reasons for this), two sentences of evidence per argument, and one closing sentence that repeats the core message. Length: at most 250 words, so the piece lands in the first thirty seconds.
[######] 6/6 done · every [open point] is left empty on purpose: the skill does not invent evidence
6P.01 · steps in the method
5P.02 · arguments at most (3 to 5, MECE)
250P.03 · words at most in the final text
0P.04 · euros, MIT licence, no account
P.01 to P.04 · properties of the skill file, not results or usage figures.
$ cat 01-what-the-skill-is.md
DOC.01

What the AI Pyramid Principle Writer is

The AI Pyramid Principle Writer is a free skill from the TheSEO skill library. A skill is an instruction file, SKILL.md, that gives an AI assistant a fixed way of working for one task. Not software you install, not a subscription, not a template: one text file of 883 words that tells Claude how to rebuild any text into Barbara Minto's structure, in which order the six steps run, and which output format always comes out. You download the zip at the top of this page, put it in your Claude environment, and from that moment Claude structures your texts along this logic. If you are not yet sure what a skill is, start with what Claude skills are.

You recognise the problem the skill solves by one symptom: your text is long, but the core is unclear. The email to the investor starts with the founding of your company. The memo to the board builds three pages of tension towards a conclusion nobody reads any more. The annual plan lists twelve points, three of which say the same thing. The reader has to wade through every side issue before understanding what it is about, and most readers never get that far. This one turns the order around: the answer first, then the arguments, and only then the evidence.

By its own description the skill is meant for anyone who hears: write this more clearly, my message is not landing, my email is a mess, make this pitch ready. It names its strongest uses itself: an email to an investor about a pivot, an internal memo about a reorganisation, the executive summary of a slide deck, a pitch for a new client, an annual plan for the board, or a blog article that has to persuade. It belongs to the library of 100 free skills we make available from our AI and automation service, with no account and no sales email afterwards. All of them sit on the hub skills.

One thing to know in advance: the skill structures, it does not invent. The arguments come from your own material and the evidence has to come from you. If data is missing under an argument, no elegant sentence appears in its place but an open point with a question for you. That is not a limitation, it is the reason the final text survives a critical reader.

$ cat 02-why-texts-do-not-land.md
DOC.02

Why most business writing does not land

Barbara Minto developed the Pyramid Principle in the 1960s at McKinsey, where as one of the first female consultants she had to edit her colleagues' reports. Her observation was simple and has since become the standard for consultants, executives and managers: most people write in the order in which they thought, not in the order in which a reader decides. They start with the background, work through the analysis and end with the conclusion. The reader wants exactly the opposite.

The core arrives too late. A busy reader gives a document thirty seconds to make clear what it is about. If the answer sits on page three, the reader has already stopped on page one or, worse, started guessing at a conclusion. Which is why the governing thought has to be the very first sentence of the final text, before the opening.

The core sentence is not an answer. Many texts do have something like a core, but it is a description: we have analysed the market. That is not a message, that is an activity. So the governing thought has to be an opinion or a recommendation, fit in one sentence, is concrete and measurable, and answers the question from the opening. Marketing plays an important role is rejected in the file itself as a weak example.

The arguments overlap or are missing. Twelve loose points are not a line of reasoning. So it enforces 3 to 5 arguments that are MECE: they exclude each other and together cover the whole, on the same logical level. Apples next to apples, no half argument that really belongs under another one.

The evidence is thin air. Studies show, everybody knows, it is common knowledge: the skill rejects that kind of support. Under every argument belongs a concrete figure, example or source, and where that is missing an open point appears instead of a guess.

In diff form, with the examples from the file: what goes out and what comes back in its place.

~/skills/ai-pyramid-principle-writer/02-pyramid[ok]
Section 02 · The anatomy

Three layers, one fixed order

reg. P.002

Every output of the skill has the same shape: a top, a middle layer and a foundation. The reader reads from top to bottom and may stop at any point, because the most important part has already been said. The content in the blocks follows the example from FIG.01.

FIG.02: The anatomy of the pyramid[##---] sheet 2/5
Layer 1 · the governing thought One sentence. A recommendation, not a description. Concrete and measurable. Answers the question from the SCQA opening.
ARGUMENT 1Excludes the other arguments, overlaps nowhere. Same logical level as 2 and 3.
ARGUMENT 2Together the arguments cover the whole case: nothing doubled, nothing forgotten.
ARGUMENT 3Three to five of them. Take one away and the case has to collapse, otherwise it was redundant.
evidence > concrete figure, example or source. No "studies show".
evidence > two sentences per argument in the final text, no more.
evidence > if data is missing, an open point and a question for you.
# reading direction from top to bottom · the reader may stop after every layer, the core is already in
$ less ai-pyramid-principle-writer/SKILL.md # 883 words of instruction
DOC.03

What the SKILL.md actually says

A skill is only as good as its instructions, so we simply describe them here. The file opens with a description of when Claude should pick the skill up: on explicit terms such as pyramid principle, Minto method, MECE thinking or McKinsey structure, but also on everyday complaints such as write this more clearly, my email is a mess, pull out the main message, or when a text is long but the core is unclear. Say something like that to Claude while the skill is loaded and you get it automatically.

Then come the six steps. Step one is the SCQA framing: the skill asks for the situation everyone takes as given, the complication that changed it, the question that follows from it and your short answer to that question. Step two turns that answer into the governing thought and tests it against four demands: an opinion or recommendation, in one sentence, concrete and measurable, and a real answer to the question. Step three has Claude propose 3 to 5 arguments and tests them for MECE: no overlap, covering the whole, on the same logical level.

Step four chooses the form of reasoning: deductive, with premises leading to a conclusion, or inductive, with parallel points carrying the same message, where the file notes that induction is usually faster and stronger for executive communication. Step five demands evidence per argument and rejects vague support. And step six builds the final text along a fixed shape of seven parts: the governing thought as the first sentence, the SCQA opening in at most three sentences, a signpost, three arguments with two sentences of evidence each, and a closing sentence that repeats the core message.

The output format is fixed as well, and for daily use that is perhaps the most useful part. You always get three blocks: the pyramid as a diagram with the arguments and the evidence below them, the full text in executive summary form of at most 250 words, and the SCQA opening of three sentences as a separate block. You paste the diagram into a slide, the text into an email, the opening above a report. Want to build such an instruction file yourself? The structure is explained step by step in writing a SKILL.md.

It closes with a source list of three titles: Minto's own book, the structured communication guidelines from the McKinsey Quarterly archive, and Power Cues by Nick Morgan. More on those below.

~/skills/ai-pyramid-principle-writer/03-mece-test[ok]
Section 03 · The test

MECE: nothing doubled, nothing forgotten

reg. P.003

MECE is the strictest part of the method and the point where most self built pyramids collapse. Every argument gets tested in two ways, exactly as the file prescribes them: if two arguments overlap they are merged, and if the case does not collapse when you remove an argument, that argument goes. The content below is schematic, the tests are literal.

FIG.03: mece-test.log[###--] sheet 3/5 · schematic example
mece-test.logthe two tests from step 3 of the SKILL.md
argument 2: customers drop out while orderingargument 4: the checkout has too many steps MERGED Overlap test. These two say the same thing at a different height: the second is the cause of the first. Overlapping arguments are merged into one argument at the right level, otherwise the reader counts it twice.
argument 5: the newsletter could be better too DROPPED Removal test. The question from the file: if I remove this argument, does the case still stand? Here the case simply stays upright, so the argument was not needed. Being true is not enough, it also has to carry weight.
arguments 1, 2 and 3: conversion, cost, capacity MECE OK Level test. Three arguments, no overlap, together they cover the decision, and they sit on the same logical level: apples next to apples. This is the set that gets into the pyramid.
$ cat 04-sources-and-theory.md # Minto, McKinsey Quarterly, Morgan
DOC.04

The theory the skill rests on

The SKILL.md names three sources by name, and that is deliberate: it lets you read them and decide for yourself whether you agree with the method.

The method comes from Barbara Minto. The Pyramid Principle: Logic in Writing and Thinking was published in 1987 by Pearson Education and is the canonical write up of what she developed at McKinsey in the 1960s. Everything the skill enforces is in there: the answer first, the SCQA opening, the pyramid of arguments, and the difference between deductive and inductive reasoning. The book is not light reading, and that is exactly why a skill that carries out the method is more practical than learning the method by heart.

The working rules come from the McKinsey Quarterly archive. The structured communication guidelines published there over the years translate Minto into daily work: how short an opening may be, how to use a signpost, why an executive summary starts with the recommendation. The fixed shape of seven parts in step six leans on that tradition.

The receiving end comes from Nick Morgan. Power Cues was published in 2014 by Harvard Business Review Press and is about how messages come across: what a listener or reader picks up in the first seconds and why the order of information weighs so heavily. It underpins the thirty second demand the skill puts on the final text.

Even without installing the skill you can improve your own text with these principles: write down your conclusion in one sentence first, cross out every argument the case does not need, and put a figure you can show under every argument that remains.

If the trouble is not the order of your argument but whether you understand the subject well enough to explain it plainly, the AI Feynman Technique Coach is the skill that finds that out. And once the structure is right, if the message itself still has to stick, the measure for that is in the Made to Stick Coach skill, and if you want to train breaking problems apart as a separate craft, start with the MECE Issue Structure Coach skill.

~/skills/ai-pyramid-principle-writer/04-rejections[ok]
Section 04 · The limits

What the skill rejects

reg. P.004

The SKILL.md has no separate list of prohibitions, but the six steps themselves are full of hard rejection rules, and those matter at least as much as what the skill does do. In a conversation they play out like this: every line is a request or a piece of material you might hand over, with the response the skill gives according to its own instructions.

FIG.04: rejections.log[####-] sheet 4/5
rejections.logthe rejection rules from the six steps of the SKILL.md
my core message: marketing plays an important roleREJECTEDToo vague for a governing thought. This is in the file itself as a weak example. A core message is an opinion or recommendation somebody can disagree with, concrete and measurable.
then this one: we have analysed the marketREJECTEDNot an answer. Also a literal example from the file: this describes an activity, not a conclusion. The question remains: and what should happen now?
as evidence: studies show that this worksREJECTEDVague support is rejected. Studies show and everybody knows do not make it. Under every argument belongs a concrete figure, example or source. If you have no data, it becomes an open point with a question for you.
these two arguments look alike, just leave themMERGEDOverlapping arguments are merged. MECE means excluding each other. Two arguments that say the same thing count as padded repetition to the reader, so they become one.
add a sixth and a seventh argument, the more the betterREJECTEDThree to five arguments, no more. And every argument has to pass the removal test: if the case does not collapse without it, it goes. Volume is not reasoning.
start the memo with some background about the past yearCORRECTEDThe text starts with the answer, not with the context. The governing thought is the first sentence, and only then the SCQA opening in at most three sentences. If you want context first, you still get it, but short and in second place.
$ unzip ai-pyramid-principle-writer-skill-for-claude.zip -d ~/.claude/skills/
DOC.05

Installing in Claude Code, Claude.ai or Codex

The zip contains one folder with the SKILL.md in it. Installing is a matter of putting the file in the right place, and that place differs per environment. SKILL.md has been an open standard since December 2025, so the same skill also works in Codex, Cursor and Gemini CLI. You are not downloading a Claude file but a working instruction any modern AI assistant can read.

CLAUDE CODE
  1. Unzip it into ~/.claude/skills/ (or .claude/skills/ in your project).
  2. Claude then picks the skill up by itself as soon as you want to structure a text.
  3. You can also call it directly, with /ai-pyramid-principle-writer.
CLAUDE.AI
  1. Go to Customize and then Skills.
  2. Upload the zip there as a skill.
  3. Or paste the contents of SKILL.md into the project instructions of a Project.
CODEX
  1. Open AGENTS.md in your repo.
  2. Paste the contents of SKILL.md into it, or put SKILL.md next to it as a separate file and point to it from AGENTS.md.
  3. Codex reads that along at every session.

After that, using it is simple: paste your messy text, email or notes and say what it has to become, for example a memo for the board or an executive summary. It asks the SCQA questions it needs to sharpen the answer and then builds the pyramid. If you get stuck installing it, the full route per environment is in installing Claude skills. The wider explanation of working with AI is in the knowledge base.

$ cat 06-when-to-use-it-and-when-not.md
DOC.06

When to use it and when not

At its strongest it is on texts that ask for a decision: the email to the investor, the memo about a sensitive change, the annual plan, the pitch, the executive summary above a long report. Anywhere a busy reader has to understand in thirty seconds what you want from them, the pyramid pays for itself immediately. It works well as a thinking tool too: anyone who cannot get their own plan into one governing thought with three load bearing arguments knows straight away where the plan itself is still shaky. That is often the biggest gain: the structure exposes what is missing in the substance.

There are also situations where you are better off leaving it alone. Texts that depend on tension and build up, a story, a case study working towards a reveal, a speech with a climax, lose their force if the answer sits in the first sentence. For descriptive documentation without a message, a manual or a report, a pyramid adds little: there is nothing to argue there. And the method orders arguments, it does not make them true: a governing thought that is wrong on the substance is only punctured faster by a tight structure, because the reader now sees it straight away.

The most honest limit sits in the evidence. Figures are demanded and gaps are marked, but nothing here can check whether your figures are right. The open points it hands back are homework, not a formality: a pyramid on unchecked evidence is a risk with a nice structure around it.

~/skills/ai-pyramid-principle-writer/05-upgrade-path[ok]
Section 05 · From skill to employee

Run it yourself or have it run for you

reg. P.005

This skill is the free do it yourself version of work we also deliver as a service. It stays complete and without catches, but understand what a skill is: it teaches your AI how to do something, while every new session starts empty. A skill is not the engine and not the memory. You prompt, you supply your figures and your context again every time, you check. If you want that differently, there are two next steps: hand over the engine, or sort out the memory.

$ cat from-skill-to-employee.mdthree steps, the same work
upgrade-path.shfree · employee · brain
STEP 1 · FREE
where you are now
The skill: you are the engine You run the AI Pyramid Principle Writer yourself in Claude, Codex or Cursor. It costs nothing, it works today, and you keep it entirely in your own hands: no trial period, no locked off parts. The limit is your own time: it only happens when you prompt, and every session you supply the figures and context the pyramid has to stand on all over again. $ claude --skill ai-pyramid-principle-writer · €0 · you prompt, you check
STEP 2 · SERVICE
have it prepared
The Quote employee: the recurring writing is ready and waiting You write an advisory note or a pitch yourself, because the core message is a choice and not a summary. What can be delivered as a service is the writing that comes back every week: incoming enquiries sorted by opportunity and urgency, with a draft answer per enquiry already built from core message down to supporting evidence. That is the Quote employee from Mansotti, the company TheSEO trades as, which also builds the other two roles and sets up the work around them to measure. The control stays with you, because output stays a draft until a person approves it. Read what an AI employee is and does. from €950 per month · a person approves, always
STEP 3 · BRAIN
everything from one source
Jarvis: all your AIs work from the same company knowledge The skill teaches the AI, the brain is where the memory lives. If you want all your AIs to work from the same company knowledge, that is Jarvis, the organisation brain. It connects ChatGPT, Claude, Codex and your people to the same projects, core knowledge and decisions, so your next AI session does not start over. Jarvis selects context, delegates work and stores only what has been approved: client knowledge stays isolated and every step leaves a checkable trail. This skill gets better from it too, because the figures and decisions your pyramids rest on no longer have to be supplied per conversation. What that delivers in practice, from the plans to your first week, is at Jarvis itself. entry plan Brain Start: €9 per month incl. Dutch VAT · pay for your brain, not per AI question
# no marketing trick: step 1 stays free and complete. The next steps are there for anyone who wants to hand this work over.
~/skills/ai-pyramid-principle-writer/06-jarvis[ok]
Section 06 · The brain

What Jarvis actually delivers

reg. P.006

Step 3 deserves more than a paragraph, because this is the difference between a clever chat and a system you can build on. Jarvis is the organisation brain: it remembers what your AIs need to know, divides the work and keeps track of what happened. With structured writing you notice that immediately, because a pyramid stands on evidence, and the evidence, the earlier decisions and the house rules for board documents are exactly what you otherwise supply again every session.

FIG.05: What a writing session gets back from the brain[#####] sheet 5/5
jarvis · organisation brain● sync
$jarvis recall "reporting agreements" # schematic example
[core]every board document starts with the answer: governing thought in one sentence at the top, context only after that
[core]house rule on evidence: every argument carries a figure with a source, vague support is sent back
[task]monthly report · pyramid stands, draft waiting for human approval
[decision]inductive build as the standard for board memos: recorded after review, approved by a person
[log]previous session: claude built the pyramid, a person sharpened argument 2, version stored
[ok]context loaded · this session does not start empty
this is how every assignment moves through the brain: recordedcontext chosendelegatedhuman approvalstored · the full trail is at jarvis/werking (in Dutch)
context.kept Your next session does not start over Today you explain how your board reports and which figures are sacred, and tomorrow a loose chat knows nothing about it. With Jarvis every session starts with the same agreements, projects and earlier decisions, as in FIG.05: recall first, then build.
ai.connected ChatGPT, Claude and Codex, one source Every connected AI works from the same core knowledge and agreements. The governing thought you settle and approve in Claude is what ChatGPT uses tomorrow in the follow up email. You explain nothing three times and three separate truths never appear.
tasks.tracked Tasks planned, tracked, reported done An assignment is recorded with a goal and a deadline, picked up by the right agent and reported done with the result attached. At any moment you see what is running, what is waiting and what is finished.
everything.logged Everything logged and open to inspection Every step leaves a checkable trail: who asked what, which sources were used, which agent worked on it and who approved. Not because it is required, but because otherwise you cannot check what happened in your name.
human.approval Nothing goes out without approval AI prepares, a person decides. Output stays a draft until somebody approves it, and only approved knowledge goes back into the brain. That line holds everywhere in the system, including work an agent prepared entirely on its own.
brain.isolated Client brain isolated If you work for several clients, the knowledge stays strictly separated per client. One client's figures do not leak into another client's report.
# THE HONEST EVIDENCE · NOT A DEMO

We have been running on this system ourselves for months, and that is the only reason we put it on this page at all. Every agent session, every task and every decision is logged in it and can be read back. A new session therefore does not start blank: it first fetches the recorded decisions, the running projects and the latest changes, and carries on where the previous one stopped. So we are not describing a promise but the way we work every day.

$ cat prices.mdpay for your brain, not per AI question
Brain Start €9 /month incl. VAT 1 organisation brain · 1 user · 1 AI employee
Brain Solo €29 /month incl. VAT 1 organisation brain · 1 user · 3 AI employees
Brain Team €99 /month incl. VAT 1 organisation brain · 5 users · 10 AI employees
Brain Business €249 /month incl. VAT 3 organisation brains · 20 users · 50 AI employees

Amounts are in euros and include Dutch VAT. All plans are listed at jarvis/prijzen (in Dutch). You choose your own plan. Paying online opens once the last checks are finished; until then we set your account up on request. For business arrangements and bespoke work there is a conversation.

~/skills/ai-pyramid-principle-writer/07-chain-of-thought[ok]
Section 07 · The chain of thought

The skills around it

reg. P.007

Structure is one link: a message goes into the pyramid and somebody has to be convinced by it. These skills from the same library each take on a different part of that work.

$ claude --interactive # seven questions, seven answers
DOC.07 · FAQ

Frequently asked questions

What does the Pyramid Principle Writer skill cost?

Nothing. The skill is free, falls under the MIT licence and you do not have to create an account or leave an email address. You download a 2.7 KB zip containing one folder and a single file, SKILL.md, and that is the complete skill. There is no paid version and no sales email follows.

Does this skill work in Codex, Cursor or Gemini CLI?

Yes. SKILL.md has been an open standard since December 2025, so the same file works in Codex, Cursor, Gemini CLI and other tools that follow the standard. In Codex you unzip it into .agents/skills/ in your project, or into ~/.agents/skills/ for all your projects; Codex has supported SKILL.md directly since the open standard of December 2025. Pasting the contents of SKILL.md into your AGENTS.md still works too. The instructions are plain readable text, so any AI assistant that accepts instruction files can handle it.

What exactly is a governing thought?

The sentence at the top of the pyramid: your core message. The file sets four demands for it. It has to be an opinion or a recommendation and not a description, it has to fit in one sentence, it has to be concrete and measurable, and it has to answer the question from the SCQA opening. Marketing plays an important role is rejected as too vague. Raise the Google Ads budget by 40 per cent in Q3 is a real one: you can disagree with it, and that is exactly the point.

What does MECE mean and why does it matter?

MECE stands for Mutually Exclusive, Collectively Exhaustive: the 3 to 5 arguments under your core message may not overlap and together have to cover the whole line of reasoning, on the same logical level. The skill tests that actively. If two arguments overlap, they are merged. And if the case does not collapse when you remove an argument, that argument was not needed and it goes. What is left is only the reasoning that really carries.

What if I have no figures or evidence for an argument?

Then the skill invents nothing. Vague support such as studies show or everybody knows is rejected by the file, and an argument without data is marked as an open point, with a question for you to supply it. That feels strict, but it is the reason the final text stays upright when somebody digs into it: every argument carries evidence you can show.

Does the skill choose between deductive and inductive reasoning itself?

The skill knows both forms from Minto's work and chooses whichever fits your message. Deductive is premise, premise, conclusion: three steps that work for logical proof. Inductive is three parallel points supporting the same message, and that works for business cases and for persuasion. The file adds the rule of thumb that induction is usually faster and stronger for executive communication, because all the arguments sit at the same level.

What exactly do I get back and how long is it?

Always the same fixed format, in three parts. First the pyramid itself: the governing thought with the arguments below it and the evidence per argument, as a diagram. Then the full text in executive summary form of at most 250 words, built along a fixed order of seven parts. And finally the SCQA opening of three sentences you can put above an email or a memo. You can paste the diagram into a slide and the text into an email, with no rework.

$ cat 08-and-now.md
DOC.08

A pyramid with a reader underneath it

A pyramid makes sure your message lands with whoever reads it. In an email you pick that audience yourself; on your website the search engine has a say in who comes in. Use this structure on pages that are barely visited and you write the sharpest core message for nobody. The free SEO scan shows in seconds which pages are in the running and which stay silent. In the guide writing models compared you can see how this model relates to the others.

Section 08 · Next stepreachable 24/7
Book a call
$ whoami
Gianluca, founder of TheSEO
Written by GianlucaFounder of TheSEO. Has been building visibility for businesses since 2017, in Google and in AI answers. More about the institute.