Why structure matters more than content when speaking
A reader who loses the thread of a written report can go back and re-read a paragraph. A listener who loses the thread of a talk cannot: once a sentence has been spoken, it's gone. This basic difference explains why the structure of a spoken talk needs to be more explicit and more disciplined than a written text meant to be read at the reader's own pace.
Barbara Minto, a consultant at McKinsey in the 1960s, formalised in "The Pyramid Principle" an idea that has since shaped communication at most consulting firms and large companies: present the conclusion or recommendation first, then walk through the arguments that support it — never the other way around. This "pyramid" format runs against the natural instinct to narrate one's thinking before arriving at a conclusion.
George Miller documented as early as 1956 that human working memory holds only a limited number of items at once. More recent work (Cowan, 2001) has revised that limit downward, to around 3 to 4 unrelated items without rehearsal support. That limit is a strong argument for a single central message when speaking, not a list of 5 to 7 points.
Miller, G. A. (1956). The Magical Number Seven, Plus or Minus Two. Psychological Review, 63(2), 81-97. Cowan, N. (2001). The Magical Number 4 in Short-Term Memory. Behavioral and Brain Sciences, 24(1), 87-114.The pyramid method applied to speaking
Minto's pyramid principle transfers directly to public speaking: identify the central message before preparing anything else, then organise the arguments so they explicitly support that message, rather than stringing them together in the order they came to mind.
Start with the answer, not the journey
In practice: state the conclusion, recommendation, or central message in the first few sentences, then spend the remaining time justifying it. This runs against the natural instinct to build a progressive argument toward a punchline — a format that works well in writing, but loses a pressed or distracted audience when spoken live.
Signposting: giving explicit reference points
Because the audience can't go back, every transition needs to be announced explicitly: "I've covered the first point, let's move to the second", "to summarise before I conclude". These signposts replace the re-reading a written text allows, and let the audience reconnect with the structure even if their attention briefly wandered.
The mistake that loses an audience halfway through
The most common mistake is structuring a talk in the chronological order of your own thinking: context first, then data, then analysis, and only at the end — the conclusion. That path, natural when you discovered the answer progressively yourself, forces the audience to hold every intermediate piece without knowing what they're building toward.
- Structuring a talk in the chronological order of your thinking rather than the order that's useful for the audience
- Defending several central messages at once instead of one clearly identified message
- Skipping explicit transitions between parts, assuming the logic is obvious to everyone
- Saving the conclusion for the end out of habit, even when the audience needs to hear it first
- Confusing richness of content with clarity of structure — the two don't offset each other
Narrating your own thought process can feel transparent and honest, but it shifts onto the audience the burden of reconstructing the logical structure while listening — exactly what the working-memory limit makes difficult.
How to embed this structure for good
Knowing the pyramid principle isn't enough to resist the instinct to narrate your own thinking when preparing a talk. Like any behavioural skill, embedding it comes from spaced repetition over time.
That's the approach behind the "Structuring Thoughts and a Talk" deck on memia: flashcards that regularly revisit identifying the central message, organising arguments, and signposting, until this principle becomes a reflex you use before every talk.
Continue on presence and eloquence
Frequently asked questions
Why does structure matter more than content when speaking?
Because a listener can't go back the way a reader can re-read a paragraph. Once a sentence has been spoken, it's gone — so structure has to carry most of the clarity, independent of how rich the content is.
What is Barbara Minto's pyramid principle?
Formalised at McKinsey in the 1960s, this principle means presenting the conclusion or recommendation first, then walking through the supporting arguments — the opposite of the natural instinct to build a progressive argument toward a punchline.
How many ideas can you defend in a single talk?
Miller's (1956) work and its revision by Cowan (2001) place working-memory capacity around 3 to 4 unrelated items without rehearsal support — a solid argument for one central message rather than a list of 5 to 7 points.
What is "signposting" in public speaking?
Explicit reference points given to the audience to mark transitions: "I've covered the first point, let's move to the second", "to summarise before I conclude". They replace the re-reading a written text allows.
What is the most common mistake in structuring a talk?
Structuring a talk in the chronological order of your own thinking — context, data, analysis, then the conclusion at the end — instead of starting with the conclusion. That path forces the audience to hold intermediate pieces without knowing what they're building toward.
Can structuring a talk be learned with flashcards?
Flashcards don't replace practising in real situations, but they anchor identifying the central message, organising arguments pyramid-style, and signposting until they become reflexes you use before every talk. That's the role of the "Structuring Thoughts and a Talk" deck in memia's Presence & Eloquence guide.