The Writer's Journal • 28 September 2026

Structure Before Detail: Stop Handing Readers Bricks With No Wall

Structure Before Detail: Stop Handing Readers Bricks With No Wall

A reader abandons your chapter on page three. Not because the information is wrong — it is accurate, well sourced and carefully phrased. They leave because they are holding eleven facts and have not been told what any of them are for.

You handed them bricks before you showed them the wall. Every fact arrived with no place to go, so each one had to be carried rather than filed — and carrying is expensive. This is not a taste problem or a style problem. It is a measurable limit on what a mind can hold, and it has a fix.

You have this problem if

  • Readers say the text is "dense" or "hard to follow" but cannot point at a sentence.
  • People ask a question you answered clearly three pages earlier.
  • You find yourself writing "as mentioned above" more than twice in a chapter.
  • Your opening paragraph begins with a definition or a piece of history.
  • A beta reader finished the chapter and could not summarise it in one sentence.

Why it happens: three findings worth knowing

Writing advice is mostly folklore. This particular rule is one of the few with actual research behind it, which is why it holds across novels, textbooks, reports and documentation alike.

Working memory is small

George Miller's 1956 paper put the span of immediate memory at about seven items, plus or minus two; later work by Nelson Cowan argued the real figure for unchunked material is closer to four. The exact number is disputed and the direction is not: it is a handful, and it is far smaller than writers assume.

Crucially, Miller's escape route was chunking — group items under a meaningful heading and the group costs one slot instead of five. That is exactly what a structure does for a reader.

A frame given first does the filing for them

David Ausubel's research in the 1960s introduced the advance organizer: a short, general statement presented before the detail, which gives the learner somewhere to attach what follows. Material introduced this way is understood and retained better than the same material presented cold.

His summary of the whole field was blunt: the single most important factor influencing learning is what the learner already knows. A structure stated up front is how you build that prior knowledge in thirty seconds.

Effort spent on navigation is effort not spent on the content

John Sweller's cognitive load theory distinguishes the load inherent in the material from the load added by how it is presented. Making a reader work out why they are reading a paragraph is purely added load — it teaches nothing and consumes the capacity they needed for the idea itself. Every "wait, how does this connect?" is a tax you charged for free.

None of this is new. Ibn Khaldun described the same principle in the Muqaddimah in the fourteenth century, arguing that a subject should be taught in three passes — a first covering the broad principles in outline, a second going deeper, a third leaving nothing obscure — and criticising teachers who opened with the difficult cases. The research caught up with the practice six centuries later.

A person writing in a notebook — give the frame before the facts
A fact with nowhere to go has to be carried. Carrying is what exhausts a reader.

The same rule at four scales

Frame before detail is fractal. It applies to the book, the chapter, the section and the paragraph, and a text that obeys it at every level feels effortless in a way readers cannot explain.

The book

An introduction that states the question the book answers and the route it will take. Not a preface about how you came to write it — a map.

The chapter

Two or three opening sentences saying what this chapter settles and why it comes here rather than earlier. Every chapter should answer "why now?" in its first breath.

The section

A heading that states a claim rather than naming a topic. "Working memory is small" is a frame; "Memory" is a filing label that tells the reader nothing.

The paragraph

The point in the first sentence, the support after it. Barbara Minto built an entire consulting method — the Pyramid Principle — on exactly this: answer first, then the reasons, then the evidence.

Journalism arrived at the same shape independently and calls it the inverted pyramid: the conclusion in the first line, because a reader who leaves after one paragraph should still have got the point. Fiction is the interesting exception, and it is a narrower one than novelists like to think — a scene can withhold facts freely, but it still has to tell you whose scene it is and what is at stake, or the tension has nowhere to sit.

The same chapter, two orders

Detail first — the reader is lost by item three
  1. The history of the method, 1890 to now
  2. Definitions of nine key terms
  3. A detailed case study
  4. A second case study
  5. What all this means
  6. How to apply it

Everything the reader needs to make sense of items 1–4 arrives at item 5.

Frame first — the same material, now usable
  1. The problem this chapter solves, in two sentences
  2. The answer, stated plainly and early
  3. Why the obvious alternative fails
  4. One case study, now interpretable
  5. The terms, defined where they are first needed
  6. How to apply it, and the history as a short aside

Nothing was cut. Two blocks moved, and the second case study turned out to be unnecessary.

That last detail is the usual bonus. Once the frame comes first, redundant material becomes visible — you were repeating yourself to compensate for a reader who kept losing the thread.

An architectural frame under construction — the wall before the bricks
Nobody hands over bricks before there is a wall to put them on.

Three objections, answered

"Giving the answer first spoils the suspense."

In a mystery, yes. In an argument, no — and the confusion between the two ruins a great deal of nonfiction. A reader who knows your conclusion reads your evidence more carefully, because they now know what it is evidence for.

"My subject is genuinely complex."

Complex subjects need this more, not less. Complexity is precisely what exhausts a reader's capacity, which leaves nothing spare for working out your organisation. Difficulty should live in the ideas, never in the arrangement.

"It makes the writing feel mechanical."

The structure should be felt, not seen. A frame can be a single vivid sentence rather than a numbered roadmap — the requirement is that the reader knows where they are, not that you announce it in a bulleted list.

The practical route into all of this is planning: see how to plan a book, and the four decisions that come before the first sentence.

See the shape while you write

You cannot fix an order you cannot see

Structure problems are invisible inside one long file. In a tree they are obvious — and fixing them is a drag, not an afternoon of cutting and pasting.

The whole outline, always visible

Read your chapter order like a table of contents while you write inside it.

Reorder by dragging

Move the frame to the front and the history to the back in seconds.

Search across the book

Find every "as mentioned above" and fix the order that made it necessary.

Version history

Try a different order without fear. The old one is still there.

Questions about structure and clarity

Why does my writing feel hard to follow? +

Usually because detail arrives before the frame that makes it meaningful. A reader holding facts with nowhere to file them runs out of working memory long before they run out of interest — and they experience that as "dense" without being able to point at a sentence. Check your opening: if it starts with a definition, a date or a piece of history, the frame is missing.

What is cognitive load in writing? +

It is the mental effort a reader spends while processing your text. Some of it is unavoidable, because the subject is hard. The rest is added by presentation — unexplained terms, buried structure, facts that arrive before their context — and that portion teaches nothing while consuming the capacity the reader needed for your actual ideas. Good structure is mostly the removal of that avoidable load.

Should I state my conclusion at the start? +

In argument, reporting, documentation and most nonfiction chapters, yes. Readers evaluate evidence better when they know what it supports, which is why journalism uses the inverted pyramid and why Minto's Pyramid Principle puts the answer first. Withholding belongs to storytelling, where the tension itself is the product.

How many points can a chapter safely carry? +

Fewer than you want. Three to five main points per chapter is a sound working limit, not because of a magic number but because unchunked items compete for a small working memory. If you have nine, they are usually three groups of three — and naming the groups is what turns nine burdens into three.

Does this apply to fiction? +

Differently, but yes. Fiction withholds plot facts on purpose, and should. What it cannot withhold is orientation: whose scene this is, where it happens, and what is at stake. A reader who does not know who is speaking is not intrigued, merely lost — and the two feel nothing alike from the inside.

Open your hardest chapter and read the first paragraph

If it begins with a definition or a date, you have found the problem — and the fix is a reorder, not a rewrite.

Start writing free

Tags

structure outlining readability nonfiction