5 min read

How to Outline a Novel (Without Killing the Draft)

Plotter, pantser, and the false choice between them. A practical guide to outlining a novel at the level of detail that helps, and no more.

The outlining debate is conducted almost entirely between people who have mistaken their working method for a moral position.

Plotters say you cannot build a house without blueprints. Pantsers say a novel is discovered, not manufactured, and an outline is just a spoiler you wrote yourself. Both are describing something true about their experience and false as a general law.

The useful question isn't whether to outline. It's how much you can know in advance without killing the reason to write it.

What an outline is actually for

Not to tell you what happens. To tell you whether what happens holds together.

An outline is a compression of the novel. Its value is that a compression can be audited in an afternoon, whereas the novel takes a year. You look at twelve lines and you see that your protagonist makes no decisions until line nine. You would not have seen that from inside chapter four.

That is the whole benefit, and it explains the whole cost. Compress too little and you can't see the shape. Compress too much — outline in such detail that every sentence is predetermined — and you have written the book once already, in a bad form, and the draft becomes transcription. Many abandoned novels are abandoned there: not because the writer got stuck, but because the book was already known, and knowing was the interesting part.

The three altitudes

Think of the outline as existing at three zoom levels. Most advice conflates them, which is why most advice contradicts other advice.

Altitude 1: The premise (one paragraph)

Who wants what, what stands in the way, and what it will cost them to get it. If you cannot write this paragraph, no amount of chapter-level planning will save you, because you don't yet have a story — you have a situation.

Write this first. Rewrite it whenever the book surprises you.

Altitude 2: The turns (five to ten lines)

Not chapters. Turns — the moments after which the story cannot go back. The lie the protagonist believes, the choice that commits them, the false victory, the collapse, the cost paid. Whatever names you use, these are the load-bearing walls, and there are rarely more than ten.

This is the altitude at which structural problems are visible and cheap to fix. Almost every "my middle sags" is really "I have two turns where I need five."

Altitude 3: The chapters (one line each)

Whose viewpoint, what changes. Written just ahead of the draft, not months before it. Two or three chapters of runway is enough — far enough to know where you're walking, close enough that the characters can still argue with you.

Why the third altitude should stay shallow

Because your characters do not exist yet.

You can plan chapter nineteen in January, but the woman who arrives in chapter nineteen will have been shaped by eighteen chapters you haven't written, and she will be more specific, more contradictory, and more interesting than the placeholder you planned around. If the outline is detailed enough to override her, you'll follow the outline. And the reader will feel it: a character doing what the plot needs rather than what she would do.

The outline should be a hypothesis, and the draft should be allowed to falsify it.

Outlining after the fact

The most under-rated method: write a hundred pages into the fog, then outline what you have, at altitude 2. Summarise each chapter in a sentence. Lay them out. Now you can see whether there's a book here.

This is not a failure to plan. It's planning with data instead of guesses. The book tells you what it is about, and the outline becomes a description rather than a prediction. Then you outline the remaining half properly, because now you actually know who these people are.

Rearranging is the point

Whichever altitude you're at, the outline earns its keep only if you move things. This revelation lands too early. This subplot goes quiet for eighty pages. These two chapters do the same work and one must die.

Which means the format matters more than people admit. A numbered list resists rearrangement — moving item 14 to position 6 means renumbering everything and holding the consequences in your head. Cards on a board don't resist it: you pick one up and put it somewhere else, and the shape re-forms in front of you.

If moving is expensive, you will convince yourself the order is fine. It usually isn't.

A working recipe

Write the premise paragraph. Rewrite until it has a cost in it.
List the turns. If you have fewer than five, you don't have a middle yet.
Outline three chapters. Draft them.
Repeat step 3 until something surprises you.
When it does, go back to step 2 and check the turns still stand.

That's it. The loop is the method. The outline is never finished, because it isn't a plan — it's your current best model of a thing that doesn't exist yet.

Narrale's planning canvas is built for exactly this loop: cards you drag rather than a list you renumber, chapter notes that sit beside the manuscript, and a story map for the threads that cross them. Plan at the altitude that helps. Draft at the altitude where the book is actually written.