Mind Map for Writers: From Premise to First Draft, Layer by Layer

A story has layers, and they don't arrive in order. The premise comes in a flash on a walk; the structure shows up weeks later when you finally see where the midpoint sits; a character's wound only surfaces when you draft a scene and realize you don't know why she lies. Trying to hold all of that in a linear document means you keep scrolling back up to fix Act 1 every time Act 3 teaches you something. A mind map lets each layer live on its own branch and stay visible at once. This guide walks the layers a writer actually builds, in roughly the order they firm up: the premise and logline that compress the whole book into a sentence, the structural beats you borrow from a real framework, the character bible, the theme map that ties scenes to meaning, and finally the move that turns all of it into prose. The example map on this page is one short novel built this way. Most of the advice is aimed at fiction because fiction has the most moving parts, but the non-fiction case appears at the end and is, if anything, simpler.

Want to build one? Open ModuMind free — no signup needed.

One map, three views

The example below is the same datashown in ModuMind’s three modes. The map is rendered by the actual product layout engine, not a mock-up — switch a real map between map, outline, and table without re-entering anything.

Map viewVisual branches

Outline viewIndented text

The Last Library (heist novel)

Logline
- An archivist learns the last library is a forgery and must expose the truth or protect the lie
Characters
- Mara - want: prove archive is real / need: accept truth's cost / flaw: trusts documents over people
- The Curator - antagonist hiding the original fire
- Soren - mentor who lies to protect her
Theme
- Question: is truth worth more than what the lie protects?
- Motifs: fire, copies vs originals, locked rooms
Act 1 - Catalyst (~10%)
- Mara finds the locked archive
Act 2 - Midpoint & All Is Lost
- Midpoint: the archive is a forgery
- All Is Lost: the mentor's lie unravels
Act 3 - Finale
- Mara burns the forgery to save the truth

Table viewStructured rows

Branch	Item
Logline	An archivist learns the last library is a forgery and must expose the truth or protect the lie
Characters	Mara - want: prove archive is real / need: accept truth's cost / flaw: trusts documents over people
	The Curator - antagonist hiding the original fire
	Soren - mentor who lies to protect her
Theme	Question: is truth worth more than what the lie protects?
	Motifs: fire, copies vs originals, locked rooms
Act 1 - Catalyst (~10%)	Mara finds the locked archive
Act 2 - Midpoint & All Is Lost	Midpoint: the archive is a forgery
	All Is Lost: the mentor's lie unravels
Act 3 - Finale	Mara burns the forgery to save the truth

ModuMind로 그린 스토리 마인드맵 — 단편 소설 '마지막 도서관'을 레이어별로 구성: 로그라인, Save the Cat 비트, 캐릭터 바이블, 테마 맵. 아웃라인 뷰로 바로 전환 가능.

Build your own — free

Open the editor, no account needed. Switch between map, outline, and table instantly.

Try it free

1. Start with the premise and a one-sentence logline

Before any structure, you need to know what the story is about in a way you can say out loud. Put a working title in the center of the map and give it two first branches: 'Premise' and 'Logline'. The premise is the engine: the situation and the question it raises. 'The last analog library is about to be destroyed, and the archivist who runs it discovers the archive she's protecting is a forgery.' The logline compresses that further into the shape a reader or agent would recognize, naming a protagonist, a goal, an obstacle, and the stakes in one sentence. 'When an archivist learns the world's last library is a fake, she must decide whether to expose the truth or protect the lie that keeps it standing.' Writing the logline as its own node early does something useful and slightly uncomfortable: if you can't fill it in, you don't have a story yet, you have a setting, and it's better to know that on day one than on page 200. Keep both nodes pinned near the root. Every later decision, whether to add a beat or whether a scene earns its place, gets checked against that one sentence, and having it on screen rather than in your head keeps the book from quietly drifting away from the idea that excited you. The premise also tells you which framework in the next step will fit: a story built on a single irreversible choice wants tight act structure, while a sprawling quest wants the hero's journey.

2. Lay structure with a real framework, not vibes

Structure is where mapping pays off most, because a framework is already a tree. Pick one that fits the story and make each beat a branch. Blake Snyder's Save the Cat beat sheet is fifteen beats and maps cleanly to a map: Opening Image, Theme Stated, Set-Up, Catalyst, Debate, Break Into Two, B Story, Fun and Games, Midpoint, Bad Guys Close In, All Is Lost, Dark Night of the Soul, Break Into Three, Finale, Final Image. What makes it more than a list is that the beats have positions. The Catalyst lands around the 10% mark, Break Into Two near 20%, the Midpoint at the center, All Is Lost around 75%, and Break Into Three near 80%, so the beats double as a rough page budget. If you'd rather use the three-act model, the load-bearing nodes are the two turning points: the first turning point closes Act 1 and forces the protagonist into the new situation, and the second closes Act 2 and launches the climax. The hero's journey gives you twelve stages instead, running Ordinary World, Call to Adventure, Refusal, Meeting the Mentor, Crossing the Threshold, Tests and Allies, Approach, the Ordeal, Reward, the Road Back, Resurrection, and Return with the Elixir. The three frameworks rhyme more than they differ: the Catalyst, the first turning point, and the Call to Adventure are the same hinge under three names, as are All Is Lost, the second turning point's low, and the Ordeal. Whichever you choose, hang the concrete scenes as leaves under each beat: 'Mara finds the locked archive' under Catalyst, 'The mentor lies about the fire' under Bad Guys Close In. Now the map shows pacing you can't feel in a synopsis. A beat branch crammed with eight scene leaves is a stretch that will drag; a beat with none is a hole, and a hole sitting at the Midpoint is the one most likely to collapse a draft. Because the beats are fixed and the scenes are draggable, you reorder by moving leaves between branches instead of rewriting a single line.

3. Build a character bible as its own branch

Plot is what happens; character is why it lands. Give every major character a branch and hang four sub-nodes under each: want, need, flaw, and relationships. The want is the external goal the character will chase: Mara wants to prove the archive is real. The need is the internal lesson underneath it, often in direct tension with the want, so that she needs to accept that truth can cost more than the lie protects even as she chases proof. The flaw is the wound or blind spot that drives the bad decisions. She trusts documents over people, which is exactly why a forgery can fool her. Relationships are the lines to other characters: who she protects, who she's lying to, who can hurt her. Building this as a branch instead of a separate document means you can draw a relationship line from Mara's flaw straight to a plot beat and see the cause and effect on screen rather than carrying it in your head. It also surfaces the most common structural problem in a first novel: a protagonist whose internal need never connects to the external plot. If you can't trace a line from a character's need to the climax beat, the arc isn't wired up yet, and the map makes that absence visible in a way a character questionnaire never does. Antagonists deserve the same four sub-nodes, because a villain with a coherent want and a defensible logic is the one who makes the protagonist's choices hard. A useful test the map encourages: give the antagonist a node for the argument they'd make in their own defense. If it's persuasive, the conflict has real weight; if it's just 'evil', the climax will feel unearned no matter how loud it is. Supporting characters can stay thinner, a single node each, but the map keeps them in the same field of view so you notice when one of them is doing two jobs and should be split, or when two of them overlap and should be merged.

4. Map theme and motifs, then wire them to scenes

Theme is the layer writers either skip or overthink, and a map handles it better than prose notes because theme is relational, not sequential. Make 'Theme' a branch and state it as a question the story argues both sides of: 'Is the truth worth more than the thing the lie is protecting?' Under it, list the motifs that carry the question physically through the book: fire, locked rooms, copies versus originals, the smell of old paper. The useful step most people skip is the wiring: draw a relationship line from each theme or motif node to the specific scenes that dramatize it. Mara's flaw, the forged archive, and the final fire all connect back to the same theme node, which is how you know the ending earns its meaning rather than just resolving the plot. The diagnostic falls out of the structure for free. A scene that connects to no theme node and advances no beat is almost certainly filler; either cut it or find the thread it should carry. Keep this branch collapsed while you draft so it doesn't clutter the structural view, and expand it when a scene feels flat and you need to remember what it's supposed to mean. Motifs reward the same wiring as theme. Once 'fire' is a node linked to its scenes, you can see at a glance whether the image recurs often enough to register or appears once and is forgotten, and whether its meaning shifts across the book the way a good motif should: fire as destruction early, fire as cleansing at the climax. The map turns a vague intention to 'use fire as a symbol' into a checkable structure you can actually verify against the scenes you've drafted.

5. Switch to outline view and write the draft

Every layer so far has been preparation. The draft is the point, and this is the step where mapping in a tool built for writing beats mapping in a diagram app. In ModuMind the nodes you arranged spatially are the same underlying data as lines in an outline, so once the structure holds, you switch to outline view and the map collapses into an indented text skeleton in reading order: beat branches become headings, scene leaves become bullets. You write prose directly under each bullet, one scene at a time, with the structure already decided and the next beat sitting right below the cursor. When pacing feels off mid-draft, flip back to map view to check which beat is overstuffed, re-slot a scene by dragging, then return to outline view and keep going. For the management layer, covering chapter status, word-count targets, and deadlines, the same nodes open in table view as a tracker, with each scene a row you can mark drafted or revised. One body of material, three ways to look at it: a map to think in, an outline to write in, a table to track in, with nothing retyped between them. That continuity is the real argument for starting in a map instead of a blank document, because the structure you reasoned through never has to be rebuilt to start writing.

6. The non-fiction case: essays, blog posts, and books

Non-fiction skips the character and theme layers and leans entirely on structure, which makes it the quicker build. Put the working headline in the center and branch into the moves of the argument: hook, problem, framework, evidence, counterargument, call to action. Leaves hold the specifics: the exact statistic, the anecdote, the objection you need to answer before a reader raises it. The hierarchy enforces a kind of honesty. An empty 'evidence' branch is a claim you haven't backed yet, found before you publish rather than in the comments, and an oversized 'framework' branch usually means you're explaining your method instead of making your point. For a book or a long report, nest one level deeper: chapter branches, section sub-branches, key-point leaves. The premise-and-logline step from section one still applies. A non-fiction logline is the one-sentence promise the piece makes to the reader, and a piece that can't state it usually wanders. As with fiction, the payoff is the switch to outline view: every branch is a heading, every leaf a paragraph's worth of point, ready to draft into without retyping a word.

Tips

•If you can't write the logline node, you have a setting, not a story. Force the one sentence (protagonist, goal, obstacle, stakes) before you add a single beat, and pin it near the root so every later decision answers to it.
•Use a known framework as your branch labels rather than inventing beats. Save the Cat's fifteen beats or the three-act turning points give you positions, not just names, so an empty or overstuffed branch immediately reads as a pacing problem.
•Under each character, force all four of want, need, flaw, and relationships. The slot that's hardest to fill is usually the need, and a missing need is why a protagonist can feel flat no matter how busy the plot is.
•Draw a relationship line from each character's need to the climax beat. If the line can't be drawn, the internal arc and the external plot aren't connected yet, which is the most common fixable problem in a first draft.
•Write scene leaves verb-first: 'Mara burns the forgery', not 'climax scene'. Verb-led nodes already read like outline bullets and survive the switch to outline view as draftable lines.
•Keep theme and character branches collapsed while plotting and expand them on demand. Same map, so the reference is one click away, but the structural view stays clean enough to judge pacing.
•When you cut a scene, drag it to a collapsed 'Holding' branch instead of deleting it. A surprising number return when a later beat needs filling, and you keep the original wording.

FAQ

Which story framework should I map with — Save the Cat, three-act, or the hero's journey?

Map with whichever you can already feel in the story, and don't switch mid-project. Save the Cat's fifteen beats are the most prescriptive and give you positions (Catalyst near 10%, Midpoint at the center, All Is Lost around 75%), which makes pacing problems obvious on a map. The three-act model is looser and hangs on two turning points, good when you don't want a beat-by-beat cage. The hero's journey suits myth-shaped and quest stories. All three are just different sets of branch labels over the same map; pick one set, hang your scenes under it, and the spatial layout does the same diagnostic work regardless.

How do I turn the finished map into something I can actually draft from?

Build the map so branches are your structural beats and leaves are individual scenes written as short verb-led summaries. When the structure holds, switch ModuMind to outline view: every branch becomes a heading and every leaf becomes a bullet, in reading order, giving you an indented outline. You then write prose directly under each bullet, one scene at a time, with no retyping and the next beat already in front of you. Flip back to map view whenever you need to check pacing or move a scene.

Where do character and theme fit if the plot is on the same map?

On their own branches off the root, kept collapsed while you plot. Each character gets a branch with want, need, flaw, and relationships as sub-nodes; theme gets a branch stated as a question with motifs and scene links under it. Keeping them in the same document lets you draw relationship lines between a character's flaw and a plot beat, or between a theme and the scenes that dramatize it. Those connections catch arc and meaning problems you'd miss across separate files. Expand a branch only when a scene needs it.

Does this work for non-fiction and blog posts, or only novels?

It works for both, and non-fiction is the faster build because you skip the character and theme layers. Put the headline in the center and branch into the moves of the argument (hook, problem, framework, evidence, counterargument, call to action) with the specific statistic or anecdote as leaves. An empty branch is a gap you've caught before publishing. The premise step still applies: a non-fiction logline is the one-sentence promise the piece makes, and the map converts to a heading-and-bullet outline you draft into the same way.

Related guides

Brainstorming Mind Map: From Wild Ideas to a Picked Next Step

A brainstorming mind map runs in two acts: diverge wide, then converge hard. This guide pairs real techniques (SCAMPER, brainwriting 6-3-5, 'How might we', affinity clustering, impact/effort) with a worked campus-coffee example you can copy in ModuMind.

Open the editor, map your ideas, then switch to outline or table view. Saves to your own drive. No credit card.

Build your own — freeNo account needed to start.