The short answer
A premise becomes an outline when you lock three decisions: genre, total length, and chapter count. An 80,000-word thriller, for example, divides into ten chapters of about 8,000 words each, with the first act closing near word 20,000 and the second act closing near word 60,000. Each chapter then receives one purpose, one conflict, and one turn, which together tell you what to draft on any given day. The tool on this page performs exactly that conversion, and the rest of this guide explains how to read its output and write from it.
What this tool actually does
Many writers stall between the exciting sentence and the first chapter. The sentence feels alive, yet it contains no instructions about Tuesday morning, when the document is open and the cursor blinks. This generator closes that gap by treating your premise as raw material and returning a working plan: a restated logline, act boundaries expressed in words, and a chapter-by-chapter table where every row names the job that chapter must perform.
It is worth understanding what the tool does not do. It does not invent your characters, your setting, your dialogue, or your theme. It structures; you supply the story. The genre you select determines which set of chapter jobs the table draws from, and the length you enter determines how many chapters appear and how many words each one should carry. Nothing leaves your browser, and nothing here drafts prose for you. The artifact you receive is a scaffold with measurements, ready for a writer to build on.
This page also differs deliberately from a chapter-expanding tool. An expander assumes you already hold section summaries and want each one grown into scenes. This generator sits one step earlier in the funnel: you arrive holding only a premise, and you leave holding chapter count, word targets, and per-chapter assignments. Once those exist, an expander becomes useful. Before they exist, expansion has nothing firm to grip.
How to use the three inputs well
The premise field accepts one or two sentences, up to a generous limit. Write the strongest version you can: name a specific person, the concrete thing they must do, and the force blocking them. A retired smuggler, one last passenger, and a closed border give the downstream beats something to attach to, because every chapter job can be tested against that sentence. A vague wish about hope or healing gives the same machinery nothing to hold, and the resulting outline will feel hollow through no fault of the structure.
The genre menu selects the beat map. Thriller, romance, mystery, fantasy, and literary fiction each carry a different twelve-chapter sequence of purposes, conflicts, and turns. Choose the shelf where your finished book would sit, not the mood of your opening pages. A fantasy novel with a romance subplot still needs the fantasy map, because the quest spine determines chapter order while the romance colors scenes inside chapters.
The length field sets total words, from 40,000 to 150,000. Enter the honest target for your category rather than an aspirational number. Adult commercial fiction usually lands between 75,000 and 100,000 words, romance often runs shorter, and epic fantasy runs longer. The count you enter drives everything downstream, so a realistic number produces a usable plan while a fantasy number produces a plan you will abandon by chapter three.
How the chapter math works
The generator assigns chapter count from length in fixed bands. Books under 55,000 words receive 8 chapters. Books from 55,000 up to 69,999 receive 9 chapters. Books from 70,000 up to 84,999 receive 10 chapters. Books from 85,000 up to 99,999 receive 11 chapters. Books at 100,000 words and above receive 12 chapters. These bands keep average chapter length inside a writable range instead of letting short books splinter into fragments or long books compress into bricks.
Once the count is fixed, the total is divided evenly. The base target equals the total divided by chapters, rounded down, and any leftover words distribute one extra word to the earliest chapters. Readers never see this remainder handling, because at book scale it shifts a chapter by a single word at most. What you see is a clean column of targets, each within a word or two of the average, summing exactly to your entered total.
Act boundaries follow a stated craft convention rather than a discovered law. Act 1 is treated as ending near 25 percent of the total, and Act 2 as ending near 75 percent. The tool multiplies your total by those fractions and rounds to whole words. Each chapter is then labeled by the word where it starts: chapters beginning before the first boundary belong to Act 1, chapters beginning between the boundaries belong to Act 2, and chapters beginning after the second boundary belong to Act 3. Conventions vary across teachers, and some place the first break nearer 20 or 30 percent, but the function of a boundary stays constant: it tells you which chapters must set up, which must complicate, and which must resolve.
Worked numbers for the default input
Take the default case the page loads with: the smuggler premise, told as a thriller, at 80,000 words. The band table assigns 10 chapters, because 80,000 falls inside the 70,000 to 84,999 range. The average equals 8,000 words per chapter. The base division is exact here, so every chapter target reads 8,000 words and the column sums to 80,000.
The act boundaries land at word 20,000 and word 60,000. Chapter 1 starts at word 0, chapter 2 starts near word 8,000, and chapter 3 starts near word 16,000, so the first two and part of the third sit inside Act 1 depending on exact starts. Chapters beginning at or after word 20,000 carry the Act 2 label through the middle of the book, and chapters beginning at or after word 60,000 carry the Act 3 label. Concretely, with ten even chapters the starts fall at 0, 8,000, 16,000, 24,000, 32,000, 40,000, 48,000, 56,000, 64,000, and 72,000, which places three chapters in the first act, five across the long middle, and two in the closing act. That distribution matches how most commercial novels behave: a compact setup, an extended tangle, and a swift resolution.
The restated logline for this input reads as the premise sentence followed by a framing clause naming the genre, the rounded total, and the chapter count. Each of the ten rows then pulls the first ten beats from the thriller map, pairing each chapter with its purpose, its conflict, and its turn. Chapter 6, the midpoint, carries the passenger-identity reveal, which sits near word 40,000 — the structural center of the book. That placement is not mystical; it simply puts the largest reversal where the reader is most ready for the story to change direction.
Worked numbers for a second length
Change only the length to 120,000 words with a fantasy premise, and every downstream number moves. The band table assigns 12 chapters, because 120,000 meets the 100,000 threshold. The average equals 10,000 words per chapter. Act 1 ends near word 30,000 and Act 2 ends near word 90,000. Chapter starts fall every 10,000 words from 0 through 110,000, placing three chapters in the setup, six across the middle quest, and three in the closing battle and return.
Notice what length buys: the middle act now holds six full chapters instead of five, which is why long-book verdicts warn about sag. Six chapters of complication need six distinct turns, or the quest feels like one long corridor. The table answers this by giving each middle chapter a different job — wonder with cost, costly victory, scattering, an ancient place, darkest defeat, and rally — so the writer can check variety at a glance before drafting.
Drop instead to 45,000 words with a romance premise, and the engine assigns 8 chapters averaging 5,625 words. Act 1 ends near word 11,250 and Act 2 ends near word 33,750. Every chapter must now advance the relationship and the outer situation simultaneously, because there is no spare room for a chapter that only deepens mood. The warning verdict for short books says exactly this: merge any chapter whose turn merely repeats its neighbor, since brevity punishes redundancy first.
Reading the four columns of the table
Each row carries purpose, conflict, turn, and target. The purpose states the chapter's assignment in the larger arc: what this stretch of pages must accomplish for the book to hold together. Read it before drafting and again after, and cut any scene that serves a different assignment, however charming that scene feels in isolation.
The conflict names the resistance inside the chapter. It is deliberately smaller than the book-level obstacle from your premise. The border is closed for the whole novel; the checkpoint guard with sharp eyes is this chapter's problem. Keeping the two levels distinct prevents every chapter from restating the premise at full volume, which exhausts readers quickly.
The turn is the most important column. It describes the change the chapter must produce: a trust broken, a truth revealed, a plan inverted, a price paid. Draft toward it the way a cook works toward plating. If you reach the target word count without the turn happening, the chapter is unfinished regardless of how polished its sentences sound. If the turn happens early, you may close the chapter early and bank the saved words for a chapter that needs them.
The target column gives the budget. Treat it as a fence, not a prison. Landing within a few hundred words either way changes nothing structural. Landing 2,000 words over or under repeatedly will shift act boundaries and pacing, so track the running total every few chapters and adjust scene length rather than chapter count.
Illustrative hypothetical example one: the border crossing
Consider an illustrative hypothetical writer named Mara, invented for this guide, who pastes the smuggler premise, selects thriller, and enters 80,000 words. The generator returns ten chapters averaging 8,000 words with breaks near 20,000 and 60,000. Mara reads chapter 4, whose purpose covers pursuit closing in, whose conflict admits an ally lied about the passenger, and whose turn breaks trust mid-journey. She decides the ally is a dispatcher who claimed the passenger was a courier but who knew about a bounty. Chapter 4 will therefore open with the dispatcher missing a check-in, escalate through a roadside stop where papers nearly fail, and close with Mara's protagonist finding the bounty notice. The turn is concrete, filmable, and checkable: by the final page, trust is broken.
Mara then tests chapter 5 against chapter 4 before writing it. Chapter 5 demands a compromised safe house and a trade of rest for a night run. Because chapter 4 already broke trust, chapter 5 can exploit that breakage: the safe house was the dispatcher's address, so of course it is watched. The outline earns its keep here, because the turn of one row becomes the fuel of the next. Writers who skip this chaining step often produce chapters that each start from rest, which makes even eventful books feel oddly static.
At the midpoint Mara finds the identity reveal near word 40,000. She chooses to make the passenger a former border officer who closed the crossing herself. This choice inverts the delivery plan into a protection plan exactly as the beat requires, and it reframes every earlier checkpoint scene on reread. That reframing power is the practical test of a midpoint: if revealing it changes the meaning of act one, it belongs at the center; if it merely adds noise, it should move.
Illustrative hypothetical example two: the festival stall
Consider a second illustrative hypothetical writer named Dev, also invented for this guide, who enters a romance premise about a widowed baker forced to share a festival stall with an old rival, selecting romance at 72,000 words. The engine assigns 10 chapters averaging 7,200 words, with act boundaries near 18,000 and 54,000. Chapter 3 carries the agreement beat: an arrangement to keep meeting despite intrusion from a duty or rival. Dev decides the intrusion is the festival committee threatening to reassign the stall unless daily sales targets are met. The arrangement becomes a pact to hit the numbers together, which forces proximity with a measurable goal.
Chapter 6, the midpoint near word 36,000, demands a truth about the past that redefines the relationship. Dev assigns it the real story of the ruined opening day: the rival did not sabotage the baker but took the blame to shield her from a harsher truth about a supplier. Told badly, over coffee, it lands as an excuse; the fallout chapters then have somewhere painful and specific to sit. Chapter 7 becomes distance rather than generic gloom, because the baker must decide whether the revelation rewrites years of grief or merely complicates them.
Dev's chapter 9, the darkest separation, carries a failed apology. He drafts the apology scene first, lets it fail honestly — wrong words, wrong hour, an audience of market stalls — and only then writes the self-change that follows without witnesses. By the time chapter 11 arrives, the declaration under pressure has been earned by visible practice rather than asserted by narration. None of these decisions came from the tool; all of them were organized by it. That division of labor is the entire point.
Genre-specific guidance
Thriller and adventure outlines run on pressure and reversal. Keep chapters free of mood-only stretches: every row should move the pursuit, shrink the options, or raise the price. Guard the midpoint reveal carefully, because thrillers live or die on whether the center reframes the opening. If your chapter 6 turn could be swapped with chapter 4 without damage, escalate it until the swap becomes impossible. Short chapters with hard turns beat long chapters of atmosphere in this shelf, and the 8,000-word targets should feel like upper bounds rather than quotas to fill with scenery.
Romance outlines run on proximity, vulnerability, and repair. The outer plot — festival sales, a border, a competition — exists to force the pair together and then test what grew there. Protect the failed-apology beat in chapter 9 from premature forgiveness; readers trust reconciliations that survive a genuinely botched attempt. Keep the midpoint truth specific and costly rather than symbolic. A secret with a date, a name, and a consequence outperforms a vague admission of fear every time, and it gives the fallout chapters concrete material to process.
Mystery outlines run on questions, lies, and timelines. Treat the false solution in chapter 4 as load-bearing: it must be plausible enough that readers briefly accept it, so the detective's rejection teaches them how to read clues. The trusted-person implication at the midpoint should narrow the suspect circle emotionally even as it widens the case logically. Keep a private timeline document beside the outline, because mystery readers forgive many things but not clock errors. When the table says alibis are re-tested in chapter 10, actually re-test them against the hours you established in chapters 2 and 3.
Fantasy outlines run on wonder rationed against cost. The map gives twelve full beats for long books, including threshold crossing, price of power, costly victory, scattering, ancient place, and return. Spend wonder early enough to earn the reader's travel: chapters 2 through 5 should each contain one memorable strangeness. Then charge for it, because costless magic deflates quest stakes. The scattering beat in chapter 7 exists so the party's reunion carries weight; do not let secondary characters teleport conveniently across it. At 10,000-word chapter targets, plan each chapter as two or three scenes with a turn at the end, not one continuous march.
Literary fiction outlines run on attention and reckoning. The beats here are quieter — a snagged detail, a contradicted self-story, a named truth, a failed gesture — and they demand precise observation rather than event density. Do not mistake quiet for formless: chapter 6 still names the central truth plainly, and chapters 7 through 9 still process it through withdrawal, failure, and grief. The climactic meeting in chapter 11 should speak plainly rather than symbolically, since a book of careful attention earns one direct scene. Keep chapter targets modest and let white space do structural work between sections.
Edge cases worth handling before drafting
A thin premise is the most common edge case. The tool flags any premise under 30 characters as informational rather than ready, because a fragment like a thief runs cannot anchor ten chapters. The outline still renders — a scaffold is better than a blank page — but treat every purpose row as a placeholder. Before drafting, expand the fragment until it names a person, an action, and an obstacle. Rerun the tool with the fuller sentence and watch how differently the same table reads when the logline carries weight.
A sprawling premise is the opposite failure. Past about 45 words, the input usually contains two books, or a book plus its subplot wearing the spine's clothes. Symptoms include two villains, two romances, or a journey plus a home-front saga given equal billing. The warning verdict advises cutting to one sentence. A practical cut keeps the clause with the highest personal stakes and demotes the rest to subplot material for chapters 5 through 8, where the middle act has appetite for complication.
Very short books compress rather than simplify. At 40,000 to 50,000 words with 8 chapters, each chapter averages around 5,000 to 6,000 words and must multitask. A chapter cannot afford a pure travel sequence or a pure mood piece; movement and meaning must share pages. Plan scenes that do double duty, and merge ruthlessly when two adjacent turns resemble each other.
Very long books dilute rather than enrich by default. At 120,000 words and above with 12 chapters, the six-chapter middle can sag unless each turn differs sharply from its neighbors. Audit the middle four rows side by side: if two turns both read as chased but escaped, rewrite one into a betrayal, a price, or a revelation. Length is justified by variety, not by volume.
Genre mismatch is a quieter edge case. A thriller premise run through the literary map produces beautiful chapters with no pursuit, while a literary premise run through the thriller map produces chases nobody requested. If more than three rows feel wrong for your story, the input genre is probably wrong, not the beats. Switch the menu and compare the two tables before committing to either.
Failure modes even good outlines cannot prevent
The outline-as-cage failure arrives when a writer obeys purpose lines after the draft has discovered something better. Chapter 4 may reveal that the dispatcher deserves a viewpoint, or chapter 7 may surface a stronger midpoint than the planned one. When the draft teaches you something true, update the plan and continue. The table is a servant of the book, and a chapter that earns its turn by a different route has still earned it.
The sagging middle failure arrives around chapters 5 through 8 in almost every long draft. Early energy fades, the ending feels distant, and scenes start circling. Prevention beats cure: before drafting the middle, write each of those turns on a card and confirm no two cards match. During drafting, if a middle chapter bores you, it will bore readers faster, so cut or combine rather than pushing through on discipline alone.
The turn-skipping failure is the most expensive. A chapter reaches its word target with lovely sentences but no change, and the next chapter therefore starts from rest. One skipped turn is recoverable; three in a row collapses momentum. Make the turn the first thing you draft in each chapter, even in rough form, then build backward to its setup and forward to its fallout. The target count measures the container; the turn is the contents.
The premise-drift failure creeps in near chapter 8, when tempting subplots audition for the spine. Test every new idea against the logline: does it raise the price of this crossing, this courtship, this investigation, this quest, this reckoning? Material that passes belongs; material that merely entertains belongs in another book or the cutting file. Drift feels like abundance while drafting and reads as dilution later.
The word-count mirage failure confuses hitting targets with finishing chapters. Targets keep pacing honest across 80,000 words, but a chapter at exactly 8,000 words without its turn is not done, and a chapter at 6,500 words with a decisive turn may be complete. Use the numbers as fences around the work, then judge each chapter by whether its turn forces the next chapter to differ.
Turning each row into draftingday work
Convert the table into a schedule by treating every row as one work unit. A writer producing one chapter per week finishes a ten-chapter book in about ten weeks plus revision; a faster pace of two chapters per week compresses the draft into five. Either way, begin each unit by copying its purpose and turn onto a fresh page or card, then list the two or three scenes that will carry the chapter from its opening state to its turn. Draft the turn scene first in rough form, because everything before it is setup and everything after it is fallout.
Scene length follows naturally from chapter targets. An 8,000-word thriller chapter often splits into three scenes of roughly 2,500 to 3,000 words each: setup with pressure, complication with reversal, and the turn with its immediate cost. A 5,625-word short-book chapter may hold only two scenes, each carrying double duty. A 10,000-word fantasy chapter may hold three scenes plus a brief interstitial of travel or lore, placed where it delays but does not stall the turn.
Dialogue and description serve the row, not the reverse. Before keeping any exchange, ask whether it advances the purpose or sharpens the conflict; before keeping any vista, ask whether it raises the price or deepens the wonder the chapter owes. Material that serves neither can be cut without grief, because the outline tells you exactly what the chapter owes. Writers who outline this way often find revision faster: the cutting decisions were pre-authorized by the table.
Track the running total as you go. After chapter 3 of an 80,000-word book, you should stand near word 24,000; drifting 3,000 words over at that point compounds into a 10,000-word overrun by the end. Corrections are cheapest early: trim a scene, combine two conversations, or move a subplot beat into the chapter where it belongs. The act labels help here, since crossing into Act 2 before word 20,000 tells you the setup ran rich and the middle must tighten.
Word targets as pacing discipline
Targets encode pacing judgments that are hard to feel from inside a scene. A chapter that runs double its budget steals words from later chapters, forcing the climax into summary. A chapter that runs half its budget starves its turn, forcing readers to accept change without evidence. Neither sin is fatal once, but repeated imbalance reshapes the book: persistent overruns in Act 1 produce a magnificent setup attached to a rushed ending, which is the most common shape of abandoned drafts.
Use the average as your stride and the act boundaries as checkpoints. Reaching the Act 1 boundary more than a chapter early means setup is efficient and the middle can breathe; reaching it more than a chapter late means setup is indulgent and the middle must compress. Neither verdict requires rewriting the plan, only adjusting scene ambition in the chapters ahead. Numbers cannot tell you what to write, but they can tell you, early enough to matter, how much room remains to write it in.
Short-book writers should treat targets as ceilings, since every extra thousand words in a 45,000-word book is proportionally larger than in a long one. Long-book writers should treat them as floors in the middle act, since thin middle chapters at 6,000 words inside a 120,000-word plan signal skipped complications. In both cases the table plus the running total gives an honest mirror: not a grade, but a measurement.
Revising the outline after a first draft
No outline survives contact with a finished draft unchanged, and that is success, not failure. Once the draft exists, reread the table against what you actually wrote. Mark each turn as earned, fudged, or skipped. Earned turns need only polish. Fudged turns — changes asserted by narration rather than demonstrated in scenes — need one added scene of evidence each. Skipped turns need either a new scene or a deliberate decision to merge that chapter with its neighbor, reducing the count by one and redistributing its words.
Then check the act boundaries against the draft's real shape. If the story's true point of no return lands at word 30,000 instead of 20,000, the setup ran long; consider moving an early complication forward or cutting a throat-clearing chapter. If the climax starts before the Act 3 label, the middle sagged; consider promoting a subplot turn into the gap. The outline becomes a diagnostic instrument in revision, comparing the planned skeleton with the grown body and naming exactly where they diverge.
Finally, revisit the premise sentence itself. Drafts often discover a sharper logline than the one they started from: the passenger matters more than the border, or the stall matters less than the supplier's lie. Rewrite the premise to match the best version of the book, rerun the generator, and compare the fresh table with your draft chapter by chapter. Where they agree, confidence rises; where they differ, you hold a concrete choice rather than a vague unease.
Questions writers ask at this stage
How many chapters should a first novel carry? The band table answers with ranges rather than rules: eight for short books, ten around 80,000 words, twelve past 100,000. More important than the number is the fit between count and material. A story with three genuine reversals stretches thin across twelve chapters, while a quest with six distinct trials crams uncomfortably into eight. Count your turns before choosing your length, or choose your length and then ensure you own enough distinct turns to fill it.
Are act breaks fixed at 25 and 75 percent? They are conventions stated as such, not measurements of nature. Teachers differ, placing the first break anywhere from a fifth to a third of the way through. What matters is the underlying function: setup must end before readers tire of orientation, and resolution must begin with enough pages left to earn its weight. This tool uses 25 and 75 so every output is comparable, and the chapter labels follow arithmetically from those lines.
Can a weak premise still produce a useful outline? Yes, with a flag. The thin-premise verdict renders the full table but labels it a scaffold, advising you to add want and obstacle before drafting. Many writers find this scaffold genuinely helpful, because blank rows with clear jobs make the missing specifics visible. The danger is mistaking the scaffold for the story and drafting ten chapters of purposeful vagueness, so strengthen the sentence first.
What happens if I switch genres mid-project? The chapter count and targets stay identical at the same length, but every purpose, conflict, and turn changes, because each genre draws from its own beat map. Comparing two genre tables for the same premise is a legitimate diagnostic: the map whose rows feel inevitable is usually the book you are actually writing. Commit to one map per draft; blending beat maps row by row produces tonal whiplash.
Should each chapter hold one viewpoint or timeline? The table is neutral on that choice, and either approach fits inside the targets. Single-viewpoint chapters make turns easier to track, since one consciousness registers each change. Multi-viewpoint chapters can deliver bigger turns but must still produce exactly one decisive change per chapter, or the row's job splits and neither thread earns its weight. Choose per book, announce the pattern early, and keep it.
What comes after the outline? Expansion and drafting. Each row becomes a chapter plan with scenes, each scene becomes pages, and the running total keeps pacing honest. When a chapter's turn proves stronger in a different position, move the row rather than forcing the draft to obey stale order. The outline is a living index, consulted daily and revised monthly, until the manuscript stands on its own and the scaffold can fall away.
Where focused drafting happens
Once the table exists, the real work moves to sustained drafting sessions where each chapter grows from its row into finished pages. That writing belongs in a dedicated drafting space that keeps the outline visible and the manuscript central. Co-Writer at the /write route is where to draft the chapters this tool plans, carrying each purpose, conflict, and turn from the table into prose. Open the outline beside the draft, work one chapter target at a time, and let the turns accumulate into a book.
A closing checklist before chapter one
Read the logline aloud and confirm it still excites you; ten chapters is a long companionship with a sentence. Confirm the genre map matches the shelf you aim for, not merely the opening mood. Confirm the chapter count fits your genuine reversals, merging or splitting rows until each turn differs from its neighbors. Confirm the middle four turns vary sharply, since the middle is where drafts go to stall. Set a drafting pace in chapters per week, schedule the turn-first habit for each unit, and track the running word total at every third chapter. Then begin, remembering the division the whole page insists on: the tool structures, and you supply the story.