The short answer
An 80,000-word triumph ending with five open threads should start its climax near word 68,000, reserve about 4,000 words for resolution, and close every thread on a payoff ledger with a named owner, a visible outcome, and a scene slot. That placement is not a rule about art. It is a budget. The final 15 percent of a novel is where readers stop collecting questions and start cashing them in, and each open thread needs stage time to convert. Twist endings need two or more early plants to feel fair, because a reveal that re-reads earlier scenes satisfies while a reveal that replaces them annoys. More than seven threads risks a rushed finish, since the climax words divided seven ways leave each payoff closer to a mention than a scene.
The planner on this page turns those budgets into numbers for your draft. Enter total words, ending type, thread count, and genre. It returns a climax start position, a resolution word budget, a per-thread payoff allowance, a ledger with owner and timing slots for every thread, and flags for the risks that sink endings most often. The scores are deterministic heuristic checks against stated craft conventions, not measurements of quality or predictions of sales. They tell you whether the structure gives each promise room to land.
What a satisfying ending actually is
Readers call many different feelings satisfying, but the mechanics underneath are consistent. A satisfying ending closes the loops the book opened, in proportion to the emphasis each loop received, at the moment of maximum pressure, with a cost that proves the outcome mattered. Each clause in that sentence does work, so take them one at a time.
Closing the loops means every setup the story emphasized gets a payoff the reader can point to. A setup is any promise: a missing sister, a loaded debt, a forbidden door, a rivalry, a romance spark, a prophecy. If the narrative spent pages making the reader care, the ending must spend at least a scene showing the result. Threads the book mentioned once in passing can fade. Threads with chapters of buildup cannot.
In proportion to emphasis means the biggest promise gets the biggest payoff slot. The central question of the novel — will she clear her name, will they survive the crossing, will the town accept the truth — should close last and largest. Subplots close earlier and smaller. When a minor thread eats the final scene while the main question got a paragraph, readers feel the imbalance even if they cannot name it.
At maximum pressure means payoffs land during the climax, not before and not in summary after. A villain defeated in chapter nine while the final three chapters cover paperwork wastes the structure. The climax is the stretch where opposing forces collide with no exit left, and each thread should resolve inside that collision or in its immediate aftermath.
With a cost means victory, reconciliation, or escape should require something the character would rather not pay. The cost can be a relationship, an identity, an illusion, a limb, a fortune, a future. Costless wins feel weightless because nothing was risked. Even gentle genres benefit: the cozy mystery solver risks a friendship to name the culprit, the romance lead risks pride to speak first.
How the tool math works
The planner uses three inputs to produce three budgets. First it sets the climax start as a percentage of total words. Triumph uses 85 percent, bittersweet 86, tragedy 87, twist 88, and open 85. Then it sets the resolution budget as a second percentage: 5 percent for triumph, 6 for bittersweet, 4 for tragedy, 5 for twist, 7 for open. Whatever sits between the climax start and the resolution start is the climax itself, and dividing that by the thread count gives payoff room per thread.
Walk through the default case with worked numbers. An 80,000-word novel with a triumph ending and five threads: 85 percent of 80,000 is 68,000, so the climax starts near word 68,000. Five percent of 80,000 is 4,000, so the last 4,000 words are resolution, running from word 76,000 to word 80,000. The climax itself spans 68,000 to 76,000, which is 8,000 words. Divide 8,000 by five threads and each thread holds about 1,600 words of payoff room — a full scene per thread. Divide the 4,000 resolution words by five and each thread gets about 800 words of aftermath. Both numbers sit above the planner danger lines, so the verdict passes.
Change one variable and watch the budget move. Keep 80,000 words but switch to a twist ending with the same five threads. The climax percentage rises to 88, so the climax starts near word 70,400. Resolution stays 5 percent, or 4,000 words, from 76,000 to 80,000. The climax now spans 70,400 to 76,000, which is 5,600 words, and per-thread room falls to 1,120 words. Still workable, but tighter, which is honest: twist structures spend middle words on misdirection, so the closing collision compresses. The planner also raises the foreshadowing checklist for twist inputs, because the shorter climax must do double duty as reveal and emotional landing.
Change the thread count instead. An 80,000-word tragedy with ten threads: climax at 87 percent is word 69,600, resolution at 4 percent is 3,200 words from 76,800 to 80,000, climax spans 7,200 words, and per-thread room is 720 words while resolution per thread is 320. The planner fails this input. Seven hundred words can stage a beat but not a scene, and 320 words of aftermath is a paragraph. The prescription is concrete: close at least five threads before the climax or add resolution words until each surviving thread holds 500.
A second set of worked numbers
Take a 100,000-word fantasy novel with a bittersweet ending and six threads. Climax at 86 percent lands near word 86,000. Resolution at 6 percent is 6,000 words, from 94,000 to 100,000. The climax spans 8,000 words, so each of six threads holds about 1,333 words. Resolution per thread is 1,000 words. This is a comfortable budget for the genre, where bittersweet closings need aftermath space for grief alongside victory. The planner passes it and suggests ordering the six payoffs from smallest to largest, with the cost beat landing last.
Now take a 60,000-word romance with an open ending and four threads. Climax at 85 percent is word 51,000. Resolution at 7 percent is 4,200 words, from 55,800 to 60,000. The climax spans 4,800 words, giving 1,200 per thread. One of the four threads stays open as the sequel hook, so in practice three threads share the climax at 1,600 words each while the hook thread gets a partial-progress beat and deferral. The planner warns on all open endings regardless, because the form carries inherent risk, and instructs the writer to close one full arc inside the climax so the current book feels complete.
Notice what the percentages assume. They assume the draft is structurally complete rather than episodic: that words before the climax build toward it instead of wandering. A picaresque middle that introduces new threads at 70 percent will break any ending budget, because threads opened late have no buildup to pay off. If your middle keeps spawning promises, fix the middle before tuning the end.
Illustrative hypothetical example: the harbor mystery
This first example is illustrative and hypothetical, invented to show the ledger method. Imagine a writer finishing an 82,000-word mystery with a triumph ending and five threads: the missing brother, the smuggled cargo, the detective debt to a retired cop, the strained marriage, and the rookie partner proving herself. The planner sets the climax near word 69,700, with about 4,100 words of resolution. Climax room per thread is roughly 1,600 words.
The writer assigns owners and beats. The cargo thread goes to the antagonist as a dockside confrontation near word 71,000. The debt thread goes to the mentor figure as an irreversible choice near word 72,500: the detective burns a favor to get the warrant. The rookie thread gets a hidden-information reveal near word 73,800 when she spots the doctored manifest. The marriage thread gets a cost beat near word 75,000 when the detective misses the ferry home to hold the scene. The brother thread, the central promise, closes last inside the resolution opening: a reunion scene that the previous four payoffs made possible. Resolution then spends its 4,100 words on the landing (the drive home) and the proof (a small final case detail showing the detective changed methods). Every promise the cover copy implied gets a scene a reader could summarize.
Illustrative hypothetical example: the twist that earns it
This second example is also illustrative and hypothetical. Imagine a 90,000-word thriller with a twist ending and four threads: the unreliable narrator account of a trial, a vanished witness, a bribed juror, and a buried recording. The planner sets the climax near word 79,200 with 4,500 words of resolution, giving about 1,125 words per thread, and raises the foreshadowing checklist because two threads would have flagged thin plants.
The writer audits the plants. The recording mechanism appears in chapter three as a background detail (a courtroom dictaphone), again near the midpoint as a disputed exhibit, and once more around word 40,000 when a clerk mentions missing tapes. The juror thread resolves straight: confrontation and arrest, no trick. The witness thread resolves straight: she testifies, plainly. The narrator thread carries the reframe: the account the reader trusted was the edited tape, not the event. Because three threads closed honestly, the fourth can reframe them without collapsing trust. The 4,500-word resolution holds one scene of shock management (characters reacting to the reframe) and one scene of consequence (the retrial motion). Without those two scenes the twist would read as a prank. With them it reads as a second ending nested inside the first.
The payoff ledger: who, what, and when
The ledger is the center of the method. Each open thread gets one row with three assignments. Who carries the payoff names the character whose agency closes it. What closes it names the visible beat: confrontation, choice, reveal, sacrifice, or reckoning. When slots it at a word count inside the climax or early resolution. A thread with all three assigned is plannable. A thread missing any of the three is a risk.
Assign owners first, and spread them. If the protagonist personally closes all six threads, the supporting cast exists only as furniture. Let the ally close one thread, the mentor force one choice, the rival concede one point. Assign beat types second, and vary them: a fight, then a confession, then a discovery, then a payment, then a reunion. A reveal works best early in the climax, because later beats can react to it. A cost beat works best late, because it converts victory into meaning right before the resolution.
Assign timing third, and order by importance ascending. Close the smallest thread first and the central promise last. Readers forgive a quiet early payoff inside a loud climax, but they do not forgive a loud minor payoff stepping on the main one. As a practical split, place the first half of threads in the climax front half and the rest in the late climax shading into resolution. The tool when column does exactly this, spacing slots evenly across the climax span.
Ending types and what each demands
Triumph needs a clean win plus proof. The climax defeats the opposition decisively, and the resolution shows the win holding under a small final pressure: the repaired bridge carries weight, the town keeps the reform after the hero leaves. Without the proof scene, triumph feels asserted rather than demonstrated.
Bittersweet needs the win and the cost in the same frame. The medal and the empty chair. The wedding and the farewell. The planner gives bittersweet the largest resolution share, 6 percent, because grief needs stage time beside celebration. Draft two aftermath scenes minimum and resist merging them: one scene for what was gained, one for what was paid. Merged scenes blur both feelings.
Tragedy needs inevitability, not surprise. The reader should reach the final pages feeling the loss could not have gone otherwise given who these people are. That requires a visible turning point upstream of the climax: the message not sent, the bridge crossed, the lie defended once too often. Place that point before the climax start the tool returns, then let the climax execute what the turning point decided. Endings that spring tragedy in the last chapter without a turning point read as accidents, and accidents do not satisfy.
Twist needs fairness plus aftermath. Fairness means plants a re-reader could find: the mechanism shown twice early and once mid-book. Aftermath means scenes after the reveal where characters and readers absorb the new frame. The planner checklist enforces both. A frequent error is spending the whole resolution on explaining the trick. Explanation is necessary but not sufficient; consequence is what converts surprise into satisfaction.
Open needs one finished arc plus one honest hook. The finished arc proves this book was worth reading alone. The hook earns the next purchase without punishing this one. Give the hook visible partial progress — the map found but not followed, the message received but not answered — so the deferral feels like momentum rather than withholding. Never leave two major arcs open. One hook is a promise. Two is an apology.
Foreshadowing fairness for twist endings
Writers often ask how much planting a twist requires, and the honest answer is countable. Two early plants establish the mechanism exists. One mid-book reminder keeps it retrievable without spotlighting it. One late confirmation inside the climax lets the reveal feel discovered rather than delivered. That is four touches total, with at most one of them emphatic.
Vary the channel of each plant so no single scene carries the whole load. A prop in chapter two, a line of dialogue near the midpoint, a behavior that seems like characterization until the reframe. Readers who catch one plant feel clever; readers who catch none can still verify the plants on re-read, which is where fairness lives.
Edge cases and failure modes
Too many threads is the most common failure. Ten threads in 80,000 words cannot each hold a scene, and the planner fails the input on purpose. The fixes in order of preference: merge two minor threads into one shared closing scene, close one thread before the climax so the ledger shrinks, or cut a thread whose promise the book never emphasized. Adding words is the last resort, because longer rushed endings are still rushed. Five threads is the comfort zone; seven is the ceiling for most commercial structures.
Too little resolution is the second failure. Drafts that end two pages after the final confrontation leave no room for the landing or the proof. Symptoms include epilogues that summarize emotions instead of staging them, and final lines that announce themes. The repair is mechanical: protect the resolution budget the tool returns and forbid new complications inside it. Resolution scenes should contain only aftermath: reactions, reunions, burials, departures, small demonstrations of change.
The late-opened thread is the third failure. A setup introduced past 70 percent cannot accumulate the emphasis a payoff requires, so closing it in the climax feels perfunctory and leaving it open feels like a hole. Either promote it earlier with at least two buildup scenes or demote it to background color that needs no payoff. The final pages stage an image or reflection while concrete threads hang: the debt unpaid, the sister unfound, the trial unheld. Imagery can carry the last paragraph but not the last act. Close the concrete promises first, then earn the reflective final beat. The planner ledger helps here by forcing every thread to name a visible closing beat before any thematic material is drafted.
Genre-specific guidance
Mystery and crime readers grade endings on explanation completeness. Every clue the detective used must be shown fairly, the culprit reveal must re-read earlier scenes coherently, and the motive must be proportionate to the crime. Budget one full reveal scene plus one motive scene; merging them rushes the motive. Cozy mysteries can close the community thread (the street, the shop, the family) inside resolution as the proof-of-change. Hardboiled mysteries should let the cost land on the detective personally.
Romance readers grade endings on the couple earned state. The black-moment separation must be resolved by changed behavior, not by new information alone. Draft the grand-gesture or honest-conversation scene as the central payoff, then a proof scene showing the couple handling a small conflict with new tools. For open or series romance, the current couple must reach a complete committed beat; the hook belongs to the next couple, never to the current relationship status.
Fantasy and science fiction endings carry system debts: magic rules, technology limits, prophecies, costs. The climax must obey the established system visibly, because readers track rules as promises. A power that solves the finale without prior cost or limit reads as cheating regardless of word budget. Epic lengths above 120,000 words scale all budgets linearly, but keep thread counts near six; longer books tempt thread sprawl that length alone does not fund.
Thriller and suspense endings are graded on pace and reframe control. The climax should be the fastest section of the book, so per-thread beats must be leaner and more kinetic: chases, breaches, live decisions. If running a twist, resolve action threads straight and fast, then spend resolution words on the reframe consequence rather than more action. A final quiet scene after sustained velocity lands harder than one more explosion.
Literary fiction endings are graded on resonance, but resonance still needs closed loops. Identify which threads are concrete (money owed, person missing, decision pending) and close each concretely before the final reflective passage. The reflection then summarizes a demonstrated change rather than substituting for one.
Resolution pacing beat by beat
A workable resolution has exactly two movements. First the landing: an emotional scene at reduced intensity where characters register what happened. Pace it slow. Let characters sit, grieve, laugh, eat, drive home. Second the proof: a small-pressure scene demonstrating the change holds. The detective declines the old shortcut. The couple navigates a disagreement cleanly. The survivor refuses the second dare. Pace it at medium intensity with one minor obstacle, not a new antagonist.
Common padding to cut from resolution includes new subplots, surprise arrivals, extended travel, and theme speeches. Each steals words from the two movements that matter. If the resolution budget exceeds 6,000 words, consider whether the climax actually ended where you think; long resolutions often conceal an unfinished climax whose final payoff leaked past the boundary.
Questions writers ask next
Do all subplots need payoffs? No, but all emphasized setups do. A single-sentence mention creates almost no expectation. A three-scene buildup creates a debt. Walk the manuscript and grade each thread by page emphasis; anything with more than a few pages of buildup needs a ledger row. When in doubt, either pay it off or trim its buildup in revision so the promise matches the delivery.
Can the climax start later than 90 percent? Occasionally, in books whose pleasure is an extended aftermath: family sagas, travel epics, aftermath-centered literary novels. The risk is a compressed collision that summarizes confrontations. If the climax starts past 90 percent, keep thread counts at three or fewer and give each a compressed but staged beat. The planner percentages will flag the tightness; treat the flag as information about trade-offs, not a prohibition.
Should every thread close on the page? Nearly. Off-page implied closures work only for minor threads whose outcome is fully determined by a shown event: the eviction stops because the case collapsed on the page. Major threads closed by implication read as abandoned.
What if beta readers report feeling rushed but the budgets pass? Trust the readers and audit scene density. Passing budgets mean room exists, not that the draft uses it. Check whether payoffs share scenes that should be separate, whether closings happen in narration rather than dramatized beats, and whether the resolution spends words on new material. Reallocate within the budget before expanding it.
Drafting the artifact in Co-Writer
The ledger this tool outputs is a planning document, and it becomes useful when it turns into scenes. Draft each ledger row as a full scene in Co-Writer, using the slash command /write with the beat type and owner named in the prompt, so every planned payoff becomes playable prose before revision. Work the rows in the ledger when order, smallest thread first and the central promise last, which keeps the finale building instead of peaking early. Once the scenes exist, read the resolution aloud inside the same project and trim any sentence that introduces a new complication, because the closing pages belong to aftermath and proof.
A revision pass for the final act
When the draft is complete, run one dedicated pass over the final 15 percent with the ledger beside the manuscript. Check each row: is the owner the acting agent, is the beat visible on the page rather than summarized, does the timing fall inside the climax span. Mark any payoff delivered in narration for expansion into scene, and mark any thread closed twice — once in dialogue and once in action — for compression into the stronger single delivery.
Then read only the closing pages and list every promise they appear to answer. Compare that list against the ledger. Promises answered that were never set up indicate climax clutter to cut. Ledger rows unanswered indicate either a missing beat to draft or buildup to trim upstream. This two-direction check is the practical meaning of satisfaction: everything promised is delivered, and everything delivered was promised. The planner budgets give each delivery its room; the revision pass confirms each room was used.