The direct answer: how you raise the stakes
You raise the stakes by making each act cost more than the one before it, at a word position the reader can feel. A useful default for an 80,000-word novel in three acts is pressure level 3 near word 26,700, level 7 near word 53,300, and level 10 at word 80,000. Each step should name a concrete thing that is threatened and a concrete thing that disappears if the hero fails. Each step should also remove one way out, so the next danger cannot be solved the same way the last one was.
That is the whole method this page teaches. Name the loss. Place it on the calendar of the book. Take away an escape route. Repeat until the climax charges the full price.
The planner on this page does that arithmetic for you. It takes your target length, your genre, your core stake, and your act count, then returns a ladder with a stake level from 1 to 10 per act, a word position per rung, and a flag wherever two neighboring rungs sit too close together. Everything below explains how to read that ladder, how to write each rung so it lands, and what to change when the ladder reports a flat spot or a short book.
This tool organizes words you supply and applies fixed rules to them. It does not fetch market data, it does not contact any service, and it does not draft publishable scenes for you. It assembles your inputs into a structured plan with deterministic arithmetic, and the same inputs always produce the same ladder.
What stakes actually are on the page
Stakes are the named loss waiting behind the current struggle. Conflict is the locked door. Stakes are who sleeps outside if the door never opens. Readers often confuse the two, and drafts suffer when every scene has loud conflict but nothing clearly gets taken away.
A working stake has three parts. First, a person who wants something specific. Second, a threat that can take something specific from that person. Third, a consequence the story will enforce if the threat succeeds. When any of the three stays vague, tension leaks out even if the action looks busy.
Consider the difference between general dread and a priced loss. General dread says the hero might fail and things could get bad. A priced loss says the hero will lose the apartment on Friday, the sister will stop speaking to her, or the hearing will record a guilty finding that ends her license. The second version can be staged, counted down, and paid. The first version can only be described louder.
Stakes also need an owner. A threat to the world rarely moves a reader unless one household, one crew, or one partnership carries the bill. Large dangers work when they arrive through a small door: a missed payment, a broken promise, a revoked permission. Build the ladder from losses the hero personally absorbs, and let the larger danger sit behind those losses as context.
Time pressure belongs on the ladder too. A loss that can happen at any moment creates less urgency than a loss scheduled for a known point. That is why the planner pins every rung to a word position. The position tells you when the bill comes due, which tells you how many scenes you have to prepare the reader for payment.
How the tool math works
The planner uses four inputs and a small set of fixed rules. The inputs are total target words, genre, core stake, and act count. The outputs are a stake level per act, a word position per act, a threatened item and a lost item per act, a flat-rung flag wherever growth stalls, and a verdict about climax pressure.
Word positions come from even division. The tool divides total words by the act count and places each rung at the end of its span. An 80,000-word book with three acts produces rungs near word 26,667, word 53,333, and word 80,000. The page rounds those to readable numbers such as 26,700 and 53,300. A 90,000-word book with five acts produces rungs near word 18,000, word 36,000, word 54,000, word 72,000, and word 90,000. Every rung therefore owns an equal span of scenes, and you can see at a glance how much room each escalation gets.
Stake levels run from 3 at the first rung to 10 at the final rung. The formula spreads the climb evenly: start at 3, add the act index multiplied by 7 divided by one less than the act count, and round to the nearest whole number. With three acts the levels are 3, 7, and 10. With four acts the levels are 3, 5, 8, and 10. With five acts the levels are 3, 5, 7, 8, and 10. With two acts the levels are 3 and 10, which is a deliberate warning case discussed below.
Threat wording comes from a fixed table that pairs genre with the chosen core stake. The five core stakes are survival, love, identity, justice, and freedom. Each one defines what is threatened and what failure costs in plain terms: survival threatens bodily safety and shelter and costs life, health, or a safe place; love threatens the bonded relationship and costs trust, partnership, and a shared future; identity threatens the sense of self and costs belonging, role, and personal history; justice threatens a fair outcome and costs truth, restitution, and accountability; freedom threatens movement and choice and costs autonomy and the right to decide. The genre then supplies the surroundings: fantasy adds the realm and companions, romance adds the household under construction, thriller adds the operation and its cover, mystery adds the case and the reputation behind it, science fiction adds the ship and mission supplies, literary adds family ties and the livelihood beneath them. No external data enters this step. The wording is assembled from your two selections through that table, so identical selections always return identical threat lines.
Flat-rung detection compares neighbors. Any rung after the first that rises by 1 level or less gets flagged as flat. In the five-act ladder above, the step from 7 to 8 gains only 1, so that rung is flagged. The flag means the reader will likely experience two consecutive sections as equally dangerous, which reads as stalling. The fix is editorial, not numerical: rewrite one of the two rungs so it removes an escape route the other rung left open.
The verdict follows three checks in order. Books under 30,000 words with more than two acts fail, because each span is too short to stage a full escalation. Two-act structures warn, because a jump from 3 to 10 with no middle step forces the ending to feel sudden. Any flagged flat rung warns, because the climb pauses. Everything else passes, which means the ladder climbs cleanly from 3 to 10 with room per rung.
Worked numbers make this concrete. Take 80,000 words, fantasy, survival, three acts. Words per act equal 26,667. Levels are 3, 7, 10. Gains are 4 then 3, so no flats. Positions are about 26,700, 53,300, and 80,000. The verdict passes. Take 90,000 words, romance, love, five acts. Words per act equal 18,000. Levels are 3, 5, 7, 8, 10. Gains are 2, 2, 1, 2, so the fourth rung is flat. The verdict warns. Take 24,000 words, literary, identity, three acts. Words per act equal 8,000. The verdict fails even though the levels still read 3, 7, 10, because 8,000-word spans cannot comfortably hold setup, pressure, and payment three times over.
Reading your ladder result
The result opens with a verdict line that states the outcome in one sentence. A passing ladder says how many acts climb from 3 to the final level with no flat rungs. A warning names the flat count or the two-act jump. A failure names the shortfall in words per act. Read this line first, because it tells you whether to draft forward or repair first.
Four metrics follow. Book length restates your target with the act count beside it. Words per act shows the usable span per escalation, which is the number that decides whether a rung has room to breathe. Final pressure shows the closing level out of 10 with the opening level as context. Flat rungs counts the flagged steps, with a hint that points at rewriting rather than renumbering.
The ladder table carries the working detail. Each row gives the act number, the stake level, the approximate word position, the threatened item, and the loss if the hero fails. Act rows flagged with the word flat deserve attention before drafting. The threatened column answers what hangs in the balance during that span. The lost column answers what disappears at the rung if the effort collapses. When both columns name specific people, places, or permissions, the ladder is ready. When either column drifts into mood words, sharpen it before you write scenes from it.
The pressure curve underneath translates the same numbers into a visual span across the word count. Early segments render in a calm tone, middle segments step up through stronger tones, and the closing segment renders at full intensity. Use it to check pacing at a glance: a long calm opening followed by a sudden dark finale means the middle rungs need heavier consequences, while a curve that darkens early and stays dark suggests the opening gives away too much.
Recommendations close the result with two concrete next moves tuned to your verdict. Passing ladders get drafting guidance for the opening scene and the climax. Failing ladders get combining guidance for merging middle rungs. Two-act ladders get midpoint guidance for inserting a partial loss. Flat ladders get a named word position to repair plus the gain sequence to verify. Treat these as assignments, not suggestions, and complete them before expanding the outline.
Illustrative hypothetical example one: the mountain courier
This example is illustrative and hypothetical. It describes an invented story used only to show the ladder in action.
Imagine a fantasy courier named Tove who carries medicine between high villages. Her core stake is survival. The book targets 80,000 words in three acts. The planner returns levels 3, 7, and 10 at roughly word 26,700, word 53,300, and word 80,000.
In the first span, Tove loses a mule and half her stock to a river crossing. What is threatened is bodily safety and shelter tied to the realm and her companions; what failure would cost is not yet her life but her route income and her standing with the villages waiting on doses. The level 3 rung works because the loss is specific and countable: one animal, half the packs, three villages now short. An escape route remains, because she can borrow her cousin's mule and walk the low pass.
The midpoint rung must burn that escape route. Near word 53,300 the low pass closes with snow, the cousin recalls the borrowed mule for his own run, and Tove must cross the high col alone with a cracked boot and a fever starting. The threatened item is the same category but heavier: continued crossings now risk her health, not just her pay. Level 7 fits because the story removes backup, access, and easy retreat in one sequence rather than merely repeating the river accident louder.
The climax charges the full price. In the final 12,000 words the medicine she carries is the only course for her own village, and the storm forces a choice between sheltering with the packs or abandoning them to reach the hut line before dark. Failure now costs life, health, and any safe return, plus standing with allies who trusted her with the run. Level 10 fits because no rescue remains: the cousin is snowed in, the low pass is gone, and the hut line keeps limited bunks for carriers who arrive with packs intact.
Notice the pattern. Each rung threatens the same core stake but enlarges the bill and deletes one alternative. The river takes stock. The snow takes the alternate route. The storm takes shelter itself. A writer following this ladder knows exactly what each third of the book must accomplish, and the reader feels the climb because options visibly narrow.
Illustrative hypothetical example two: the shared bakery
This example is illustrative and hypothetical. It describes an invented story used only to show emotional stakes climbing without physical danger.
Imagine a romance about two partners, Priya and June, renovating a shared bakery while deciding whether to marry. The core stake is love. The book targets 90,000 words in five acts. The planner returns levels 3, 5, 7, 8, and 10 at roughly word 18,000, 36,000, 54,000, 72,000, and 90,000, and it flags the fourth rung as flat because the step from 7 to 8 gains only 1.
The first three rungs climb well. Act one threatens the bonded relationship through the household they are building: a missed loan payment puts the ovens at risk and exposes different money habits. Act two widens the cost: Priya accepts night shifts that solve cash flow but erase every shared evening, so trust frays in a countable way. Act three reaches the point of no return: June signs a supplier contract Priya opposed, and breaking it would cost the deposit that funds the wedding. Each rung removes an easy fix. Extra hours, then separate schedules, then a signed obligation.
The flagged fourth rung is where many drafts stall. A weak version repeats act three with another argument about the same contract at the same intensity. The reader senses déjà vu because nothing new is taken away. The repair is to delete a remaining escape route. Perhaps June's mother, who had offered to cover the deposit, withdraws the offer after witnessing a quarrel, or the landlord converts their month-to-month lease into a year term that locks them into the space together. Now level 8 differs from level 7: the couple cannot pause, borrow, or separate the business from the wedding. The threatened item is still the partnership, but the cost of failure now includes a workable daily life together, not just one event.
The finale then earns level 10. In the last 12,000 words the ovens fail inspection days before opening, and only one partner can attend the reinspection because the other must cover a shift that pays the repair. Whoever shows up decides alone whether to open late or open small. Failure costs trust, partnership, and the shared future they defined the bakery to mean. The ladder works here because every rung priced love in practical terms: payments, schedules, signatures, leases, inspections. None of it required a chase or a weapon.
What each level from 1 to 10 means in practice
Levels are shorthand for how much of the hero's life is engaged. Use the following descriptions to calibrate scenes so a level 7 chapter does not read like a level 4 chapter with louder adjectives.
Level 1 means background unease. Something could matter later. A bill arrives, a look lasts too long, a rule is mentioned. Nothing is threatened yet in a way the hero must answer this week.
Level 2 means a named inconvenience. The hero must spend time or money but keeps every option. A delivery is late. A form needs refiling. Write this briefly or skip it; few acts should rest here.
Level 3 is where ladders on this page begin. A real risk appears inside the hero's ordinary life. Income wobbles, a relationship strains, a permission is questioned. One concrete scene should show what could be lost, while backup plans still exist.
Level 4 means the inconvenience turns costly. The hero pays in a way that cannot be refunded: savings dip, a shift pattern breaks, a friend keeps distance. The escape routes still function but each now has a fee.
Level 5 means the cost spreads. A second area of life absorbs damage from the first. Work trouble reaches home, or case trouble reaches reputation. Scenes should show two fronts requiring attention at once.
Level 6 means a partial loss becomes permanent. Something the hero expected to recover is gone: a deposit, a reference, a season. The hero adjusts plans downward and stops talking about full recovery.
Level 7 means the point of no return. A signature, a statement, or a departure locks the hero into the remaining path. Retreat would now cost nearly as much as advance. This is the natural home of a midpoint reversal.
Level 8 means collapse threatens everything built so far. The business, the case, the voyage, or the household could end entirely. Allies recalculate whether to stay. The hero operates with no slack and borrowed resources.
Level 9 means the hero pays in advance. A sacrifice is made before knowing whether it works: savings emptied, a relationship paused, a cover blown deliberately. The story should stage this as a choice, not an accident.
Level 10 means the full price is due with no rescue available. What was threatened is taken unless the hero succeeds in the scene. Endings earn this level by deleting every alternative assembled earlier, then asking for one final effort at maximum cost.
Most failed escalations come from skipping the middle. A draft jumps from level 3 worry to level 10 catastrophe with no levels 5 through 8 in between, and the finale feels unearned. Another common failure parks at level 6 for half the book, repeating partial losses without locking anything. The planner guards against both by spacing levels and flagging repeats.
Building each rung so it holds weight
Every rung needs four elements on the page: a visible threat, a scheduled payment date, a narrowed option set, and a scene that proves the loss is real.
The visible threat is a person or force taking steps the reader can observe. A landlord files papers. A rival hires the witness first. A storm line moves down the valley. Avoid threats that exist only in the hero's thoughts for pages at a time. Internal worry can color a rung, but the rung itself should contain an action by someone or something other than the hero.
The payment date is the moment the loss becomes final. Rent court convenes Friday. The supplier deposits the check Monday. The pass closes with the next snow. Place that date inside the rung's word span and count down to it across two or three scenes. Deadlines convert abstract danger into calendar pressure, and calendars are easy for readers to track.
The narrowed option set is what separates neighboring rungs. Before writing a rung, list the hero's current outs: who could help, what could be sold, where the hero could retreat, which rule could bend. Then close exactly one door in a way that cannot be reopened. Burn the letter, spend the reserve, offend the helper, sign the clause. If a rung closes no door, it will likely read as flat even if the prose is vivid.
The proof scene shows a smaller version of the loss landing on someone nearby. A neighbor loses the same hearing. A fellow carrier turns back with frostbite. A friend's engagement ends over the same quarrel pattern. Proof scenes teach the reader the price without exposition, and they make the hero's continued effort look brave rather than oblivious.
Dialogue earns its place in this design when it changes options. A conversation that restates feelings without altering permissions, money, or access does not raise a rung. A conversation that withdraws an offer, sets a condition, or issues an ultimatum does. When revising, test every major talk by asking what the hero can no longer do afterward. If the answer is nothing, shorten the talk or attach a consequence to its end.
Edge cases and failure modes
Short books with many acts form the most frequent failure. A 24,000-word story divided into five acts gives each rung under 5,000 words, which is roughly two scenes. Two scenes cannot establish a threat, count down to a payment date, close a door, and prove the loss. The planner fails such combinations outright. The repair is to reduce the act count or lengthen the manuscript, not to write faster. One full escalation in 8,000 words outperforms three rushed ones.
Two-act structures form a special case. They can work for novellas, long stories, and books deliberately built as setup plus extended finale. The risk is a missing middle where the reader never sees levels 5 through 8. If you keep two acts, insert a clear midpoint reversal inside the first act: a partial payment around the halfway word mark that converts early worry into locked commitment. Without that beat, the jump from 3 to 10 snaps.
Five-act structures fail differently. They rarely lack room; they repeat. With five rungs the arithmetic forces at least one gain of 1, which the planner flags. Writers often try to fix this by inflating adjectives on the flagged rung. Stronger words cannot fix a repeated price. Only a removed alternative fixes it. Find the escape route both neighbors left open and close it in the flagged span.
Repeating the same loss type across rungs is another stall. Three consecutive scenes about money, or three about arguments, blur together even when amounts grow. Vary the currency of loss while keeping the core stake constant. A love ladder can charge money in act one, time in act two, pride in act three, housing in act four, and the relationship itself in act five. The stake stays love throughout, but each rung bills a different account.
Escalation without rest exhausts readers. A ladder that climbs every chapter without a breather turns tension into noise. Plan a short recovery scene after each payment: a meal, a walk, a joke with a friend, a competent small win. Recovery does not lower the rung; it lets the reader feel the new height before the next climb. Books that skip rest often get described as tiring rather than gripping, even when the plot logic is sound.
False escalation damages trust faster than slow escalation. If the story announces a terrible price and then discounts it through coincidence, the next announced price will not be believed. Avoid rescues that arrive without setup or cost. Every survival should consume something: supplies, goodwill, health, time. The planner assumes enforced consequences, so honor that assumption in the draft.
Protagonist passivity breaks ladders from the inside. When events escalate around a hero who takes no new action, the levels describe weather rather than story. Each rung should contain one decision the hero makes that contributes to the next danger: signing, hiding, pursuing, refusing, confessing. The decision can be sympathetic and still costly. Readers accept rising pain when the hero's own commitments help cause it.
Climax crowding is the late-book version of the same error. Writers save three unpaid bills for the final chapter and detonate them together. The result reads as coincidence rather than culmination. Instead, pay most bills on schedule across the middle rungs and reserve the finale for the single largest one. A clean level 10 built on four paid smaller losses overwhelms a finale that introduces fresh debts.
Genre-specific guidance
Fantasy escalates through obligations and terrain. Early rungs threaten pay, post, or apprenticeship. Middle rungs threaten oaths, strongholds, and companions. Late rungs threaten the realm through the hero's household or crew rather than through speeches about kingdoms. Remove magical or material outs one at a time: the charm cracks, the ally is posted elsewhere, the pass closes. Because fantasy readers accept high final prices, the greater risk is starting too high. Keep act one near level 3 with a countable village-scale loss, and the level 10 finale will feel vast by contrast.
Romance escalates through entanglement. Early rungs threaten convenience and routine. Middle rungs threaten shared plans that cost money to exit. Late rungs threaten daily life itself. Useful doors to close include separate leases, backup childcare, flexible schedules, and family subsidies. The finale should force a decision that cannot be split: one home, one calendar, one public commitment. Keep physical safety out unless the premise promises it; importing danger late to inflate a love climax usually breaks the contract with the reader.
Thriller escalates through exposure and time. Early rungs threaten the cover: a camera angle, a curious clerk, a phone record. Middle rungs threaten the operation: safe houses burned, contacts doubled, funds frozen. Late rungs threaten the person behind the role. Each rung should shorten the timeline while removing a resource. If the opening already burns the cover, later rungs have nowhere to go, so stage exposure gradually and save identity itself for the finale.
Mystery escalates through access and credibility. Early rungs threaten the case: a witness cools, a record seals, a client wavers. Middle rungs threaten the investigator's standing: credentials questioned, allies warned off, a complaint filed. Late rungs threaten the truth itself going unrecorded, plus the investigator's livelihood. Avoid raising bodies when the ladder calls for raising costs; a second crime does not automatically raise stakes unless it removes an option the first crime left open. Closing doors matters more than counting victims.
Science fiction escalates through systems and supplies. Early rungs threaten comfort and schedule: rations trimmed, a burn window missed, a module sealed. Middle rungs threaten redundancy: backup oxygen, spare parts, alternate landing sites. Late rungs threaten the mission and the crew together. Track consumables on paper so each rung spends a named reserve. Readers of the genre notice when scarcity is announced but never audited, and the ladder fails if declared shortages never constrain a decision.
Literary fiction escalates through irreversibility. Early rungs threaten routine and face. Middle rungs threaten employment, housing, and family patience. Late rungs threaten the terms on which people will continue knowing one another. Doors here are social: an apology refused, a confidence repeated, a guardianship transferred, a house sold. Because the prose often carries nuance well, the risk is vagueness about price. Name what ends, on which date, and who will not return, and quiet stakes will hold as firmly as spectacular ones.
Horror and suspense, for writers blending them with the listed genres, escalate through proximity and knowledge. The threat moves closer while the hero understands more about what contact costs. Close distance first, then light, then witnesses, then exits. The finale should combine full understanding with zero distance, so the hero faces the thing clearly with nowhere to stand apart from it.
Frequently encountered questions in the workshop
Writers often ask whether every subplot needs its own ladder. It does not. One main ladder carries the book. Subplots should attach their payments to the same rungs so the whole story tightens together. A subplot that peaks between main rungs can work as a breather, but a subplot that demands its own level 10 in the final pages competes with the ending. Schedule subplots to resolve one rung before the finale, and let their resolution remove or grant one resource the climax needs.
Another common question concerns likable heroes and harsh prices. Readers do not require the hero to suffer constantly; they require the suffering to mean something. A well-built ladder alternates payment with competence. Let the hero solve real problems inside each rung while the rung still charges its fee. Competence plus cost produces admiration. Cost without competence produces pity, and pity rarely sustains a novel.
Many writers ask how stakes differ for series books. The main ladder of each volume must still close at level 10 within that volume. A volume that ends at level 7 because the true price waits two books ahead will feel unfinished. Series escalation sits above the per-book ladder: each installment can threaten a larger circle — household, then town, then realm — while every installment still pays its own bill. Promise the larger circle early, but charge the current circle now.
A related question asks about stakes in quiet books where nobody faces ruin. The ladder still applies because ruin is relative. Losing a garden plot, a church role, or a weekly dinner can carry level 10 weight inside a life organized around those things. The requirement is proportion, not spectacle. Show the small thing holding up everything else, remove the props beneath it rung by rung, and the final loss will register fully.
Writers also ask when to reveal the full price. Reveal the shape early and the exact amount late. The opening should teach what kind of loss this book charges: money, belonging, freedom, love, or life. The middle should show the price rising through partial payments. The finale should state the exact bill and enforce it. Hiding the currency until the end confuses preparation, while announcing the exact finale price in chapter one removes dread. Shape early, amount late.
A final workshop question concerns revision order. Fix the ladder before fixing sentences. When a draft feels flat, writers often polish prose on rungs that charge the wrong price at the wrong position. No sentence-level work rescues a mispriced ladder. Reassign each major loss to its correct rung, confirm every rung closes a door, verify gains of at least 2 outside the flagged spans, and only then revise language. Structure first, music second.
Turning the ladder into chapters
Translation from ladder to outline is mechanical. Divide the word count by the act count to get each span. Inside each span, assign three to five chapters depending on your chapter length. Give the first chapter of the span the threat's entrance, the middle chapters the countdown and partial payments, and the last chapter the door closing. The next span then opens with the hero living inside the smaller option set.
A workable chapter target helps. At 2,000 words per chapter, an 80,000-word book holds 40 chapters. Three acts give roughly 13 chapters per act. That allows four chapters to introduce the rung's threat, six to spend down options, and three to enforce the payment and rest briefly. At 3,000 words per chapter the same book holds about 27 chapters, or nine per act: three to enter, four to tighten, two to pay. Either rhythm works. What fails is assigning one chapter per rung, which compresses setup, countdown, and payment into a summary.
Track options on a single page per act. List helpers, funds, permissions, refuges, and tools at the span's start. Cross one out at the span's end with the scene number noted. When every act crosses out at least one line, the ladder is enforced regardless of how elegant individual chapters feel. When an act crosses out nothing, flag it the way the planner flags a flat rung and revise before moving on.
Place the climax by subtraction. Reserve the final 10 to 15 percent of total words for the full-price sequence and its immediate aftermath. For 80,000 words that means the last 8,000 to 12,000 words. Everything before that zone builds and pays smaller bills. Writers who start the finale at 60 percent of the book usually discover they have no larger bill left for the actual ending, so they invent one. Plan the zone first and the middle will size itself correctly.
Doing the drafting work in Co-Writer
Once the ladder is set, the remaining job is writing chapters that enforce it scene by scene. That drafting work belongs in Co-Writer at the writing page at /write, where chapters can be built from the outline you just priced. Bring each rung's threatened item, lost item, and word position into the drafting session, and generate chapter by chapter so every span closes its assigned door. When a flagged rung needs repair, revise its chapters there until the escape route is gone and the next level rises by at least two.
Revision checklist before calling the ladder done
Confirm each rung names a person who pays, a thing taken, and a date it is taken. Confirm each rung after the first removes an alternative the previous rung left available. Confirm gains of at least 2 on every step except spans you deliberately designed as breathers. Confirm the finale sits inside the last 15 percent of the word count with no rescue unspent. Confirm the opening rung reads near level 3 rather than level 8, so the book has somewhere to climb. When all five checks pass, the escalation is ready and remaining revision can focus on voice, detail, and rhythm.