A setup without a payoff is a promise the book makes and never keeps: the locked drawer in Book 1 nobody ever opens, the prophecy that never comes due, the gun on the mantel that never fires. In a standalone, a reader will forgive one dangling thread. Across a series they keep a running list, and they expect every line you underlined to come back. Tracking foreshadowing means recording each setup, marking where it pays off, and finding the ones that never do.
Chekhov's rule is usually quoted at the scale of a single play: if a rifle is hanging on the wall in the first act, it has to go off by the third. The trouble with a series is that the first act and the third act can be a thousand pages and three years of writing apart. The promise you plant in Book 1 has to fire in Book 4, and by the time you get to Book 4 you may not remember you ever loaded it. Setups dangle in four predictable ways.
The deferred payoff you never circle back to. You plant a detail early on purpose, the strange birthmark, the sealed letter, the debt the hero swears to repay, intending to cash it in "later in the series." Later arrives, the plot has its own momentum, and the planted detail is still sitting there, loaded and untouched, three books on.
The mystery the cast stops caring about. Book 1 makes a question feel urgent: who killed the steward, what is behind the locked door, why does the old map have a seventh kingdom. The characters obsess over it for a hundred pages, then the story moves on to a new crisis and the original question is quietly abandoned. The reader, who was promised an answer, has not forgotten it.
The prophecy or oath that outlives its arc. Prophecies are setups with a deadline. "The last of the line will fall before the second frost." If the series ends and the line never falls, or falls for unrelated reasons that have nothing to do with the second frost, the prophecy was a load-bearing promise that held up nothing. Oaths and curses work the same way: they raise a flag the reader watches for.
The Braindump note that never made it onto the page. Series authors plant setups in two places, the manuscript and their own notes. You jot "remember the locked drawer pays off in Book 3" in your planning doc, fully intend to write it, and then the scene that opens the drawer never gets drafted. The intent exists; the payoff does not. A reader only ever sees the page.
A setup is a contract the book signs in Chapter 2. The reader holds you to it long after you have forgotten you signed.
Foreshadowing is not one list of "cool details." It is a set of paired entries, each a promise and its eventual receipt, with a span of pages or books between them. This pairing is what StoryHelm's Thread Coverage dimension measures: which narrative threads open, which close, and which are left hanging. Six things are worth recording explicitly.
The discipline is not recording each setup once. It is pairing it with its payoff and re-checking, after you have written four books, that nothing was left open by accident. A setup you plant deliberately and never resolve is a loose thread; a setup you forgot you planted is the same loose thread, just one you cannot defend.
You can track foreshadowing by hand, and disciplined authors do, in a "promises and payoffs" spreadsheet or a planning doc full of "DON'T FORGET: the drawer." It works on day one. The trouble is that the list records what you meant to write, while the manuscript records what you actually wrote, and those two drift apart the moment a revision cuts the payoff scene or a sequel adds a setup you never logged.
An automated read does not decide for you whether a thread should pay off, or how. Some setups are meant to stay ambiguous, and a deliberate unanswered question can be the point. The read removes the bookkeeping: confirming that the threads you intended to close actually close somewhere in the prose, and surfacing the ones that quietly didn't.
An orphaned setup is rarely a mistake of craft. The setup itself is usually the best writing in the book, vivid, specific, deliberate, which is exactly why the reader files it away to be paid back. No single chapter is broken. The flaw only appears when you read Book 1's promise against the four books that fail to answer it, and that is the one view a chapter-by-chapter pass never takes.
Here is the kind of dangling thread that survives every chapter-level read, because the setup reads beautifully in isolation. Book 1 plants the drawer; nothing in the four books that follow ever opens it:
Read on its own, the Book 1 passage is excellent: "waiting" is doing real work, telling the reader to remember this. The craft is precisely what creates the debt. The prose has signed a contract, and a reader who carried that locked drawer through five books will reach the end feeling shorted, even if they cannot point to the page where the story let them down. The series opened a thread and never closed it.
StoryHelm reads every book in your series together, whether you wrote them in its Scene Editor or imported finished files. Its Foreshadowing analysis pulls setups straight out of the prose: the locked drawer, the half-spoken prophecy, the planted object. The Thread Coverage dimension of your Plot DNA then pairs each setup with its payoff and flags the ones that pay off nowhere, pointing you to the exact book and chapter where the promise was made. It surfaces the dangling thread; the fix is yours to write, because StoryHelm reads and analyzes your prose and never drafts a line of it. For the wider picture of what this kind of read does, see what an AI continuity check actually does.
That is a legitimate choice, and ambiguity can be the point. StoryHelm flags unpaid setups so you can decide; it does not insist every thread close. The value is that the choice is now deliberate. You are looking at the list of open threads and signing off on each one, rather than discovering an accidental dangler in a one-star review.
Yes, and the wide span is the whole point. A setup planted in Book 1 and resolved in Book 4 is matched as a closed thread no matter how many chapters sit between them, which is the match a memory or a manual spreadsheet is most likely to lose. The thread that stays open in every book is the one that gets flagged.
A continuity check (the Coherence Guardian) catches contradictions: a name that changes, a timeline that doesn't add up, an eye color that drifts. Thread coverage catches the absence of a payoff, a promise the prose made and never kept. One finds the line that conflicts; the other finds the answer that never arrives.
StoryHelm reads what is on the page. If you logged "the drawer pays off in Book 3" in a Braindump or planning note but never drafted the scene, the manuscript has a setup and no payoff, and that is exactly what gets flagged. A reader only ever sees the prose, so that is what the analysis reads.
StoryHelm reads your whole series at once, pairs every setup with its payoff, and hands you the list of threads that pay off nowhere, pointed to the exact book and chapter where the promise was made. Catch the dangler now, not in the one-star review from the reader who carried it through five books waiting for the answer.
Request a founding seat