The sample chapter ends mid-sentence—or worse, mid-gesture, a character's hand on a door handle that will never turn. You flip the page and find the copyright notice. What you feel in that moment isn't just frustration; it's a small, involuntary experiment in how the mind handles an unfinished loop. That unfinished loop is one of the most reliable tools in a bookseller's arsenal, and understanding why it works tells you a lot about how readers actually decide what to buy.
The Zeigarnik Effect and the Cost of a Closed Door
Nearly a century ago, psychologist Bluma Zeigarnik noticed that waiters remembered unpaid orders better than settled ones. The finding became one of the most replicated ideas in cognitive psychology: interrupted tasks occupy working memory more persistently than completed ones. An unfinished scene is an open task. The mind keeps rehearsing it—who is at the door, what they're about to say—because closure hasn't been delivered.
This is why a sample chapter that ends at a natural pause (a scene resolved, a chapter break) produces a polite shrug, while one that ends mid-scene produces a physical itch. The reader isn't being manipulated so much as handed an unresolved cognitive debt. The only way to pay it off is to keep reading, which usually means buying the book.
Why Mid-Scene Endings Feel Like a Fair Trade
There's a difference between a cliffhanger and a cheat. A cheat withholds information the reader has earned; a fair mid-scene cut delivers enough momentum that the interruption feels like good editing rather than a marketing trick. Daniel Kahneman's work on loss aversion is relevant here: readers weigh the prospect of losing a story they've already invested in more heavily than the prospect of gaining a new one. A sample chapter that ends mid-scene creates a small, specific loss—the next line—and losses loom larger than gains of equal size.
Booksellers see this play out at the register. A customer who reads a sample ending at a chapter break often sets the book down and wanders. A customer who reads a sample ending mid-scene tends to ask a different question: "Do you have another copy of this one?" The decision has already been made; they're just completing the transaction.
The Publisher's Variable-Ratio Problem
Publishers have experimented with sample structures for decades, and the results map onto a well-known behavioral pattern: variable-ratio reinforcement, the schedule in which rewards arrive unpredictably. A sample that ends at a predictable point (always the end of chapter one) trains readers to expect a clean stopping place—and clean stopping places are where readers stop. A sample that ends unpredictably, sometimes mid-scene, sometimes at a chapter break, keeps the reader uncertain about when the next satisfying beat will arrive. That uncertainty is mildly aversive, and readers resolve aversive uncertainty by continuing.
The catch is dosage. End every sample mid-scene and readers learn the trick; the itch becomes irritation. The most effective samples I've seen vary the cut point across a catalog, so no single reader can predict where the excerpt will stop.
What This Means for the Shop Floor
If you run a bookstore or a reading series, the practical takeaway is to treat sample endings as a design decision, not an afterthought. When you write shelf talkers or staff picks, end your description mid-thought—give the setup, name the tension, and stop before the resolution. When you host a reading, read the passage that ends on the unresolved gesture rather than the one that ties off neatly. And when a customer asks whether a book "gets good," hand them the sample that ends mid-scene and let them find out for themselves. The unfinished loop does the selling; your job is simply to place it in the right hands.