There's a tiny word that does more work in a story than any clever phrase available. It's three letters long. It appears twenty times in everyday speech without notice. The word is "but."
Listen to the difference. "A person went to the store. They bought milk. They came home." Nothing. Now: "A person went to the store for milk — but the lights were off, the door was chained, and someone was standing in the parking lot waving at them." Suddenly the listener wants to know what happened. Same trip. The only thing that changed is that something got in the way. That little obstacle is the whole engine. Without it, it's just narration of errands.
That's the move this section is built around. A story holds a listener only as long as there's a question hanging in the air that needs answering — and the word "but" is the simplest way on earth to open that question. The shape from the last section gave a story a beginning, a middle, and an end. Tension is what makes anyone care about getting from one to the next.
Here's the thing about attention that nobody likes to admit. A listener is not obligated to stay. Every second they're listening, some part of their brain is running a quiet cost-benefit calculation — is this still worth it, or should I check my phone, scan the room, plan dinner? They stay for exactly one reason: there's something they don't know yet and want to. That wanting is tension. The moment it's gone, so are they.
So think of tension less like drama and more like a debt opened at the start of a story and paid off at the end. A promise is made — a question, a problem, a "wait, how did that turn out?" — and the listener extends credit. They'll keep listening as long as they believe the payoff is coming. Welch on that debt, take too long, or let them figure out the ending early, and the credit dries up. They tune out, politely, while still nodding near the snack table.
This is exactly why the rambling weekend story fails. There's no debt. The friend describing the broken elevator never opened a question that needed closing, so there was nothing pulling the listener forward. Contrast that with a story that starts, "So I almost got arrested last Tuesday, and it was completely my own fault." The listener isn't going anywhere. A debt has been opened, and now it needs settling.
Which brings us to the cleanest checklist available on this topic. Chris Anderson, the head curator of TED — the person who's watched more people succeed and fail at telling a story on stage than almost anyone alive — boils a good story down to four things, and they map directly onto tension. Writing in a piece adapted from his book on public speaking, Anderson identifies four elements: a character the audience can empathize with, tension built through curiosity or intrigue or actual danger, the right level of detail, and a satisfying resolution.
Sit with that middle pair for a second, because they're where most stories live or die. Build tension. Offer the right level of detail. Those two are in constant negotiation with each other, and getting the balance wrong is the most common way a promising story collapses.
Start with the character, though, because tension needs somewhere to live. Anderson's first element is a relatable character — someone the listener can root for or worry about. This matters more than it sounds. Tension can't be felt about a stranger there's no stake in. If told "a man was running late," most listeners shrug. If told "my sister, who has never been late to anything in her life, was about to miss the one flight she'd been planning for two years," now there's a person to care about, and the lateness suddenly has weight. The tension isn't in the event. It's in the gap between what this particular person wants and what's standing in their way.
So here's the most practical move in this whole section, and it's almost embarrassingly simple. Plant the question early, then make the listener wait for the answer. The Heath brothers, Chip and Dan, in their book "Made to Stick," call this a curiosity gap — the space between what someone knows and what they want to know. Open that gap up front, and the listener's own brain does the work of keeping them engaged. They're not staying because the storyteller is charming. They're staying because a question was left open and the human brain hates an open loop.
Think about how a good murder mystery works. A body appears on page one. The reader doesn't find out who did it for three hundred pages. Everything in between is just the writer keeping that gap open, dangling it, refusing to close it until the last possible moment. The exact same technique works when telling someone about a job interview. Don't open with "I got the job." Open with "I was eleven minutes late to the most important interview of my life, and the elevator was broken." Now the listener has to know how it ended. The body is on page one.
This is the part where most people get it backwards, so stay with this for one step. The instinct, when nervous or excited, is to lead with the resolution because it's the good part. "I got the job! So anyway, here's what happened." And the second that's said, the gap closes. The debt is paid off before it opens. Everything after that is the listener waiting for the speaker to finish so they can react politely. The resolution is the thing they're buying — never give it away in the first sentence.
Now, releasing tension too early is one failure. The opposite failure is just as deadly, and it's the one Anderson flags as the third element: detail. Offer too little and the story isn't vivid. Offer too much and it gets bogged down. This is the trap, and it's a sneaky one, because the details that feel important to include are almost never the ones the listener needs.
Anderson is blunt about why. He says many storytellers overstuff their narratives with details that matter to them, but a wider audience just doesn't need to know. That's the heart of it. A person remembers the weekend in full resolution — every floor of the parking garage, the name of the elevator company, what they had for lunch. All of it feels load-bearing because it was lived. To the listener, every detail that doesn't feed the central question is just noise on the line, drowning out the signal.
Here's the test. Every detail in a story is either tightening the tension or releasing it. There's no neutral. The broken elevator tightens it — it's the obstacle between the sister and her flight. The color of the elevator doors releases it, because while doors are being described, the listener's brain quietly notices that nothing is at stake in a door, and the pressure leaks out. A story bleeds tension one irrelevant detail at a time, and usually it can't be felt happening because each individual detail seems harmless.
So if trimming a story and unsure whether a detail earns its place, ask one question — does this make the listener more anxious to know what happens, or less? If it's less, cut it, no matter how vivid it is. The discipline isn't adding good detail. It's deleting the detail that isn't pulling the listener forward.
Now, there's a real argument about how far to stretch the wait, and serious storytellers don't fully agree on it. The screenwriting tradition pushes toward delay, delay, delay — maximum suspense, hold the answer until the last possible breath. Anderson's framing leans the other way. His whole list ends on a satisfying resolution, and he warns that a story which intrigues but never delivers something the listener can walk away with is, in his word, heartbreaking — a fascinating person who built tension and then paid off nothing. The lean here is toward Anderson. In a casual conversation, milking suspense the way a thriller does just reads as withholding, and people resent being toyed with over a story about a commute. Stretch the tension as far as the stakes will honestly carry, and not one beat further. A two-minute story can hold a question for ninety seconds. It cannot hold one for nine.
There's a corollary the TED material is firm on, and it's worth saying plainly. Whatever tension gets built has to be true. People sometimes exaggerate or fib precisely because tension lands — they sense the elevator wasn't actually broken, so they break it for the story. Anderson calls this a temptation that has to be resisted. A manufactured obstacle is the fastest way to lose someone, because the listener can feel the seam where reality stops and the embellishment starts, and the moment they feel it, every other word becomes suspect.
So strip this down to what's doing the real work. Tension is a debt — it opens with a question and requires a satisfying answer. It opens early, with a "but," with an obstacle, with a body on page one, and it's protected by cutting every detail that doesn't make the listener more desperate to know what happens. The character gives the tension somewhere to live; the curiosity gap is what keeps them in their seat; the resolution is the only thing they were ever really waiting for, so it never, ever gets given away first.
Here's the line worth carrying out of this one: a story isn't held together by what gets told — it's held together by what the listener is made to wait to find out.
And that wait only works if the listener actually has a stake in the answer. Which raises the question the next section takes head-on — because the quickest way to make someone need to know how a story ends is to quietly make the story about them.