How to Tell Stories in a Speech

The advice is always the same: tell stories in a speech and people will remember what you said. What nobody explains is what makes something a story rather than a list of things that happened.

That gap matters, because a speaker told to add a story usually produces events in the order they occurred. Audiences hear the difference immediately, even when they could not name it. Research can name it, and can also say how much a story is really worth — less than the internet claims, and still worth the work.

OneHow much a story is actually worth

Raymond Mar and his colleagues collected every study they could find that tested the same kind of information twice: once written as a narrative, once as an explanation. That came to thirty-seven articles, seventy-eight samples and just over thirty-three thousand people.

Narrative won on both measures. For memory the difference was a Hedges’ g of 0.72; for comprehension, 0.48. Those are moderate to large differences — the size you would notice in a room, not the size that only shows up in a spreadsheet.

Now the finding that works against all of this, and the same team went looking for it. They tested their own collection of studies for publication bias and found it: the less precise a study was, the bigger the effect it reported. That is the pattern you get when results showing no difference quietly never reach print. The real advantage is probably smaller than 0.72.

They add a caution most storytelling advice ignores: their result should not be read as a reason to force all information into narrative form. So the honest claim is the modest one. A story is remembered better than an explanation of the same material — which is not a reason to make your whole talk a story. A talk made only of stories has no point left in it.

Two“And then” is not a story

Tom Trabasso and Paul van den Broek went back over story-recall data from earlier experiments and mapped each story as a network: which event caused which other event. Then they checked what predicted the events people remembered, and the events people rated as important.

The answer was causal connection. Events linked by cause to more of the other events were recalled better and judged more important. Events that merely happened next were neither.

This was a reanalysis of other researchers’ data rather than a new experiment, so treat it as a strong pattern rather than a controlled test. It still gives you the most useful editing rule here. Between any two sentences of your story, try to put the word “so” or “because”. If only “and then” fits, that sentence is doing no work.

• Weak: I got to the office. And then I set up the room. And then people came in. And then I started.
• Strong: I got to the office late, so I never tested the microphone — which is why the first two minutes were silent for everyone at the back.

Same events, half the words, and only the second one is still in anybody’s head on Friday.

ThreeEvery story needs a “so what”

In the 1960s, William Labov and Joshua Waletzky recorded ordinary people telling stories about things that had happened to them, and found the same shape under almost all of them: a one-line abstract, an orientation, the complicating action, an evaluation, a result, and a coda that brings the listener back to the present. Their framework describes what fluent tellers do; it does not test whether copying them works, so treat it as a map rather than a result.

The part speakers leave out is the evaluation — the section where the teller says why this is worth hearing. It answers the question a listener never asks out loud but always has: so what?

The fix takes one sentence. When your story ends, say your point out loud. Not implied, not left for them to draw. Said.

FourWhat absorption does, and why to be careful with it

Melanie Green and Timothy Brock gave a name to the state a good story puts people in: transportation. Being absorbed in a story, with your imagination, your feelings and your attention all pointed the same way.

Across four experiments, the more transported readers were, the more they held beliefs matching the story and the more warmly they judged its main character. When the researchers reduced transportation by changing how people were told to read, those beliefs weakened too.

Two smaller findings are more useful to you than that headline. In one experiment, with sixty-nine readers, the most transported people noticed fewer false notes in the story — fewer details that were wrong or did not fit. And across the studies, telling people the story was fact or fiction generally made no difference to how transported they were, or to what they came to believe.

That is a reason to handle this carefully, not a reason to be pleased with it. Tell stories that happened, and say when you are simplifying. And if a storytelling claim sounds too neat to be true, note that the most famous one failed a careful replication.

FiveFive moves, and the English to build them

Everything above turns into a shape you can build in about ninety seconds of speaking:

• Orient in one sentence. One time, one place, one person.
• Introduce the problem. Something has to go wrong or there is no story.
• Say what you did, using “so”. This is where the chain of causes gets built.
• Mark the turn. The moment it worked, or the moment you knew it would not.
• State the point. The evaluation. One sentence, said out loud.

The English you need is smaller than you expect. Past simple carries the events, past continuous carries whatever was already going on around them: I was waiting for the file when the client walked in. That pair will carry almost any story you want to tell, which is also why a story is easier to hold in your head than a list of points — a subject with its own article.

Keep it near ninety seconds. A story that runs past its point loses the attention it earned, and how much time the rest of the talk can spare is a separate question.

The short version

Narrative beats explanation for memory and for understanding by a solid margin — though the published literature leans toward that result, and the researchers who measured it warn against forcing everything into story form. What turns events into a story is cause: the parts joined by “so” and “because” are the parts people keep. What makes a story worth telling is the evaluation, the sentence where you say why you told it. And an audience inside your story is not checking whether it is true, which is a responsibility rather than a technique.

One true thing that happened to you, ninety seconds, joined with “so”, ending with your point. That is the whole method.