Storytelling statistics: what the research actually proves — and the 22x myth

The effect of storytelling on belief and memory is well supported by peer-reviewed research. The specific statistics circulating in marketing content mostly are not. The most-quoted figure — that stories are "22 times more memorable than facts" — is attributed to Stanford professor Jennifer Aaker and repeated by Stanford's own channels, but we could not locate a published study, sample size or method behind it anywhere. Four findings do hold up, and they are more useful than the myth: narrative transportation shifts beliefs (Green & Brock, 2000), character-driven narrative triggers oxytocin release that predicts helping behaviour (Zak, 2015), story-structured information survives delayed recall where rote study collapses (Bower & Clark, 1969), and thought-leadership content measurably influences B2B buying groups (Edelman–LinkedIn, 2025).

The four findings that hold up

Each of these is a real study you can look up, and each licenses a narrower claim than the internet makes on its behalf. The narrow version is the useful one, because it survives being questioned.

1. Absorption in a story shifts what people believe

Green and Brock (2000), Journal of Personality and Social Psychology 79(5), 701–721, ran four experiments with samples of 97, 69, 274 and 258 participants. Readers who reported being more transported into a narrative showed more story-consistent beliefs afterwards and evaluated the protagonist more favourably. When the researchers deliberately reduced transportation, the belief shift shrank with it — which is what makes this a mechanism rather than a correlation.

What it licenses you to say: that absorption in a narrative measurably changes belief. What it does not: anything about revenue, conversion, or how much better a story performs than a bullet list in an inbox. And one finding deserves to be quoted more often than it is — the effects held whether the story was labelled fact or fiction. That is not permission to invent. It is the reason your stories have to be attributable, because persuasion is clearly not doing any truth-checking on your behalf.

2. Character-driven narrative changes willingness to act

Zak (2015), Cerebrum 2015:2, summarises work showing that character-driven narrative causes oxytocin release, and that the amount released predicted participants' willingness to help people connected to the story — including donating money.

What it licenses: "character-driven narrative increases willingness to help." What it does not: "storytelling increases sales by 30%." The measured behaviour was helping, in a lab, in the minutes after exposure. That is a real and interesting effect. It is not a conversion rate.

3. Story structure makes information last

Bower and Clark (1969), Psychonomic Science 14(4), 181–182, gave students twelve serial lists of ten nouns each. One group chained each list into a narrative; the yoked control group studied the same words for the same amount of time. Immediate recall was effectively identical — 99.9% against 99.1%. On the delayed probe, the narrative group recalled 93% and the control group 13%, six to seven times more, with no overlap between the groups.

What it licenses: that story structure dramatically improves durable recall. What it does not: that stories are easier to understand, or that this transfers cleanly to business communication — this was serial word-list learning with students, and extending it to a sales meeting is an inference, not a finding. Almost nobody who quotes the 93% mentions that immediate recall was a tie. That detail is the whole insight: story does not help people take it in, it helps them still have it next week.

4. Thought-leadership content moves B2B buying groups

The 2025 Edelman–LinkedIn B2B Thought Leadership Impact Report surveyed 1,934 management-level professionals across seven markets, fielded 17 March to 3 April 2025. Two findings matter for anyone building a story to be forwarded internally: more than 40% of B2B deals stall on internal misalignment, and "hidden buyers" in finance, legal and procurement influence outcomes while having no relationship with the vendor at all.

What it licenses: that the audience for your story includes people you will never meet, and that internal disagreement — not competitor preference — is the most common place deals die. What it does not: any claim that narrative specifically outperforms other formats on revenue.

The 22x claim, traced

The claim is that stories are 22 times more memorable than facts alone, usually attributed to Jennifer Aaker, General Atlantic Professor of Marketing at Stanford Graduate School of Business. Here is what we found when we went looking for the study.

1. Where it appears. Stanford GSB's own channels, Stanford's Women's Leadership Lab material, a Lean In education module, MIT's communication lab pages, Forbes, and — conservatively — thousands of agency and vendor blog posts. It is genuinely everywhere, including in places that ordinarily check things.
2. What is consistently missing. No paper. No journal. No year. No sample size. No method, no instrument, no measure of what "memorable" meant. Sources cite each other, or cite Professor Aaker directly, with nothing underneath. We could not find a publication in which the figure is derived.
3. The companion figure has the same problem. The pairing often quoted alongside it — that in a pitch exercise 63% of the audience recalled the story while 5% recalled the statistic — is described in the same secondary sources and, as far as we can establish, was never published either.
4. What we are and are not saying. We are not saying the claim is false, and we are not accusing anyone of fabricating anything. Professor Aaker is a serious scholar, the intuition behind the number is supported by other work, and figures do sometimes originate in classroom exercises before anyone thinks to publish them. What we are saying is narrower and, we think, undeniable: after a genuine search, the number has no citable methodology. A figure you cannot cite is a figure you should not put in a client pitch.
5. What to say instead. Use Bower and Clark, with its limits stated: story-structured information was recalled at 93% against 13% for equal-time rote study on a delayed test, in a 1969 word-list experiment. It is less dramatic than 22x, it is roughly the same order of magnitude, and it will survive someone asking you for the paper.

The numbers we could not verify

Every claim below appears constantly in storytelling content, including in decks sold to businesses. None of them survived a search for a primary source.

A related correction worth making, because it circulates in the same rooms: the widely repeated claim that Gartner forecast a 25% drop in website traffic by 2026 is a misstatement. The February 2024 press release said traditional search engine volume would fall 25% — not traffic to commercial sites. Those two are conflated constantly, that horizon has now passed with an ambiguous outcome, and the durable part of the advice was never the number anyway.

Why unverifiable claims never get caught

This is the part that connects a citation problem to our own working life, and it is uncomfortable to publish.

We film customer testimonials for a living, which means we spend our weeks in a room with someone who is describing something that happened to them. The most consistent thing we have learned is that subjects are overwhelmingly willing to be directed. Tell a person their answer sounded good and they will offer a stronger, tidier, more quotable version of it — and thank you for the guidance. And almost nobody ever comes back afterwards to say that isn't quite what I meant. We can put a number on that one, because we counted.

Now set that beside Green and Brock's finding that transportation moves belief regardless of whether a story is labelled fact or fiction. Neither end of the chain is a check. The audience will not catch a story that has been shaped past the truth, and the subject will not either — they will approve it warmly. That is exactly how an unsourced statistic behaves in the wild. It is plausible, it is useful, it flatters everyone who repeats it, and no one in the chain has any incentive or any mechanism to stop and ask for the paper. Years of repetition without a citation is not a conspiracy. It is what happens when a claim is agreeable and nobody is checking.

Which puts the responsibility in an awkward place: on the person telling the story. In our work that means the producer's restraint is the only real safeguard on accuracy. In yours it means the discipline of checking your own numbers, especially the ones that help you.

Why a bad statistic costs you the room

There is a practical argument here as well as an ethical one. Senior audiences check things, and they check the thing that sounds most impressive first. When one number in a deck gets challenged and cannot be defended, the cost is not that one number — it is that every other claim you made now carries an asterisk, including the true ones.

It gets worse in a buying group. Edelman and LinkedIn found that more than 40% of B2B deals stall on internal misalignment, and that hidden buyers in finance, legal and procurement shape outcomes without ever speaking to a vendor. Your champion is the person who forwards your material into that room, without you, to be read by people whose job is to find the weak seam. A statistic your champion cannot defend does not just fail to persuade. It becomes the reason the deal goes quiet.

How to cite storytelling evidence honestly

Five rules we hold ourselves to on this site.

• Name the study in the sentence. Not "research shows" — "Green and Brock, 2000". If you cannot name it, you do not have it.
• State N and method when the claim is load-bearing. Wyzowl's State of Video Marketing 2026 reports that 85% of people say they have been convinced to buy after watching a video; it is a survey of 266 respondents fielded in late 2025, and saying so costs you eleven words.
• Never round up, and never let a figure drift. Stale numbers travel further than fresh ones — that 85% was 82% in the previous edition, and the old figure is still being quoted as current all over the internet.
• Distinguish "the effect exists" from "the effect is this big." Nearly all the abuse in this field happens in the second half of that sentence.
• Prefer your own attributable outcome to any industry average. One named customer saying what changed, on camera, is more persuasive than every statistic on this page — and it is the only evidence a competitor cannot also quote.

Sources

• Green, M. C., & Brock, T. C. (2000). The Role of Transportation in the Persuasiveness of Public Narratives. Journal of Personality and Social Psychology, 79(5), 701–721. Four experiments, N=97, 69, 274, 258.
• Zak, P. J. (2015). Why Inspiring Stories Make Us React: The Neuroscience of Narrative. Cerebrum, 2015:2. PMC4445577.
• Bower, G. H., & Clark, M. C. (1969). Narrative stories as mediators for serial learning. Psychonomic Science, 14(4), 181–182. doi:10.3758/BF03332778. Twelve lists of ten nouns; yoked controls with equal study time; immediate recall 99.9% vs 99.1%; delayed recall 93% vs 13%.
• 2025 Edelman–LinkedIn B2B Thought Leadership Impact Report. 1,934 management-level professionals, seven markets, fielded 17 March – 3 April 2025 (7th edition).
• 2024 Edelman–LinkedIn B2B Thought Leadership Impact Report — for the 95:5 rule, via Ehrenberg-Bass.
• Wyzowl, State of Video Marketing 2026. 266 respondents, surveyed late 2025; 12th edition.
• Gartner press release, 19 February 2024, from Predicts 2024: How GenAI Will Reshape Tech Marketing (Alan Antin, VP Analyst) — traditional search engine volume, not website traffic.
• SeventhSlate first-hand count: 31 customer-story films delivered to date; one subject requested a change to the edit after review. Own internal record, not an audited sample. Cited above with its limits; we will update the denominator as it grows.
• Unverified claims above: searched for primary sources in academic databases and publisher archives; no originating study located. If you can point us to one, we will correct this page and say so.