Nobody shares ads. They share stories they were part of.

Nobody wakes up hoping to share your advertisement. They wake up hoping to have something worth saying. That distinction is uncomfortable for marketers because it moves the job from persuasion to participation.

People rarely share advertisements. They share stories they happened to be part of: the unexpected invitation, the moment they looked clever in front of friends, the small public win, the ritual they can retell without sounding like a brochure. If your campaign cannot produce that kind of story, you are buying distribution for a monologue.

This piece is about designing for talkability-word-of-mouth as a designed outcome, not a hopeful side effect. Frameworks, examples, and a few exercises you can run before you spend another pound on creative that nobody will forward.

Ads ask for attention. Stories repay status.

Sharing is a social act. When someone posts your ad, they risk looking like a billboard. When they share a story they were inside, they look like a person with taste, timing, or luck. Design for the second identity.

• Status: does sharing make them look sharp, generous, or in-the-know?
• Emotion: is there a spike-delight, surprise, mild outrage, pride?
• Utility: can a friend act on this without a long explanation?
• Identity: does the story fit who they already believe they are?

Most startup “viral” plans fail because they optimise for the brand’s need to be seen, not the sharer’s need to be someone. That is why you get clever campaigns that die in the group chat. Related: why I care more about systems than campaigns.

A framework: moment → role → relic → relay

1. Moment: design one story-worthy beat

A campaign can have many assets. It should have one moment: the thing that happened. Not a theme. Not a moodboard. A beat with a beginning and an end that a human can retell in thirty seconds.

Examples of beats: “I got a custom roast named after my street,” “I fixed my onboarding live with the founder in fifteen minutes,” “I left with a map of three places that actually use this product.” Vague “brand experiences” are not beats. Beats have nouns and verbs.

2. Role: give people something to do

Participation is the difference between audience and cast. Roles can be light: choose, vote, co-create, invite a friend, complete a challenge, leave a mark. Heavy roles (full user-generated campaigns) fail when the ask exceeds the reward. Match effort to status returned.

1. Define the smallest meaningful action that creates ownership.
2. Make the action public enough to feel real, private enough to feel safe.
3. Reward with recognition, access, or a better outcome-not only a discount code.
4. Remove steps that exist only for your CRM comfort.

3. Relic: leave proof they can show

Stories travel better with artefacts: a stamped card, a photo of a named thing, a personalised URL, a certificate of absurdity, a physical object that looks intentional. Relics are not swag. Swag is logo merch. Relics are evidence of having been there.

4. Relay: make forwarding effortless

Design the handoff. Pre-written invite language, share cards that do not look like ads, a clear “bring someone” mechanism, a landing page that continues the story rather than resetting to corporate tone. For narrative continuity across surfaces, see marketing that feels like product design.

Word of mouth is not a channel you rent. It is a behaviour you design for-and then get out of the way.

Why most “UGC strategies” fail

User-generated content briefs often ask strangers to perform unpaid labour for brand awareness. That is not participation; it is outsourcing. People create content when it serves their identity, their friendships, or their craft. Hashtag mandates rarely qualify.

Better approach: design an experience so worth documenting that documentation is a side effect. Then make the documentation path frictionless. Ask less; enable more. If you need a 12-step UGC brief to explain the joke, the joke is not ready.

Examples: story-worthiness in practice

Airbnb’s early “belong anywhere” worked because hosts and guests already had stories; marketing curated and amplified them. Duolingo’s mascot chaos works because fans feel like co-conspirators in a running joke. Pop-up restaurants and limited drops create scarcity stories people can claim: “I was there on the first night.”

Fictional contrast: “Ledgerly,” a finance app. They launch a polished ad about “taking control.” Nobody shares it. Then they run a “money date night” kit for couples-prompts, a shared goal tracker, a photo booth prop that is slightly ridiculous. Suddenly people post because they look fun together, not because they love APR. Same product category. Different casting.

Designing talkability without being gimmicky

Gimmicks get attention once. Story systems get retold. The test: does the moment teach something true about the product, or could any brand swap logos and keep the stunt? If the latter, you built entertainment with a sponsorship sticker.

• Tie the moment to a real product capability or customer job.
• Prefer wit over spectacle when budget is limited.
• Make the story portable across cultures you actually serve.
• Avoid cruelty, embarrassment, or bait that punishes participants.

Measurement that respects human behaviour

Track referral invites, share rate, phrase mentions in support tickets and sales calls, secondary attendance (“brought a friend”), and qualitative retellings. Vanity reach still matters for paid amplification-but only after you have a story worth amplifying. Otherwise you are paying to distribute a shrug.

If growth is stuck, diagnose whether you have a story people can carry-or only a funnel. Start with If your product isn’t growing, start here. If you are pouring budget into noise, read before you spend £10,000 on marketing.

Building a repeatable story engine

One viral spike is luck. A story engine is operations: a calendar of moments, a capture kit, a response ritual for people who share, and a feedback loop into product. Treat talkability like activation-something you instrument and improve weekly.

This is also where product and marketing should stop arguing. Features that create shareable outcomes are growth features. Features that only look good in a changelog are inventory. For the deeper diagnosis when “marketing” is actually a clarity or product issue, see every startup looks like a marketing problem until you dig deeper.

What to read and watch beyond this essay

Jonah Berger’s research on why things catch on remains a useful checklist for talkability drivers-start from Contagious / Wharton resources. For practical brand community craft, First Round Review and Y Combinator’s Library still publish high-signal essays on distribution that does not pretend virality is a strategy. For behavioural UX that supports sharing flows, Nielsen Norman Group helps you stop designing share buttons nobody trusts.

How this shows up in my work at nau

I help founders design belief systems-product, web, and story as one. Campaigns that nobody can participate in are expensive monologues. Campaigns that cast the customer as the protagonist tend to need less shouting later. That is not mysticism. It is sequencing: story first, megaphone second.

If you keep hiring more marketing when you need a clearer offer, read what founders actually need isn’t more marketing. If your first spend went to the wrong layer, see the first £1,000 startups spend in the wrong place.

Cast your customer. Then get quiet.

People rarely share advertisements because advertisements cast the brand as the star. Design a moment, a role, a relic, and a relay. Give someone a story they happened to be part of-and then let them talk. Your job is to make that talk accurate, easy, and true.