Narrative Embargo

Alfred Hitchcock had a distinction he used constantly, and most marketers have never heard of. Put a bomb under a table and blow it up with no warning: that's surprise, ten seconds of shock, gone by the next scene. Show the bomb with a clock, then have two people talk about lunch: that's suspense, holding the audience for twenty minutes because they know something the characters don’t.

Brand marketing uses neither correctly. It defaults to a third, weaker mode: hide everything, announce nothing, then explain it all at once in a press release nobody was primed to want. Call it disclosure. It's efficient. It's safe. It produces no narrative charge, because nobody was given anything to hold onto while they waited. Disclosure is what happens when a brand mistakes the absence of risk for the presence of a strategy. Nothing can go wrong with a press release that says everything, which is precisely the problem. Nothing can go right with it either.

The useful withholding isn't about surprise or disclosure. It's Hitchcock's mystery: the audience receives a puzzle, not a shock or briefing, and is invited to solve it with the story, not handed fully formed. This distinction—not "scarcity" or "buzz"—defines what a narrative embargo is built on.

Why an unresolved gap does something a resolved one doesn't

George Loewenstein's research on curiosity makes an unglamorous point that most citations of it skip: an information gap doesn't feel good while it's open. It's closer to an itch than a pleasure. People act on it because it's uncomfortable to leave alone, not because uncertainty is fun. That reframes what a brand is doing when it embargoes part of a story. It isn't delighting anyone. It's putting a small amount of cognitive tension into circulation and asking people to carry it around until the brand decides to relieve it.

Which means the ethics of this are simpler than the usual "don't be deceptive" caveat suggests. The audience isn't owed protection from confusion. It's owed a resolution proportionate to the itch it was asked to carry. A brand that induces three weeks of speculation and pays it off with a font change hasn't been cute. It's taken something from people and given back less than it asked for. This is also why the standard advice to "build anticipation" is useless on its own. Anticipation for what, specifically, is the entire question, and most campaigns never answer it.

The gap has to be about something, not about nothing

Most teaser campaigns fail for a boring, structural reason: there's no reference point. "Something big is coming" isn't a mystery, because a mystery requires the audience to already know something specific enough to notice a piece is missing. Nobody can be curious about a shape they've never seen half of.

Cadbury's Caramilk Secret in Canada is a cleaner illustration of getting this right than most of what circulates in case studies today, because it's old enough that nobody's tired of it yet. The premise was mundane: how does the soft caramel get inside the chocolate. Decades of campaigns treated that manufacturing detail as a mystery worth solving, complete with hidden keys in bars leading to a vault nobody was ever meant to open. One activation period saw brand equity jump twelve points, the largest movement the metric had seen in a decade, and single-bar sales rose more than fifty percent year over year. The secret itself was trivial. The audience already had a reference point (they'd eaten the chocolate, they knew caramel doesn't behave that way) and the gap was specific enough to hold.

KFC Australia ran a leaner, more deliberate version. A set of off-menu items, built by staff, buried inside the app with zero official announcement, discoverable only through scattered clues in existing channels, including a frame or two hidden in outdoor ads. No press release. App downloads jumped over a hundred percent on the day the clues resolved, and a large share of the people who found the secret menu were still ordering from it more than a month later. Nobody was confused about what KFC does. They were given a specific, small, solvable thing to go find, which is a different and considerably harder brief than "generate hype."

Neither of these worked because the brand went quiet. They worked because the brand stayed specific while quiet, which is the part usually lost in translation when someone tries to copy the tactic without understanding the mechanism underneath it.

The part that gets left out of the framework slides

Here's what most breakdowns of "curiosity gap marketing" don't mention, because it's not a creative problem, it's an org chart problem: the reason most brands can't sustain an embargo has nothing to do with the campaign concept. It's that the moment speculation starts, someone internally reads ambiguity as failure. A stakeholder sees flat conversion numbers in week one and assumes the strategy isn't working, when flat conversion plus rising branded search plus a Reddit thread arguing about what the brand is doing is exactly what a functioning embargo looks like at that stage.

This has to be settled before the campaign goes live, in writing, with a named person accountable for holding the line: what gets confirmed if a journalist asks directly, what the social team is allowed to say when speculation starts (nothing, usually, though "interesting theory" carries a brand's whole voice in three words if done right), and what specific event triggers the reveal. Not a date on a calendar. An event. A date lets the reveal land whenever internal nerves give out, which is always too early. An event, like speculation crossing a volume threshold or a piece of press finding the wrong answer, lets it land when the tension has peaked.

There's a version of this failure that's less about panic and more about drift: nobody explicitly caves, but the embargo leaks through small decisions by those unaware of it. A regional social account posts early due to lack of briefing. A retail partner shelves product early. It's not sabotage, but the default when restraint isn't formalized as a rule with an owner, just an assumed vibe.

When the gap closes badly, or doesn't close at all

There are two distinct ways this goes wrong, and they get treated as the same failure when they're not.

The first is the never-closing gap. Valve never announced Half-Life 3 or said it wasn't coming, illustrating what happens when an information gap lacks an owner. For over a decade, "Half-Life 3 confirmed" became internet shorthand not for anticipation but for a promise unlikely to be kept. The gap didn't build goodwill; it turned into a joke and resentment, as an unmanaged gap erodes trust over time.

The second is the gap that closes on schedule and still fails, because the reveal itself becomes the story for the wrong reasons. A product unveiling can hit its date, satisfy every internal milestone, and still collapse in the room if the moment of resolution introduces a new, unplanned question the brand never intended to raise. That's a different mistake than mistiming the embargo. It's forgetting that the reveal is a performance with its own risk, not an information dump that closes out the campaign.

That's the boundary condition on all of this, more useful than the standard line about never embargoing pricing or safety information (true, but nobody serious was going to do that anyway). The boundary is: somebody has to own the gap, intend to close it, and control what happens in the ninety seconds after it closes. An embargo with no closing event isn't restraint. It's an unanswered question with a company's name attached, and companies don't get to leave those lying around indefinitely without the silence starting to mean something the brand didn't choose.

What this replaces, not what it adds

None of this is a case for brands saying less. Most brands should say the same amount they currently do about pricing, functionality, and what a product does for someone. What changes is what happens to the ten percent of the story that isn't required for a purchase decision: the meaning, the reveal, the reason something exists now rather than a year ago. That's the part currently being spent in one flat disclosure, on day one, alongside everything else, and it's the only part with any narrative value left to extract.

A gap that nobody's watching isn't a strategy. It's a job nobody finished.

‍