Part I, The Thinking · Chapter 1

The Three-Second Truth

How shoppers decide on a sales promotion

One idea, many logos

Between the start of July and the end of August 2026, anyone buying a kitchen or laundry appliance in Australia was walking into a very polite argument. Beko would give you a digital Visa card worth up to two hundred dollars if you bought a washer and a dryer together, claimed online with the serial number and the receipt by the end of September. Omega would give you a prepaid Visa card worth up to five hundred dollars if you spent five thousand, claimed by the middle of August. Buy a fridge from any of seven brands sharing one promotion and a knife set arrived by redemption. The sums were different. The sentence underneath every one of them was the same: spend the money, then get a bit of it back in a different shape.

That is not a competitive set. That is a category that has agreed on a mechanic and stopped arguing about it.

We do not say this to mock anyone. We have built offers like these and will build more. A bonus card on an appliance is a sensible thing, and in a category where the buyer is often not the user, where the person at the counter is furnishing a rental or replacing a dead machine on a Tuesday night or buying a present, a certain reward is exactly what a nervous purchase wants. The trouble is not that the gift card is wrong. The trouble is what happens when everybody on the same floor is right in the same way at the same time. The offer stops breaking a tie. It only stops you losing one. A brand in that position is paying full promotional freight for a defensive outcome, and the budget still leaves the P&L on time.

Two other promotions ran through the same two months and did something different with the same problem. Petbarn, for the tenth birthday of its own Leaps and Bounds range, offered five prizes of ten thousand dollars cash to members who spent forty-five dollars on the range in one transaction, entries open from 30 June to 31 August. Cash. To the human. In a category where the obvious prize is a year of dog food, the money went to the person who pushes the trolley, and the member card made every entry a record the retailer could use. Over at Mattel, a Barbie purchase with the receipt kept went into weekly draws from 9 July to 27 August for double passes to the Australian Ballet’s Nutcracker, with a bigger family experience in the final draw. The consumer is a child. The prize is an evening the parent takes the child to, which is a reward pointed squarely at the person holding the wallet, dressed as a reward for the person who is not.

Neither of those is a bigger gift card. Both of them cost real money and neither is automatically cheaper than a bonus card; a capped prize pool and an open redemption liability simply have different shapes, which is a conversation for chapter 4. What they have in common is that somebody asked a question the appliance floor had stopped asking: who, exactly, is standing in the aisle, and what are they actually weighing up in the few seconds they give us?

That question is the whole of this chapter, and most of the book.

What Is A Sales Promotion?

A sales promotion is a short-term offer that gives a shopper an extra reason to buy now rather than later, or this brand rather than that one: a prize draw, an instant win, a cashback, a gift with purchase, a bonus pack, a reward for coming back. In Australia the ones run as competitions are usually called trade promotions, and the ones decided by chance are trade promotion lotteries, regulated state by state, with a permit still needed in some places. This book uses sales promotion for the whole family. Its subject is the part the definitions leave out: why some promotions move stock and others only move paperwork.

The bottle that danced

About ten years ago we built a promotion for a drinks brand that we were certain would perform. It was beautiful. A bespoke app, augmented reality, point your phone at the bottle and a character climbed out of the label and danced on the table in front of you. We spent properly on the build and more again on media. The client loved it. We loved it.

Launch day, we sat around the dashboard waiting for the servers to groan. By five in the afternoon there were fourteen downloads. By the end of the week, a hundred and two. The cost per download does not need stating; you can do it in your head, and we did, repeatedly, for some weeks.

Nothing about that promotion was badly made. The app worked. The character danced. What never happened was the moment it was all for. Someone standing in a bottle shop at half past five on a Friday, drinks for the weekend in one hand and a phone in the other, looked at a shelf wobbler inviting them to download something, and could not tell in the time they were prepared to give it what they would get for the bother. They did not say no. They did not say anything. They picked up the bottle next to ours, or ours without the app, and went home, and the dashboard stayed quiet.

We had built the thing for the people in the room: the client, the agency, the awards jury that never got to see it. The shopper had not been consulted, and the shopper had the only vote that counted. We have a name for this in the office that is not fit for print, and a printable one too, which is the vanity trap. Everyone who has worked in this trade for long enough has a dancing bottle somewhere in the drawer. Ours happened to be the one that taught us to look at the aisle before we look at the idea.

How shoppers decide on a promotion

Stand in a supermarket aisle and watch people meet a promotion. It is over before you have focused. They see the flash, win a car, ten dollars back, scan to enter, and in a couple of seconds they have either reached for their phone or kept walking. Nobody reads the terms. Nobody weighs it up in any way they could describe afterwards. And yet a judgement has been made, and it is a surprisingly consistent one, and a lot of money gets compressed into it. Months of creative, a media plan, a prize budget, a retailer negotiation, all spent in the time it takes a trolley to roll past a wobbler.

Years of watching that moment, and then watching the entry numbers that followed it, left us with a rough picture of what is happening inside it. The shopper is asking three things at once, and the three combine in a particular way.

Do I want that? Do I believe I would actually get it? And how much of my time, and my patience, is it going to cost me?

We ended up writing it as a sum, because a sum is a useful way to argue in a meeting, and the sum has a name in our shop that has stuck. The 3-Second Equation: Reward, multiplied by Belief, divided by Friction. The shopper is not doing arithmetic. We are, on their behalf, to work out why they walked past.

The multiplication matters, and it is the part people most often get wrong when they repeat the thing back to us. Reward and Belief do not add up; a wonderful prize that nobody believes they could win does not come out as half wonderful. It comes out as nothing, because anything multiplied by zero is nothing, and nothing divided by any amount of effort is still nothing. A car in a single draw that reads as impossible is, at the shelf, exactly as persuasive as no car. That is not a judgement on cars. It is a statement about where the zero sits.

Reward is the term that feels easiest and is not. The instinct is that bigger must mean better, and it does not work like that at all. What matters first is whether the reward clears the bar of being worth wanting by this person, on this trip, against what they came in to buy. A reward can also be too small to carry its own claim. We started calling that point the Insult Threshold after watching a run of cashbacks die quietly: a two-dollar saving that wants a receipt uploaded and an account created is not a small win, it is a small annoyance, and shoppers answer annoyance with silence rather than complaint. And the reward has a direction as well as a size, which is the whole lesson of the appliance floor. A card worth the same as everyone else’s card has size and no direction. A night at the ballet for the parent has both.

Belief is the term brands most often forget they can move. A shopper looks at one big prize and concludes, without articulating it, not me, never me. The odds feel like zero whether they are or not. Three prizes on show and the same shopper thinks, possible. A hundred small ones and they think, probable. Same budget, a different sense of whether it is worth a go, and the entry numbers move accordingly. There is a whole craft of shaping a prize pool to change what people believe about their chances, and it has its own chapter later. For now it is enough to notice that the shopper runs this term against the way the offer is shown, not against the published odds, which they will never read.

Friction is where good promotions quietly bleed, and also where they are quietly funded, and chapter 4 is about holding both of those in your head at once. At the shelf it is simpler. Every field, every step, every create-an-account-to-continue is a small tax, and the taxes compound. But the useful question was never how few fields can we get away with. It is how much effort this reward and this belief have earned the right to ask for. A trip for four can ask for a receipt and a form. A five-dollar cashback cannot, and will be told so.

Hold the three together and the dancing bottle reads itself. The reward was a cartoon; wanted by us, unclear to them. Belief hardly came into it, because there was nothing to win. And the friction was an app download in a bottle shop, standing up, with the weekend waiting. Any one of those would have slowed it down. All three together produced fourteen downloads by five o’clock.

Three people in the aisle

The sum is run by a person, and it turns out there are three people to think about, and only one of them is physically in the aisle.

The first is the shopper: the one holding the wallet, the one whose three seconds we have been describing. The second is the consumer: the one who will eat it, drink it, wear it, feed it to the dog. In a lot of categories these are the same person and the distinction is a seminar topic. In the categories where promotional budgets run largest, appliances, pet, kids, gifting, beer bought for a barbecue someone else is hosting, they are different people, and only one of them is present when the decision gets made. The consumer research is usually the only research in the room when the mechanic is chosen, so the prize ends up designed for the user by default, and the user is not in the shop. The dog cannot read the pack. The child is not at the counter. The question that fixes this is not who is our shopper, which gets answered with a demographic and a shrug. It is: what does the buyer get out of this product if they never use it? If the honest answer is nothing, then the reward is carrying the whole load, and it has to be worth something to a buyer rather than a user. Petbarn answered that question with cash to the human. Mattel answered it with a night out for the parent.

The third person is not in the aisle on the day at all. They were there months earlier, in a category review, deciding whether your promotion would be in the aisle in the first place. The category manager at the retailer is measured on units, basket, incremental sales and whether your activity grows the category or just shuffles share around inside it, and they run their own fast judgement that has nothing to do with affection for your brand. Can a shopper understand this in three seconds? Can staff run it without ringing head office? Does it move units on this trip, for this buyer? We call that test S.O.S., for simple, operational, sales, and the person who applies it the Gatekeeper, and they have their own chapter in Part III because they deserve one. The point for now is that the three-second sum gets run twice before a single entry arrives: once by the shopper on the day, and once, in a different dialect, by the person who decided whether the shopper would ever see it.

Design for the consumer and you can lose both of the other two at once. Design for the shopper and the Gatekeeper tends to come with you, because a promotion a shopper understands in three seconds and acts on in sixty is also a promotion that moves units on the trip, which is the sentence a category manager wants to hear.

What the numbers can say, and what they cannot

We run and fulfil a lot of these promotions on the Trevor platform, so we can see the far side of the three seconds: every entry and every claim. You will not find one of those numbers in this book. They belong to our clients, and behind them are our clients’ customers. Every figure here about a real promotion comes from somewhere you can check for yourself: the published terms, the promotion page, the brand’s own announcement, a company’s results or independent research.

What running them teaches can be said without the numbers. The first lesson is that an always-on program and a six-week promotion are different animals. A gift-with-purchase offer that stays open for years will collect more claims than any short promotion could, and put the two in one table and every other figure on the page stops meaning anything. The second is that comparing mechanics across categories mostly tells you about the categories. An appliance cashback is claimed by people who have just spent thousands of dollars; a wine draw is entered by people who spent twenty on a Friday. The comparisons worth trusting hold the category still: the same brand, the same idea, the same few weeks, one setting changed.

What no count of entries can tell you is how many sales the promotion caused. Some of the people who claimed were going to buy anyway, and a receipt cannot tell you which. That figure, the extra stock that moved because the promotion was there, lives in the brand’s own sell-out data, set against the weeks before and after, and for most jobs it is the number a promotion is finally judged on; a promotion built for fame is judged on fame, which chapter 2 comes to. Chapter 15 is about getting the sales number. What entries and claims can show is which designs got the most people to act at the shelf, and a design that gets more people to act has more chances to sell something they would not otherwise have bought.

What all of that does support is the thing this chapter is actually about. The three terms are not three separate problems to be solved by three separate people. A prize team polishing the reward while a digital team builds the form and a sales team negotiates the shelf will produce a promotion in which each term has been optimised and the sum has quietly collapsed. The equation earns its keep by making you hold all three at once and be honest about which one is dragging.

What the sum is for

It is worth being clear about what the 3-Second Equation is not. It is not a predictor. It will not tell you the entry count. Promotions are a creative trade; you cannot always tell in advance what will work, and the occasions when the trade surprises us are the reason we are still in it. What the sum does is narrower and more useful. When a promotion feels off and nobody in the room can say why, it tells you which term has gone to zero. And once you know which term is dragging, you know what the move is, and the conversation changes from whether to run the idea to how to run a version of it that works.

Take it back to the appliance floor. A dozen logos, one mechanic, one answer. The reward term is identical everywhere, so it cancels; the belief term is as high as it gets, because the reward is certain; the friction is near enough the same claim form with a different logo. Everyone is running the same sum and getting the same result, and the only way to change the result is to change a term the others have not touched. Bidding two hundred and fifty against two hundred changes nothing except the shopper’s habit of waiting for the offer. Pointing the reward at the buyer instead of the user changes the direction of the first term. Showing three things to win instead of one changes the second. Letting the claim happen at the counter instead of at home changes the third. None of those is a verdict on the gift card. Each of them is a version.

That is the posture of this book, and it is as good a place as any to say so. Almost every good idea has a version that works. The craft is finding it, and the three terms are where we start looking. In our office that craft has always had a name, half a joke and half not: the dark art. It is not dark because anything is hidden. The web has a phrase, dark patterns, for designs that bury the catch, and that is the opposite of what we mean; every condition in a promotion we build is stated before the shopper is asked to want anything. It is dark the way any trade is dark from the outside. It is years of knowing which settings work together, which rules moved since the last job, and where the give is in an idea that looks impossible on paper. This book is the art written down, with the lights on. The next chapter is about the one thing a promotion should be trying to do, because the terms get weighted differently depending on the job. The two after that are about who the shopper is when they run the sum, and about friction, which turns out to pay for everything. Then the full board of levers arrives, five of them, and the moment a client puts something impossible on the table and asks us to make it work.

But it starts here, in the three seconds a trolley takes to pass a wobbler, with a person we have not met deciding whether we were worth the bother.