I've spent twenty-odd years watching people make decisions they couldn't explain afterwards. Clients picking the wrong agency. Boards backing the safe idea. Customers buying the second cheapest thing on the list, every time, forever.
Marketing plans still get written as though none of that happens. As though there's a rational buyer at the end of the funnel, reading the options and weighing them up.
There isn't. There's someone tired, on a phone, half-watching the telly, trying to get out of the decision as fast as possible.
That's true of buying. It's true of consent too, and consent is now the most regulated choice in marketing, which means the sloppy version of this gets you fined rather than ignored.
Thirty-five of them below. Definition first on each, then what it actually does. Usual dyslexia warning applies.
Part one: the mechanics of choice
1. The default effect
The pre-selected option gets chosen far more often, simply because it was pre-selected.
Austria runs opt-out organ donation and gets near universal consent. Germany ran opt-in and sat around 12%. Same people. Same values. Different tick box.
That's the whole of behavioural science in one example, and if you only take one thing from this list, take that one.
2. Status quo bias
People prefer things as they are and treat change as a risk needing justification.
Every agency review I've ever been near, the incumbent starts about 30 points up for doing nothing in particular. I've been the incumbent and enjoyed it. I've been the challenger and felt the heat.
What loses the challenger the business is almost never the work. It's that nobody wants to explain to a CFO why they moved. Or that they feel safer or more comfortable staying put.
As a brand or business you have to demonstrate there's more danger staying put, than stepping away.
3. Choice overload
More options increase interest but reduce the chance anyone decides at all.
Iyengar and Lepper's jam study. 24 varieties pulled the crowd. 6 varieties sold ten times more.
I've sat through so many credentials decks with nine services on the capabilities slide. Nine. Nobody in that room remembers one of them. Curation reads as confidence. A list reads as insecurity.
4. The decoy effect
Add a deliberately poor third option and it changes which of the other two people pick.
The Economist's famous test. Web only, print only, or both for the same price as print. Almost nobody bought print alone, but it made the bundle look obvious and sales jumped.
Three tiers beat two. The middle one is rarely the one you're selling.
5. Anchoring
The first number sets the reference point for everything after it.
RRPs. "Was £199." The premium tier listed first. Reference pricing is regulated in the UK precisely because it works.
Here's the agency version. Show the rate card before you show the fee and you're negotiating from your number. Show the fee cold and you're negotiating from theirs. I've watched good agencies give away six figures a year by getting that sequence the wrong way round.
6. Framing
Identical facts produce different decisions depending on wording.
90% fat free outsells 10% fat. Same yoghurt.
"Save £20 a month" becomes "keep £240 a year". "We collect your data" becomes what they get back. Nothing changes but the words, which is either depressing or the entire reason our industry exists, depending on the day.
7. Loss aversion
Losing hurts roughly twice as much as the equivalent gain pleases.
"Your points expire Friday" beats "earn points now" every time. Trials convert on the day access disappears, not on the day the product proves itself.
Use it to protect something people already hold. Manufacture the loss and you'll get the sale, then the refund request, then the review.
8. The endowment effect
Once it feels like theirs, they value it more than they did an hour ago.
Test drives. 30-day returns. Trials pre-loaded with their own data. The wardrobe they built themselves and now won't throw out.
This is also why the best pitch tactic I know isn't a tactic. Do some of the work. Turn up having already made the thing. For an agency making a story board an ripomatic is gold. Doing pitch decks in the brand's own style, now - that's rocket fuel. By the time you leave the room it's half theirs and they're defending it to each other.
9. The zero price effect
Free isn't a very low price. It's a different, more emotional category of decision.
Free delivery beats an equivalent discount nearly every time.
Free also has a cost, and it's usually paid by the agency. Free ideas, free thinking, free chemistry sessions. I've written about this before and I'll keep writing about it. Stop giving shit away.
10. Present bias
Rewards now get wildly over-valued against bigger rewards later.
Buy now pay later. 0% finance. Start today, pay in January.
It's also why brand loses the budget argument to performance every single quarter, in every single business, regardless of what the evidence says. The finance director isn't stupid. He's just human, and next quarter is real to him in a way that year three isn't.
11. Sunk cost
People continue based on what they've already spent, not what they'll get.
Concorde. A loyalty balance. Three years with a supplier everybody's outgrown.
I've seen agency owners hold a loss-making client for years because of the history. The history is gone. It's spent. The only question is what the next twelve months look like.
12. The goal gradient effect
Effort increases as the finish line gets closer.
A loyalty card with twelve slots and two stamps already given beats an identical card with ten empty slots. People finish what already feels started.
Progress bars. Onboarding checklists. Show them how far they've come.
Part two: the social shortcuts
13. Social proof
When unsure, people take the behaviour of others as evidence of the right answer.
Reviews, ratings, logos, waitlists.
The cheapest conversion lift most brands are ignoring, and the one most likely to get you in trouble if you invent the numbers. The CMA went after pressure selling on booking sites for a reason.
14. Authority bias
We defer to signals of expertise rather than assessing the claim.
Nine out of ten dentists. The dermatologist. The name on the byline.
This is the whole engine behind founder-led marketing, and why a signed point of view outperforms an anonymous brand blog by an embarrassing margin. Borrowed authority is rented and fragile. Built authority compounds, slowly, for years, and then all at once.
15. In-group bias
People favour those they read as their own tribe and discount everyone else.
The strongest brands don't describe a product. They describe a group worth being in. Patagonia. CrossFit. Apple, back when it was still a bit rude about everyone else.
Get it right and your customers do your marketing for you. Get it wrong and you're a commodity with a logo and a tone of voice document.
16. The halo effect
One strong impression bleeds across every other judgement.
A beautiful site makes people assume the thinking is sharper. A typo in the proposal makes them quietly question the strategy.
Wildly unfair. Completely real. Your weakest visible touchpoint is setting your prices.
17. Mere exposure
Repeated exposure on its own increases liking. No new information required.
This is what brand fame actually is, underneath the language we wrap around it.
It's also the argument I've had a hundred times with marketing directors who are bored of their own campaign in month four, while roughly nobody outside the building has noticed it yet. Rebrand every eighteen months and you reset the compound interest to zero. Consistency is dull, and it's the closest thing to a free lunch in this industry.
18. The illusory truth effect
Repeat something often enough and it starts to feel true, evidence unnecessary.
Category claims. Political slogans. "The nation's favourite."
Which makes this a responsibility, not just a tactic, because the same mechanism is what makes misinformation stick. Repeat things that are true.
19. Negativity bias
Bad weighs more and lasts longer than good.
One angry review outweighs ten glowing ones. One missed deadline outweighs six quiet months of delivering.
Every brand or agency I've advised through a reputation problem had the same blind spot. They'd budgeted heavily for winning clients and almost nothing for putting things right when they went wrong. Recovery is marketing. Fund it like marketing.
Bad situations turned good are some of the most powerful moments in marketing. Never waste a crisis.
20. Reciprocity
Receiving something creates a felt obligation to give back.
The waiter who leaves mints gets bigger tips. Genuinely, measurably.
Real tools, real audits, an ungated paper, a useful chapter. The catch is it has to be an actual gift. A brochure dressed as a gift produces the opposite feeling, and people can smell the difference in about four seconds.
21. Commitment and consistency
Agree to something small and people act consistently with it later.
Email before card. Quiz before quote. Paid pilot before retainer.
Foot in the door research is decades old and it still runs most good funnels. A sequence of small honest yeses beats one enormous leap of faith.
Part three: belief and attention
22. Confirmation bias
People seek out what fits their existing beliefs and dismiss what doesn't.
Changing a mind costs about ten times what reinforcing one does. Nobody puts that in a plan, but it's the maths every challenger brand is up against.
So if you're challenging, don't argue. Demonstrate. Make people experience something that their existing belief can't explain. If you're the leader, your job is different and easier. Keep handing believers reasons to be right. When we launched Ovo, we focused on what people hated about energy companies. Complexities. No green energy option. Price.
Aligning against a customer's existing enemy gets you closer fast.
23. The availability heuristic
Whatever springs to mind easiest is judged most likely and most important.
Flood insurance sells after floods.
Categories get bought from whoever is mentally available at the moment of need, which is almost never the moment of your campaign. Being remembered beats being persuasive. It's also slower, more expensive and harder to sell internally, which is why so few people do it.
24. Ambiguity aversion
Unknown risk gets rejected faster than known bad odds.
Vague pricing. "Contact us for a quote." Woolly data policies. All of it reads as risk.
Apple's tracking prompt was a masterclass in plain English, and most people said no, because for the first time they understood what was being asked. That's not a failure of the prompt. That's what informed consent looks like.
25. Decision fatigue
Decision quality degrades with the number of decisions made.
By screen four of a consent flow, people click whatever ends it fastest. That's not consent. That's surrender.
Same in a pitch. The fourth agency of the day is being judged by a tired room that stopped listening properly at about 3pm. I used to fight hard to go first or last and never in that dead middle slot. It is not superstition. Just fatigue. That said, if you know you're bringing fireworks (sensory, not just content) any slot is the right slot. But first or last make you about 20% more likely to win.
26. The peak-end rule
Experiences are remembered by their most intense moment and their last one, not their average.
Which means your offboarding, your returns process and your final invoice matter more to your reputation than most of the service in between.
The end of the relationship writes the review. Almost nobody designs the exit.
Part four: pressure, permission and friction
27. Scarcity
Limited availability lifts perceived value and shortens deliberation.
Drops, waitlists, closed cohorts, ten places only.
It works exactly as long as it's true. Countdown timers that reset on refresh are an enforcement issue now in the UK and EU, and long before the regulator turns up, a customer has screenshotted it.
28. The deadline effect
Open-ended decisions get deferred indefinitely. Dated ones get made.
Proposals without a decision date drift for months. I've never seen one age well.
Put the expiry on something that genuinely expires. A start window. A pricing round. Fake urgency trains people to ignore you.
29. Confirmshaming
Wording the decline option so refusing feels stupid or mean.
"No thanks, I don't want to save money."
It lifts opt-in rates. It also turns a neutral visitor into someone who thinks you're a dick. If your yes needs a shamed no to work, the yes isn't good enough yet.
30. Sludge
Friction deliberately placed in front of an action the business doesn't want.
Nine-screen cancellations. Unsubscribe links demanding a login. Banners where Accept is one click and Reject is four.
Amazon settled with the FTC over its cancellation design at enormous cost. Regulators have stopped treating this as clever and started treating it as harm. Make leaving as easy as joining. You'll lose some people who were going anyway and keep the ones who'd have told everybody.
31. Reactance
Push someone toward a choice and they resist, even against their own interest.
Forced registration. Autoplay. The pop-up before you've read a sentence.
Pressure creates refusal. That's it. That's the bias.
32. The privacy paradox
Stated concern about data has almost no relationship to actual disclosure behaviour.
People say they care, then hand it over for a discount code.
Do not read that compliance as approval. Trust is ticking down quietly in the background and it surfaces later as dead engagement and a list that stopped working, by which point nobody connects it to anything you did.
Part five: the ones on our side of the table
33. False consensus
We assume everyone else thinks and behaves roughly as we do.
Marketers are not the market.
The room that loved the idea was full of people who work in advertising, in a nice building, looking at a big screen. Your customer sees it for two seconds on a cracked phone in a queue.
Burger King's mouldy burger was a great example of this. It got great applause from us agency lot - all the awards. But it was fucking gross.
34. The curse of knowledge
Once you know something, you cannot imagine not knowing it.
This is why category jargon survives in copy that nobody outside the building understands.
You've said the proposition four hundred times. They have never heard it once.
35. Survivorship bias
We study the winners and ignore the identical strategies that failed.
Every case study, award entry and best practice deck has this baked into it, including plenty of mine.
The bold move worked, so we called it strategy. Always worth asking what else was true. Who else did exactly the same thing and disappeared. That's where the real lesson is, and nobody writes that one up because there's no trophy in it.
The line
Every one of these is a tool. None of them is a trick until you use it like one. They're core programming.
The test I'd apply is simple. If you showed the customer exactly what you'd done and why, would they nod, or would they feel had?
Defaults that serve them (good things), nod. A reject button buried four clicks down (bad things), they feel had. The French regulator fined Google and Meta a combined €210m for the second one. Which tells you where this is heading.
All of this points to one big important truth. The better you know your customer and understand behaviour, the better the outcomes you'll get.
Choice architecture isn't manipulation. It's design.
People are going to use these shortcuts whether you plan for it or not, so the only real question is whose side the design is on.
Make the right choice the easy one.