Short answer: A cognitive bias is a predictable mental shortcut people use to decide quickly. In an ad script, each bias has a natural home: attention biases belong in the hook, trust biases in the body and conversion biases in the close. A bias stack is one bias per layer, chosen so all three tell the same story.
Lists of "persuasion tricks" are everywhere. Most of them miss the point that matters for a script: a bias that works brilliantly in a hook can feel pushy in a close, and a trust bias in the first second just slows the ad down. This hub sorts the 29 biases we use into the layer where each one does its best work, explains how to stack them and links to a full article on each. The brands in the examples are made up.
What is a cognitive bias stack?
A cognitive bias stack is a set of up to three biases, one for the hook, one for the body and one for the close, chosen to work together in a single ad.
Each part of an ad has a different job, so each needs a different kind of help.
| Layer | Job | Bias family | What the viewer is asking |
|---|---|---|---|
| Hook | Stop the thumb | Attention | "Why should I look?" |
| Body | Earn belief | Trust | "Why should I believe this?" |
| Close | Make action feel right | Conversion | "Why now, why this?" |
The layers line up with the arc every script follows. The hook is peripheral capture, the fast reaction in the first 3 seconds. The body is the central argument, where the viewer actually weighs what is being said. The close needs to work for both kinds of viewer, the one who thought it through and the one who did not. That arc is explained in peripheral vs central route, and the whole pipeline is in the Script Vibe system.
How do you stack biases in one script?
Pick at most one per layer, and make sure all three point at the same story.
The problem: A script with curiosity, authority, social proof, scarcity and loss aversion all crammed in reads like a late night infomercial. Each bias is fighting for the viewer's attention, and the viewer can feel the machinery.
The solution: One bias per layer, each one handing off to the next. Here is a stack for Groundswell, a $289 raised garden bed kit: a contrast effect hook, a conservatism body and a decision utility close.
"Last spring, a dead patch of lawn. This spring, salad."
"I'm not a garden person. I kill basil."
"But this is just a wooden box, like the one my dad built behind the garage."
"Except the corners slot together. No tools. Took me an afternoon."
"And there's a drip line built in, so it waters itself."
"The first time I walked out barefoot and picked dinner, I just stood there."
The hook shows two states side by side. The body makes something new feel familiar and safe. The close sells one specific peak moment instead of a feature. Three biases, one story.
A few stacking rules:
- You do not need all three. Many strong scripts use one deliberate bias and let the rest happen naturally.
- Match the mindstate. The same bias reads differently to a promotion buyer (chasing a gain) and a prevention buyer (avoiding a loss). Social proof is momentum to one and safety to the other. See promotion vs prevention.
- Never fake it. No fake scarcity, no fake authority, no fear without a fix. A bias that is not backed by something true is a lie with a label on it.
Attention biases: which biases belong in the hook?
Attention biases stop the scroll in the first 3 seconds. They work on reflex, before the viewer has decided to think. Our hook library covers the hook types these biases power.
Curiosity Gap
People feel a pull to close a gap between what they know and what they want to know. A good hook opens a gap the body actually closes, quickly. If the payoff never comes, you have trained the viewer to scroll past you. Example: "There's a reason every sleep mask lets light in."
Pattern Interrupt
The brain filters out the expected and notices the unexpected. A hook that breaks the pattern of the feed (an odd image, a strange first line, a surprising action) earns a second look. It must connect to the point, or it is just noise. Example: "I used to tape my sleep mask to my face."
Salience
People give more weight to whatever stands out most vividly. A salient hook zooms in on one sharp, concrete detail instead of a general problem. The more specific the detail, the more it feels like the viewer's own life. Example: "That thin line of light under your sleep mask at 5:30."
Optimism Bias
People tend to believe good outcomes are more likely to happen to them. In a hook, that means selling the transformation, not the tool: show where they could be, not what the product is. Example: "Picture picking dinner from your own backyard this summer."
Negativity Bias
Bad news grabs attention faster and harder than good news. It is powerful and easy to overuse, so we use it carefully: name a real downside, then make sure the ad resolves it. Example: "Your filter pitcher might have stopped working months ago."
Contrast Effect
We judge things by comparison, not in isolation. A hook that puts two states side by side (before and after, theirs and ours) makes the difference obvious without a single claim. Example: "My homemade cold brew versus this. Same ice, same milk."
Nostalgia
Memories of the past carry warmth and trust that new things have not earned yet. It works especially well for food, gifts and anything tied to home and family. The memory must be true to the product. Example: "My grandpa grew tomatoes in a box just like this."
Trust biases: which biases belong in the body?
Trust biases earn belief once the viewer is paying attention. They do their work in the central part of the ad, where the mechanism and proof live.
Authority
People give extra weight to sources they see as expert. Real authority (a credential, a profession, years of hands on experience) shortcuts the viewer's skepticism. Fake or implied authority is both unethical and easy to spot. Example: "I'm a night shift nurse. I sleep at 9 in the morning. Light is my whole problem."
Social Proof
When unsure, people look at what others like them are doing. The strongest social proof talks about the people, not the star rating. Real numbers only. Example: "My sister bought one after staying with us. Then her neighbor did."
Conservatism
In scripts, we use conservatism to mean framing the new inside the familiar. People resist change, so a new product described as a better version of something they already trust meets less resistance. Example: "It's just a pitcher. You fill it from the tap like always."
Labor Illusion
People value outcomes more when they can see the effort behind them, an effect studied by Ryan Buell and Michael Norton. Showing the process makes the result feel earned. Example: "They steep it for hours, then filter it twice. That's why it's not bitter."
Illusory Truth Effect
Repeated statements start to feel more true. In a script, that means repeating the core message in different clothes, not repeating the same sentence. Example: "It waters itself. I don't touch the hose. I haven't thought about watering once."
Processing Fluency
Things that are easy to process feel more true and more likable. One idea per line, plain words, readable on mute. Example: "No gap at the nose. Total black. That's it."
Tribal Belonging
This is our working label for the pull to belong to a group that shares your values or habits. When the speaker sounds like "one of us," the message gets past the guard the viewer keeps up for outsiders. Example: "If you're the person who reads the label on everything, you'll get this."
Choice Supportive Bias
People like to feel good about choices they have made. A script can validate the decision the viewer is already leaning toward, or the smart choices they have made in the past. Example: "You already switched to filtered water. You just couldn't see if it was working."
Conversion biases: which biases belong in the close?
Conversion biases make action feel right, right now. They do their work at the end, once the viewer believes. Used too early, they feel like pressure.
Loss Aversion
Losses feel stronger than equal gains, an idea from Daniel Kahneman and Amos Tversky. A close that names what the viewer keeps losing by waiting can move people a gain never would. Always pair it with the fix. Example: "That's another week of bad sleep I'm not getting back."
Outcome Bias
People judge decisions by their results. In a script, lead with the result and explain the mechanism second, because the viewer cares about the outcome first. Example: "Slept through the garbage truck. Then I figured out why."
Mere Token
This is our label for promising the first small win. A believable early win (the first night, the first cup) feels more reachable than a big transformation. Example: "Just try it the first night. You'll know."
Zero Risk Bias
People prefer removing a risk completely over reducing a bigger one. In the close, that means guarantee framing, but only when the guarantee is real. Example: "[Your real guarantee], so the worst case is you're back where you started."
Scarcity
Things in short supply feel more valuable. Real inventory limits beat ticking timers, and fake scarcity destroys trust. Example, only if it is true: "They make one batch a season, and this one's almost gone."
Anchoring
The first number people see shapes how they judge the next one, a finding from Tversky and Kahneman. Show the expensive alternative first. Example: "I was paying for café coffee every morning. One bottle of this is my whole week."
Framing Effect
The same fact feels different depending on how it is presented. Reframe the unit: per night, per cup, per meal. Example: "$38 once. Over a year of nights, that's about a dime a night."
Now Bias
People weigh what they feel now far more than what they will feel later, often called present bias. Close on what the viewer will feel tonight, not in six months. Example: "Tonight could be the first dark night you've had in years."
Goal Gradient and Zeigarnik
People speed up as they near a goal (Clark Hull's goal gradient) and remember unfinished tasks (the Zeigarnik effect). The close says: you're almost there, finish it. Example: "You already bought the pitcher. This is the part that makes it work."
Hindsight Bias
Looking back, things seem obvious. A script can use that to reframe past failures: it was never your fault, it was the thing nobody told you. Example: "Looking back, of course it was the light at the nose. Nobody tells you that."
Nirvana
The nirvana fallacy is rejecting a good option because it is not perfect. Name the imperfection honestly and position the product as the closest thing to perfect. Example: "Is it perfect? No. It's the closest I've found."
Overconfidence Transfer
This is our label for how a speaker's certainty passes to the viewer. No hedging, no "kind of," no "I think maybe." Certainty must still be honest. Example: "I'm never going back to regular masks."
Decision Utility
Daniel Kahneman separated the value people expect when choosing from the value they experience later. People choose based on the peak moment they imagine, so sell that one moment. Example: "The first morning I sat down with it and nobody talked to me."
False Self
This is our label for the idea that people buy identities. The product lets them become, or be seen as, the person they want to be. Example: "I'm the guy who grows his own food now. Who knew."
How do you choose the right bias for each layer?
Choose by the job, the mindstate and the product, in that order.
- Start with the job. If the problem is that nobody stops scrolling, the hook bias is the lever. If people watch but do not buy, look at the body or close.
- Check the mindstate. Promotion buyers respond to optimism, social momentum and identity. Prevention buyers respond to loss aversion, certainty and familiarity. More on the motivators underneath in emotional motivators.
- Check the product. Impulse buys can close on a fast peripheral cue like social proof. Considered purchases need a close with a real reason in it.
- Make sure it is true. If you cannot back the bias with something real, pick a different one.
What are the most common bias stack mistakes?
- Too many biases. The viewer feels manipulated even if they cannot name why.
- Conversion biases in the hook. Opening on scarcity or urgency before the viewer cares feels like a pushy salesperson at the door.
- Fake proof. Invented authority, scarcity or numbers. It costs trust, comments and sometimes the ad account.
- Fear without a fix. Negativity bias and loss aversion without relief feels cruel.
- Biases as labels inside the script. Creators perform labels stiffly. Keep the bias in the writing, not in brackets.
Key takeaways
- Attention biases belong in the hook, trust biases in the body and conversion biases in the close.
- A bias stack is one bias per layer, all three telling the same story.
- The same bias reads differently to promotion and prevention buyers.
- Every bias must be backed by something true. No fake scarcity, authority or fear.
- You rarely need all three layers. One well chosen bias beats five crammed in.
FAQ
Do I need to use a bias in every layer? No. Many strong scripts use one deliberate bias. The stack is a ceiling, not a target.
Can a bias from one layer be used in another? Sometimes. Loss aversion can power a hook ("I lost 2 years of sleep to this"). But each bias has a home where it works most naturally, and moving it usually takes more care.
Are some of these terms made up? A few are our working labels: Mere Token, Overconfidence Transfer, False Self, Tribal Belonging and Conservatism as we use it in scripts. Each linked article explains exactly what we mean. The rest are established ideas from psychology and behavioral economics.
Is using cognitive biases in ads manipulative? It can be. Biases are how people already make decisions. Using them to make a true message easier to understand is good writing. Using them to make a false message believable is manipulation.
Which bias should I test first? Start in the layer where your ads are failing. Weak hooks need attention biases. Ads that get watched but not bought usually need help in the body or close.
If you want to know where your current ads are losing people, we offer a free Meta account audit. It includes a full account audit plus net new reach and cumulative reach reports, so you can see which creative reaches new viewers and which keeps reaching the same ones. Request one at our contact page.
Check what you learned
Tap an answer to see if you got it right.
1. Where do attention biases belong?
2. What is a bias stack?
3. What is a cognitive bias?
Ready for the full test? Take the Buyer Psychology exam and get certified. Want this run on your account? Get my free audit.
Module 7: Buyer Psychology · article 1 of 30
Next: Curiosity Gap in Facebook Ads: How to Use It in Direct Response Scripts
About the author
Jake Mehani, founder of Creative Analytics
Jake has spent 15 years running Meta and paid social ads for consumer brands. Meta published his RBX Active Advantage+ case study, he taught the DMAnc Meta and TikTok ads masterclass, and he spoke at Motion's 2024 Creative Strategy Summit. His senior team runs media buying, creative and measurement like an in house team.