99% of Marketers Are Using AI Wrong (Including YOU)

Summary

---

Brain Driven Brands – "99% of Marketers Are Using AI Wrong (Including YOU)"

Host: Sarah Levinger
Date: November 6, 2025
Episode Summary

Main Theme

This episode dives deep into the pitfall most marketers fall into with AI: using it superficially rather than as a tool to challenge assumptions, drive real neuromarketing innovation, and supercharge experimentation. Sarah Levinger shares advanced neuromarketing tactics inspired by industry leaders such as True Classic, Spotify, and Plants vs. Zombies, with a core message: AI, when used to reimagine customer experience and behavioral psychology, can transform e-commerce brands—if you use it correctly.

---

Key Discussion Points & Insights

1. The Right vs. Wrong Way to Use AI in Marketing

• Superficial Uses Are Rampant: Most marketers use AI for basic tasks—writing headlines, simple copy, or automating routine to-dos—missing its transformative power.
  • “I want everybody to realize that there is a right way and a wrong way to use AI. … We can't stop talking about AI.” (01:07, Sarah)
• AI as a Thought Partner: Sarah advocates leveraging AI to break one's habitual ‘marketing truths’ and routines, and to explore what might be possible by challenging deep-seated beliefs.

2. The Power of ‘What Ifs’: Prompting Imagination (01:19–08:00)

• AI-Generated "What If" Scenarios:
  • Sarah demonstrates prompting AI (ChatGPT) to generate wild "what if" scenarios as thought experiments, both generally and specific to behavioral science.
  • These include:
    • "What if your memories were a subscription—you could pause, resume, or refund?" (02:10)
    • "What if gravity turned off for one random minute each day?" (02:16)
    • “What if emotions glowed as faint visible colors around people?” (02:24)
• Challenge Marketing Ruts:
  • These prompts help shake loose the biases and assumptions marketers accumulate, encouraging more innovative, science-backed CRO (conversion rate optimization) tests.
  • “I sometimes get stuck in this rut of thinking that everything I'm doing is true in marketing … I want to break my own processes often so I can get to things that would be fascinating…” (02:40, Sarah)

3. Translating "What Ifs" to Practical Brand Experiments (03:19–07:44)

Sarah runs through several actionable, AI-generated "what ifs" grounded in psychology and neuromarketing:

• Checkout Frictions & Motivation:
  • “What if checkout buttons had to start as a ‘no’ and only flipped to ‘yes’ after a micro reason was entered?” (03:45)
  • “What if every cart showed two clocks: one for time since you first wanted this, and one for time you'll spend replacing a bad choice?” (03:52)
  • Use of temporal discounting and regret aversion to drive conversions.
• Social Proof Reimagined:
  • “What if ratings were hidden until you wrote your own?” (04:03)
• Aligned Loyalty Incentives:
  • “What if loyalty programs gave you points for not buying when it’s off of your goal?” (05:12)
• Personal Behavioral Cues:
  • “What if product pages showed how often you return to similar items?” (07:06)

Notable Moment:

“If you can show [customers] what their own behaviors are, you have a much better chance to convert ..." (08:46, Sarah)

4. From Theory to Testing – Implementing Advanced Neuromarketing Tactics (07:44–13:31)

• Easy-to-Test Tactics:
  • Many ideas, like tracking individual revisit data or time-to-purchase, can be implemented with current e-commerce tools (e.g., Intelligems for Shopify).
• SaaS Inspiration:
  • “If you guys are thinking about making a software at any point, these what ifs are kind of fun questions to ask…” (07:52)

5. Notable "What If" Examples (08:57–11:47)

• Discounts Tied to Intent:
  • “What if discounts only unlocked after you explain how you would use the item?” (08:57)
  • Focuses on 'implementation intention'—helping customers articulate motivation, increasing commitment and conversion.
  • Quote: "You're basically sitting in a salesperson position and just being like, you tell me how to sell to you." (09:53, Sarah)
• Self-Referencing Calls-to-Action:
  • “What if CTA text always included the verb you chose earlier?” (10:33)
• Subscription Offramps and Reengagement:
  • “What if pause buttons asked, how should we greet you when you come back?” (11:12)
  • Leverages the 'Fresh Start Effect' for re-engagement in subscription models.

6. Overarching Principle: Make Marketing More Human, Not Less

• Personalization Over Automation:
  • “Most of you are using AI to be less personal and less human. … Think about your customers like a friend or a spouse.” (11:58, Co-host)
  • Sarah: “If you're only using chat to write headlines, you're doing it wrong. … It's very, very good at pattern recognition. It recognizes things that I haven’t seen yet.” (12:20)
• AI as a Lens for Empathy:
  • AI serves to expand human thinking, not replace it, driving honest, more trusting, and authentic messaging.

7. Trust Calibration & Authenticity in the AI Era (13:02–14:09)

• Imperfect Product Images:
  • “What if product images had to have one realistic flaw by default?” (13:02)
  • Promotes trust by showing authenticity and 'anti-AI' polish.
  • Quote: "People are already starting to be very distrusting of AI because it’s just, it's way too creepy." (13:20, Sarah)
• Emerging Platform Solutions:
  • Pinterest and YouTube are rolling out more robust controls to allow users to filter AI-generated content (13:47).

8. Initiation & Customer Qualification as Neuromarketing

• Creating a sense of "earning" the right to buy—turning shopping into a club or accomplishment.
  • “I'm super fascinated by like making people prove that they're worthy of product.” (14:09, Co-host)
  • Example: Requiring customers to demonstrate knowledge or intent before unlocking exclusive product access (whiskey club example, 14:36).

9. Final Takeaways: The Real Power of AI Is Unlocking HUMAN Brilliance

• AI is a thinking partner, not a shortcut:
  • “Please stop asking chat to think for you and start using it as a thinking partner to make you think more like you. Human intelligence, I think, is still underrated and underappreciated.” (15:43, Co-host)
  • “We need more humans in marketing... this is just a tool to get you to think more broadly.” (15:59, Sarah)

---

Memorable Quotes & Moments

• On breaking assumptions:
  • “I want to break my own processes a lot, and I want to do it pretty often so that I can get to things that would be like, wouldn't that be so fascinating…” (02:40, Sarah)
• On discounts unlocking through justification:
  • “We really only want to give good deals to people that align with our ethos… Can you just write us a sentence or two that shows us you're that?” (09:30, Co-host)
• On AI as a brainstorming partner:
  • "If you’re only using chat to write headlines, you’re doing it wrong. … Let’s expand your thinking! It’s very, very good at pattern recognition." (12:20, Sarah)
• On authenticity in the AI age:
  • “We need a feed with all verified human content… People are starting to be very distrusting of AI.” (13:20, 13:39)
• On making customers earn access:
  • "I’m super fascinated by making people prove that they're worthy of product." (14:09)

---

Timestamps for Key Segments

• AI ‘What If’ exercise & breaking assumptions: [01:07–03:18]
• Behavioral Science–based ‘What Ifs’ for Brands: [03:19–08:00]
• Implementation examples for e-commerce and SaaS: [07:44–08:57]
• Discounts tied to justification and commitment: [08:57–09:53]
• Self-referencing CTAs & subscription reengagement: [10:33–11:47]
• Authenticity and trust in the AI era: [13:02–14:09]
• Making the customer ‘earn’ the product: [14:09–15:28]
• Final call to use AI for expanded, not diminished, humanity in marketing: [15:43–15:59]

---

Tone & Style

Conversational, energetic, idea-packed, and direct—with a healthy dose of tough love for marketers stuck in the AI rut. The episode is loaded with humor, tangents, and dozens of creative marketing test ideas, all with practical, emotionally intelligent explanations anchored in neuromarketing science.

---

Actionable Takeaways

• Use AI to break your habitual thinking and uncover hidden assumptions in your marketing approach.
• Challenge yourself and your team to run "what if" experiments informed by psychology and behavioral science.
• Prioritize authentic, personalized, and humanized marketing—AI should augment, not replace, your empathy and curiosity.
• Test unconventional CRO and loyalty tactics grounded in real psychological triggers, not just standard playbooks.

Get the full list of "What Ifs" in the show notes and start experimenting!

---

Transcript

[A]

Three, two. Oh, you're recording already? Welcome back to Brain Driven Brands. What's up, Sarah? How's life?

[B]

Hey, how are you? Fantastic. I'm doing really well. I like these real quick episodes that we have to do because it keeps us on track. Otherwise, we just talk for three straight hours and then go. Oh, yeah, we were supposed to record a podcast. Sometimes a time limit.

[A]

Let's keep it short and sweet. Some might say it's tactical and practical, but, you know, we'll see.

[B]

Hey, you know, here's the thing. Is your podcast doing better than mine? Hours. Hours. Mine hours.

[A]

I mean, super interesting the way you asked that. I don't know. I mean, like, sponsor revenue is higher on mine, so.

[B]

So is. It's mostly because I've tried my hardest to, like, find time to go actively be like, hey, do you guys want to sponsor us? But then I've run out of time every single week. So shout out to our sponsor. Other side. You guys are amazing. Thank you for coming to us and being like, hey, we like what you do. Can we sponsor you? Yeah. Love that Somebody else come.

[A]

More of you are allowed to do that.

[B]

We're giving you permission to come in and say whatever.

[A]

All right, let's stop begging for money. What do you want to talk about?

[B]

Okay, so I want everybody to realize that there is a right way and a wrong way to use AI And I'm angsty about it because, yes, we can't stop talking about AI. So I want to show you guys one of the ways that Sarah likes to use AI and it has nothing to do with writing headlines or doing anything productive. And I'm going to show you a couple of the prompts that I used and some of the output from this so that you guys can see what AI can actually do if you push it hard enough. So this literally came out of. I can't even remember where I saw it, but there was something I was watching a couple weeks ago where somebody just kind of was asking AI what if XYZ happened and then tried to see what it would say. Right? So this. Can you see this? I hope you can see this. Okay, this is the conversation I had with AI the other day. I want you to come up with 50 wild and weird quote, unquote what ifs. That was the prompt. Just come up with weird what if situations. Some of the what ifs that it came up with are fascinating. So the first one, what if your memories were of subscription? You could pause, resume, or refund.

[A]

This feels like something that AI has planned for us someday. I don't like that.

[B]

Number two, what if gravity turned off for one random minute each day?

[A]

Well, I don't think gravity is real. We can get on into that on another episode.

[B]

What if you were on the toilet? That's my question. Okay. Number three, what if emotions glowed as faint visible colors around people? This would be incredibly helpful for me. Please can we get this? What if every object kept kept a whispering log of every hand that touched it?

[A]

Creepy.

[B]

That. Oh, Terry. So it gave me 50 of these. And I was like, oh, I love these. Because it's literally just like, let's think of all the things that are true. Ish or true things in the world and then think about what if that wasn't a thing? Now, the reason I asked Chat to do this is because I sometimes get stuck in this rut of thinking that everything I'm doing is true in marketing. And it's not like my fault. This is just how the brain works, right? It's very biased. But I want to. I want to break my own processes a lot, and I want to do it pretty often so that I can get to things that would be like. Wouldn't that be so fascinating for us to test something that we think is true, but it just isn't at all?

[A]

Yeah.

[B]

So the next thing I did was I asked Chat and I just said, hey, these are cool. Can you do it for this topic? And I asked it to do it for behavior science. So here's some of the what ifs that you could test as a brand just based upon breaking your own psychological beliefs. So number one, what if checkout buttons had to start as a yes? No. Right. It started out as a no, and then it would flip to a yes, but only after like a micro reason is entered.

[A]

Whoa.

[B]

So you can't get to the checkout unless you ask, do you want this product? Yes or no.

[A]

Interesting.

[B]

Interesting. Test number two. What if every cart showed two clocks? The first one said time since you first wanted this and time you'll spend replacing a bad choice.

[A]

I love this one.

[B]

Isn't it interesting?

[A]

I'm like, oh, this would be so much more effective than like a countdown time.

[B]

Yes.

[A]

Would be a clock of like, do you know that you've wanted to buy something like this for 14 months?

[B]

Oh, my God, 100. Especially if you tracked how many times they came back to your website. You were like, you've been on this site a lot, honey. It's just buying, like, just go into. Isn't this fascinating? This is using temporal Discounting and regret aversion. So it's giving me all of the psychology around this too. But this is why I asked Chad to do it, because I'm like, look at all these frickin tests. Number three, what if ratings were hidden until you wrote your own? I kind of love this one because it won't, it won't change what you think, right? Social proof changes what I think about this product. Good or bad. It's going to sway me one way or the other. What if all of our social proof was hidden and it just said to see our reviews, go ahead and drop us a line about like what you think about this product, what you like about it, what you don't. Well, love that one. Number four, what if loyalty programs give you points for not buying when it's off of your goal, right when it's outside of what you're trying to achieve as a customer?

[A]

You know, that last one is something I've thought about with Adapt so far, like dugging too much into it.

[B]

Let's pause for just a minute.

[A]

Prospecting on Meta in 2025 has been a huge pain. Reaching net new people is getting harder and harder and harder and the only thing Meta seems to care about is rolling out more a slop features that take away a lot of our controls as advertisers. So brands have been looking for another side and they're turning to otherside. Other side is an agency that figured out how to turn programmatic advertising into a performance marketing channel. They're able to advertise outside the Meta eco system, show clear and accurate attribution and they have the ability to prospect and retarget across platforms so you're able to unlock huge new audiences that aren't like us. They're not chronically on Meta and Instagram. For brain driven brands listeners, they're running a crazy offer right now. They'll run your ads for free for two months to prove to you it works. Then you can either part ways or continue at their normal affordable retainer. But they're making this risk free and it's a no brainer for you to try. So go to jointheotherside.com and let them know we sent you.

[B]

And now back to the show.

[A]

We want to be like a health partner for our customers.

[B]

Yeah.

[A]

And like I've thought about like, what if we said like, hey, before you buy from us, you have to prove to us that you're eating right, sleeping right, exercising often because we want you to be fully healthy, not just be a customer. Of ours.

[B]

Exactly. We want to see whole health. Right. Like after your money, do we earn.

[A]

So much more trust and credibility?

[B]

Oh, yes.

[A]

Like that.

[B]

Yeah, 100%. So this one's really interesting. Right. So you get points basically for just saying, this product isn't what I need right now. You get points for that. That's oh, so interesting. Number five. What if product pages showed how often you return to similar islands? Not items, not how many times other people are seeing this, but just like you've looked at this item 20 times or whatever.

[A]

I think that's the most powerful one here.

[B]

Yeah. I want to see this tested and.

[A]

The time you wanted it.

[B]

Yep.

[A]

Gosh, I wonder if there's any way to do that right now.

[B]

I know, I'm like, I'm like racking my brain. Is there a tool out there that can do these type of CRO level tests? Because I'm like, oh, some of these might be hard to implement, but a lot of these you could do pretty easily, I think.

[A]

I think intelligems in Shopify could get pretty close.

[B]

Oh, you think so? Oh, I love intelligence. Oh, see, this is why I love those guys because they're doing some really interesting tests over there. If you think it could do it, we should get intelligence to try it. Because I'm like, oh, I'm gonna email.

[A]

Them as soon as this podcast ends. Okay.

[B]

I will send all of these to you because I have. We did 50 in the first round.

[A]

Yeah.

[B]

And then just ask that you choose top five. But you could just keep going and going and going with these.

[A]

Like, by the way, if you guys are thinking about making a software at any point, these what ifs are like kind of fun questions to ask if you can solve.

[B]

Yes. Oh my gosh, yes. Somebody come take this list and just build a crap ton of sass from it. Because I'm like, these are the type of tests that I want to do because they're based in. In science. Right. Because I asked it specifically do a lot of. Lots of what ifs around behavior science.

[A]

Yeah.

[B]

But I, I really want it to tell me what do I think is real, but it just isn't.

[A]

Yeah.

[B]

Because a lot of times things like this. Right. Showing people how often they return to similar items, that's fairly easy to do and implement from a tech standpoint, but it's so behaviorally focused because it shows people a mirror. I don't really give a shit about what other people are doing.

[A]

Right.

[B]

Social proof is important. Yes. But outside of reviews, I. I don't really need to know what other people doing. I'm trying to get my problem solved if I'm a customer.

[A]

Yeah.

[B]

And if you can show that to me, you have a much better chance to convert to me on the back end just because you're showing me what my own behaviors are. So interested in these. Okay.

[A]

What I want to call out here. What if discounts only unlocked after you explain how you use them or how you would use the. The item Again, I think that's like such a crazy one that a bunch of brands could. Could do it and then be like, hey, listen, we will give you a better deal, but we really only want to give good deals to people that align with our ethos and our, like, core believers in whatever our brand stands for. So can you just write us a cemetery or two that shows us you're that?

[B]

Yes.

[A]

And we'll give you the better deal.

[B]

There's so much psychology in here because implementation intention is. Is a really interesting psychological process which we'll have to do a whole episode on. But. But this comes down to, like, prove it to us that you're going to use this product. You're disqualifying people who are just like tire kickers, basically. You're also helping people understand their own motivations, which helps them commit.

[A]

Yep.

[B]

Commit. Guys, Decisions are really interesting in the brain. It takes a long time for people to kind of work out what they think about things and how far they want to go into this process. And if you're trying to get people over that line, if you're trying to increase aova, increase ltv, or get people to convert in general, half the time it really comes down to explain to us why you want this. You're basically sitting in a salesperson position and just being like, you tell me how to sell to you.

[A]

I was going to say, and think about the ad copy inspo you would get.

[B]

Yeah.

[A]

From forcing people to, like, prove that they are worthy of your product.

[B]

Blows my freaking mind. I can't even get off it. Okay, how many more of these should we. Do you want to see more? There's like a lot of them.

[A]

So I'm like two more.

[B]

Okay, here we go. What if CTA text always included the verb you chose earlier? So for instance, you clicked on a button that says build my kit. Then the CTA that would go into like, let's get this build started or whatever it is. Right. So you match the two. I kind of love this. Call my mind. Yeah. So this is self reference effect. You're basically referencing Something that you chose prior as the actual, like, end goal.

[A]

Love that. I'm going to choose the one under it. Super relevant for subscription brands, which is where I work now. What if pause buttons asked, how should we greet you when you come back?

[B]

Oh, that's this one especially because you guys are subscription based. So I'm like, oh, re. Entry cues, Fresh start effect. Yeah, yeah, I know. Like so many of these.

[A]

This is such an interesting concept that I think people can message into.

[B]

Yes, yes.

[A]

Yeah, because, like, and like, we, you know, like, that's the project we're gonna have you work on with us. Like, all right. Like when. When people are on their way out, is there anything we can do to get them back? Is the right thing to do, to wait a month or two before we try to get them?

[B]

Yes.

[A]

Yeah, that's gonna be super impactful. I love that one.

[B]

This is the type of stuff that I'm like, this is. If you're going to use AI correctly, this is how you should be using it. Right. Well, to challenge your beliefs on things.

[A]

Do you know what every one of these is suggesting to us too, is to be more personal with our customers. And like, most of you are using AI to be less personal and less human. And that is not what you should be doing at all. Every one of these is basically just being like, hey, think about your customers like a friend or a spouse. Like, that's what it is.

[B]

Oh, my God. Yes. Well, I. I love this list. And this is the reason why I'm like, ah, guys, you're only half using chat. If you're only using chat to write headlines, you're doing it wrong. Like, there's so many ways that chat can come in here and say, look at all of these interesting ideas. Let's like, expand your thinking a little bit because it's very, very good at pattern recognition. It recognizes things that I haven't seen yet. I would have never, like, thought to look at some of these, especially, like, pause buttons. I. How would I come up with that? Right, yeah. I'm a smart human being and I'm like, there's no way that I would have been able to come up with that. I love this. Okay, one of the ones I want to call out down here, number 43. What if product images had to have one realistic flaw by default?

[A]

Oh, I like that.

[B]

Trust calibration. Now, the flaw doesn't have to be on the product. It could be on anything. Right. A messy kitchen countertop. Right. Realism.

[A]

And again, like, I think this is an anti AI sentiment here of like, let's be real and authentic.

[B]

Yes. Yeah.

[A]

And like, not be so polished, which.

[B]

Is good in today's day and age because people are already starting to be very distrusting of AI because it's just, it's way too creepy. Okay, you. Yeah, go ahead.

[A]

I am so sick of scrolling Instagram already.

[B]

Like, God, it's so bad. Instagram. Tik tok. Like, you can't go anywhere like I without seeing it.

[A]

Yeah. Like, we need to hit a button somewhere that says like, no, no, no. I, I, I need a feed with all verified human content. Like, it's so much.

[B]

Did you know that Pinterest just launched this?

[A]

Did they?

[B]

They have personalized control in the back end where you can turn off Gen AI for like 17 different topics.

[A]

Yeah, that'd be great. That'd be great.

[B]

YouTube has very, very strict policy now. I think it's like 10% of your video can be AI and the rest of it has to be real, otherwise they won't let you. Good. All right, so you're gonna go test some of these. I can give you this whole list if you want.

[A]

I'm super fascinated by like making people prove that they're worthy of product. Like, I think that's like, to me, I think about like when I'm planning like a golf trip with like the boys, like, I'm like, all right, are you actually gonna be good at golf? Are you gonna be fun at the bars that night? Like, like, you need to contribute before we let you go on this trip.

[B]

Yep. You gotta do initiation first.

[A]

Yeah. And like, I feel like if, if you message like that to your customers and like, I might do this for the, the whiskey hat, like hat brand. It's like, hey, like, I really only want to sell that hat that says Chief Whiskey on it. Chief Whiskey Officer on it, to people who like, know their about to Chief Whiskey Officer. Like, I'm not giving you the C suite whiskey.

[B]

You can't just get it.

[A]

But just because you want the hat, like toxic. Like, what's your blood alcohol count right now?

[B]

Oh, so smart. So smart.

[A]

It's super.

[B]

Yeah, I like that one.

[A]

And then I am going to tell intelligence. Like, can we put the opposite of a countdown timer?

[B]

Yes.

[A]

The first time you came to our website was 39 days ago.

[B]

Yeah.

[A]

Aren't you sick of feeling like that?

[B]

You are going to make so much money. I'm upset about it. I'm like, okay, if you guys want this list, I will actually include a link to it in the show. Notes. You guys want to go test this for yourself? Please start using AI Better, guys. AI is so freaking powerful.

[A]

Yeah. Please stop. Please stop asking Chat to think for you and start using it as like a thinking partner to make you think more like you. Human intelligence, I think, is still underrated and underappreciated.

[B]

Yeah.

[A]

So please use your brains.

[B]

Please go use your brains, guys. Like, we need more humans in marketing this. This is just a tool to do this. Just to get you to think a little bit more broadly. And that's it. Like, I love it.

[A]

Thanks for listening to Brain Driven Brands. See you guys next time.

[B]

Thank you so much for joining us on the show today. Appreciate you guys listening. If you want to follow me, I'm at Sarah Levenger. Anywhere you can see content. He is at Nate Legos. If you like this show and if you like this episode, go ahead and like, subscribe. Share with a friend. Drop us a review when you have a minute. We would appreciate it. Otherwise, have a great week. We'll see you next time.