Why Jeffrey Epstein Paid For SEO

All right, Neil. So, you know what we're going to talk about? We're going to talk about 10 business moats that actually protect your marketing. Okay. So the first business moat we're going to talk about today, the first one is switching costs. And what switching costs are is when you think about, there's a, there's a great strategy book out there called The 7 Powers. And one of the powers in there is switching costs. And so one example of this would be, let's say you have an agency and you actually happen to offer a bunch of different services. Okay. So you don't just offer SEO, you offer SEO, paid media, CRO, ad creative, and a bunch of other things, right? Like influencer marketing, everything, the full stack. That makes it harder even if you're performing at a B- across the board. Even if you're number 1 at SEO, people are going to rather— they're going to— they want to go with the provider that can provide more, right? They want more full service. And so one of the moats that will protect your marketing is switching costs, right? And so when you think about— because that is protecting your business ultimately, and these things kind of go hand in hand. So that's number 1.

Number 2, Funny enough, I wasn't going to look at the list, but my number 2 was similar to the list. I think it's the same thing. I think in marketing, one of the biggest most is your brand because you've heard the saying, you don't get fired for hiring IBM. You can say the same for McKinsey, Accenture, Deloitte. Look at Deloitte and all the government projects that didn't come to fruition. I'm not saying they suck or they fail.

$74 billion.

But it's a reality and they keep getting hired.

Why?

Because Deloitte, has a really powerful brand. If someone hires you and you have a really strong brand and they don't get the results and everyone in the industry knows of your brand, the person, their boss will be like, well, you hired a really good person, really good company. So, eh, you try. If you hire a company that's not known, really tiny, unknown brand, and something goes wrong, they'd be like, why'd you hire this company? No one knows them. They don't have a proven track record. This is risky. You're taking a gamble here. Brand carries a lot of weight. It takes 10+ years to really build a strong brand, but that is a strong moat in business and marketing.

And you can also erode that brand in seconds. This is what Warren Buffett said, right? It takes a lifetime to build a brand. It takes minutes to destroy it. So be careful.

Yeah, and in his annual letters, do you know what he used to always say to people when it came to their brand?

What?

If you have a decision to make that would cost you short-term profits or long-term brand, like to hurt your brand in the long run, take the short-term revenue hit, like never do anything to mess up the brand.

Another one would be, this is digressing a little bit, but if you wouldn't even hold it for 10 seconds, if you want to hold it for 10 years, don't even hold it for 10 seconds. That relates to investing. But anyway, I digress. So the next one is on process power. And so what process power is, is the actual strength of the process you have.

Okay.

So I'll give you an example. So Toyota is known for their process power. When everyone was making— everyone still makes a lot of cars today, but everyone was looking at Toyota like, how are they so efficient? How is it that they can crank out so many cars than other people before? Nowadays, I mean, Tesla's kind of really up there when it comes to manufacturing, but also process power ties in with your marketing too. How you go about creating your content, how do you think about strategizing for, you know, your long-form content? How do you think about redistributing it? How do you think about chopping it up? Right. That can tie in with your marketing, but also process power for your business too. It could be how you do your work for your clients, how you interact with your clients. All that stuff matters. And then how do you make sure you hold people accountable so the process doesn't start to fall apart, especially if you're a very people-oriented business? Yeah.

Number 4, community. If you can build a really strong community, you'll have a really loyal following of evangelists who'll keep promoting your brand over others. That goes a really long way. But the key with the community, and this is why it's a good moat as long as you do it right, is you got to actually care for a community and be there and continually hear them out and adapt based on what they're saying and their needs are. You can't just have a community, it be really strong, and then all of a sudden if your community's giving feedback on something, you just ignore it. A great example of this is Digg back in the day. The Digg users were part of the Digg community. It was owned by Kevin Rose, extremely popular.

It was like Reddit, guys.

Yes. And he made changes that people really hated and a lot of the community left. and they went to Reddit and they didn't care to come back. And eventually Digg fell and Reddit is now a multi-billion dollar company.

Fun fact, Neil used to be a super digger. I was like, that was like a little before my time. But yeah, Neil's a super digger. He was like always on top, right? Yeah. Anyway, next one would be network effects. And so network effects, when you think about using like Instagram or you think about using WhatsApp, the more people that use the network, the more valuable it becomes. And so not every business necessarily has network effects, but I do believe that now that most people can build things, I think we're going to start to see more of these mini networks. And a community, you can say, is a mini network. But I do think that because you and I, anybody can vibe code whatever we want to build now, I think we're going to start to see more networks come up. So number 6 is you.

All right, number 6, speed. The quicker you can move, the better off you're going to be. It's shocking how slow people are, even with AI. You have all these tools at your fingertips. There's no reason why you just can't get stuff done faster.

See, isn't AI so good?

Yeah.

Okay, next one is 7, I think. So because so many people right now are able to vibe code whatever they want, because you, me, my, you know, my dad, I guess, they can— you can vibe code anything you want now. And people think that whatever you build, people are just going to come to it. That's not the reality. Distribution often beats product. And distribution tastefully, I'm not saying just go and run ads, right? Like that's certainly powerful. You gotta think about your creative, all these different things. You gotta think about where your audience is hanging out. And so any, and I'll actually flip both too, 'cause I do think, and we'll talk about the SaaSpocalypse in a little bit, but I do think in the next couple years or so, a lot of software is just gonna be given away for free, and that's gonna be the way of getting people in the door. And so in some cases, product kind of becomes distribution, but distribution is still kind of the lead domino. And you think about, I'll just use you as an example, Neil. The reason why in my mind MP Digital was able to grow so quickly in the beginning was because of distribution.

Yes.

Yeah. It wasn't because you had some kind of— the distribution was the secret sauce in my mind.

Yeah. And I would say number 8.

Yeah.

I would say a big thing is network and not network the way you described it. That is an effective moat as well. But I'm talking about who do you know? If You work with amazing people on your team and you yourself have been in the space for a while or have been networking for a while. If you have good strong contacts with people who could be your ideal customers, it helps the business grow much faster. People rather do business with people they know and like, the like part is really important, than random strangers. So build up your network. Eric and I have been networking for, call it 20+ years, and a lot of the deals we work on today are from people we connected with 10+ years ago. Your network truly is your quote unquote net worth that people have been saying for so many years. It's just more so you got to focus on not just building a network, but building those relationships with people within your network. So when it comes time to doing business, they're much more likely to do it with you than some other stranger.

Dude, let me show you this real quick. So you can see on my screen over here, this is my Telegram bot, right? I call him Alfred, like Alfred from Batman. He has a little butler hat too. Look at that.

Okay.

Right. So, So anyway, you can see it's like, oh, who this person is over here. Okay, they're in this ABM role, 3 years over here, spend is $300 grand a month, $3.6 million a year on ABM stuff, okay? Discovery call was like this time, explicitly said she's figuring out a gap, uses these tools over here, here's the gap. So it basically, talking about relationships, this is my ultimate relationship hub, right? And then not only that, look, it crafted some copy here. Hey Neil, great connecting last Friday. It's been over a year, so it was a little wrong there. Been thinking about what you said regarding the ad personalization gap, which is what she said on the call, right? You mentioned exploring competitor and want deeper Sixth Sense integration for landing pages. That's literally Karat's sweet spot. Would you be open to a quick 15-minute demo focused specifically on the one-to-one landing page personalization you asked about? And then here's the context over here. Key question, targeting enterprise accounts greater than 1,000 employees. And so this is a good example of the— it's not just building your network, the relationships, like it's also how do you mine it, right? And so when Neil told me about how he's been just making sure that he's proactive about it, I'm like, I have very bad memory. I just need something to be proactive for me. So every day I get a deal of the day thing. It's kind of like, remember, what's the website? Was it Whoop? Or not Whoop.

Woot.

Woot, yeah, yeah. Every day you get a deal of the day. Yeah, you get a deal of the day.

New TV half off, cell phone half off, headphones 1/10 off.

Exactly.

Or 1/10 the price.

That's how I see it. So anyway, I digress a little bit, but the next one, I think number 9 from my side, really this one's a toss-up. So I'm going to go with I'm going to go with talent moats. No matter how good, you know, your distribution is, how good my distribution is, everything ultimately business moat-wise comes down to the people. I don't care if you have a product business, I don't care if you have a service business, everything comes down to how good your team is. And I remember there's Andy Grove's book, High Output Management, and you know what that book is about, Neil?

What? Confrontation.

The entire book is about confrontation basically. If you're not willing to confront, it doesn't matter how good you are at, you think you are at hiring ultimately, but you're going to cap yourself out. And one thing I will say, that, you know, since I've known you in the beginning, you've always been good at confrontation. I've gotten better over the years at confrontation, but talent moats in my mind are the hardest to copy because if you have bad talent managing your marketing or your business, everything crumbles down. And you know what Warren Buffett and Charlie Munger used to say? They say, make sure your business is easy enough to run by an idiot because eventually an idiot will. I hope that doesn't happen to us, but yeah.

Hopefully not.

Yeah, maybe I'll just stay the idiot. Anyway, go ahead, continue.

Last but not least, data. In today's world, data's more fragmented. It's all over the place. Networks don't give you all the data, sadly. And the more first-party data that you own, the better off you're going to be. You saw the deal that Amazon has, right? Netflix, is it Roku? Who else is in there? Disney, I believe, is in there.

Yeah.

They're all partnering with Amazon so Amazon can run ads for them on their platforms for their streaming. Why is this important? Amazon has so much data on what people are buying. They can connect it all in and serve better ads. It monetizes better. Advertisers will pay more. It's a win-win for everyone, including the Disneys of the world.

Yep. Okay. So next topic here. So Neil, I'm going to give— I'm going to read you some examples, human versus AI here, right? So this is where I want to talk about how Claude Code makes you a one-man marketing army. So I'm going to show you some examples and you tell me what you think about these examples.

Okay.

You having a marketing brain. And so So here's the first email here. You can jump in here too, Noah.

So wait, you want me to rate 1 to 10, like what's good or bad?

Yeah, I'm going to read you 2 emails. Are you ready? Okay, it'll be quick. Okay, so one is AI generated, the other one's human generated.

Don't tell me which one is what.

Okay, I won't. And then I'll do the same thing for you. Okay, ready? Okay, here we go, guys. And then this is the— then guys, the topic here is how Claude Code can make you a one-man marketing army. Okay, so first email here, the subject line is Salesforce's LinkedIn ads are haunted. Okay. Preview of the post, the little preview by the ghost of impressions that don't convert, quote unquote. Hey Neil, I have a confession. I've been looking at Salesforce's LinkedIn presence for the last, the past 20 minutes, not in a creepy way, in a quote unquote, wait, they're doing ABM wrong and they probably don't know it, quote unquote way. Quick question. Do you know what percent of your target accounts are actually seeing your LinkedIn ads? If the answer is impressions, that's exactly my point. Most ABM teams are paying for eyeballs that will never buy. It's like running a Super Bowl ad to sell B2B software. Expensive, feels good, doesn't work. I built something that might change how you think about this. But first, am I totally off base here? Is Salesforce crushing it on LinkedIn ABM and I'm the one missing something? Eric, say. So that's number 1. Rate that one. Do you think that was human generated, AI generated? Rate it 1 through 100.

1 through 10 is easier.

Okay, 1 through 10.

Yeah. I think it was AI generated and I would rate it as a 6.

Okay.

I also think AI generated and I would put it like a 5.

Okay. So here's the second one. Okay. Subject line is 387 ads in 12 minutes, not a typo. Preview, the math doesn't make sense until it does. Neil, I timed myself yesterday. It took 12 minutes and 47 seconds to create 387 personalized LinkedIn ad creatives for 387 different target accounts, each one with the prospect's company logo, their name, a relevant pain point based on their industry, Ready to launch. I used to think personalization at scale, quote unquote, was marketing BS. Turns out it just needed better math. The old way, one designer, 387 accounts, 15 minutes per ad, 97 hours, about 2.5 weeks. The new way, 12 minutes, same quality, sometimes better. I know this sounds like I'm pitching you. I'm not yet. I'm genuinely curious if you've seen this kind of efficiency anywhere else in your ABM stack, because if you have, I want to know about it.

Eric? 6.5. I would probably say this one is also AI-generated and the first one may be human-generated. Because assuming I can only— one of them's human, one of them's AI.

How about you?

I think it was a combo. I think it was human with AI. AI started it and then human tweaked it. Great. Okay.

So yeah, they're both AI-generated. They're both AI.

Yeah.

So Neil caught on. But My point here is I think where I'm going, and I'll share some other examples, but I think the physics of business is changing. You know, like when, if you and I work together and you told me it's going to take 2 weeks to get something like this up, I can at least get this. This took me like 5 minutes, right? And I can iterate off of it. So you said 6 initially, you said 8 on one of them, right? That's enough of a good enough starting point, which is what we've been talking about. But I think the problem is, and Steve Jobs said this, right? He's never found anyone in his career that will magically want to work hard that's never worked hard before. And what's happening with this stuff is it's actually exposing people who aren't motivated a lot faster because again, the physics have changed. The laws, like I can get anything up I want, at least up to like a 6 or 7, right? Which is pretty good. And we actually got new copy up for the ClickFlow site and we got new copy up for the Single Grain site, right? Like this week we changed our popup as well. The popup looks a lot better. You know, we reworked ads, like all these, like there's this carrot ads plan, there's the onboarding email sequence for all these things. My point is not saying, guys, this is like a 10 out of 10 out the gate, but I do think that you're able to do a lot more now, a lot faster with a lot less stress. If you're a motivated person, it doesn't matter. Like when you and I were not making any, when my first job, I was making $32,000 a year, I always wanted to work hard every day. Did I play poker in the back online? Absolutely I did.

I did not expect that answer.

But I worked hard, right? That's my point. That's my point anyway.

By the way, do you want to know why I felt they were AI generated?

Why?

Too wordy.

So that's a big problem with AI.

It doesn't get to the point. I don't know why it rambles. Almost all platforms do it.

You know what's crazy? No, finish your thought.

And what we found, and this is why I gave them a 6 rating, one was 6, the second was 6.5. The first one was too storytelling. I'm not saying storytelling's bad because marketing storytelling's amazing, but in this example, Eric is a B2B company, right? He owns Carrot and he's hitting up someone who works at Salesforce. Correct me if I'm wrong. I'm assuming you hit up someone at Salesforce.

No, that was, I just inserted. Yeah.

Okay.

But hypothetically, Salesforce would be a target customer.

Okay.

Yeah.

That's another B2B company. The people Eric are hitting up, they don't want storytelling. They want stats. data and they want you to get to the point and that's it. The reason the second one, I liked it a little bit more is he used some stats and data, but again, it was too wordy. And the angle I would've taken with their approach is I would've taken how many ads you can do. And I would've talked about how they barely do any ads. And I would've talked about the ROI by just testing in a mass quantity. And I would've said, I created some more ads for you that are personalized, et cetera, based on who we think is your target audience. Let me know if you want to schedule a call. We can actually roll this out and do more for you to grow your revenue or whatever it may be.

All right, so I wanted to take a moment to tell you about my podcast co-host Neil's agency called NP Digital, and they work with a whole host of global companies or a global organization. Also, Neil has SEO tools such as Ubersuggest, Suggest and AnswerThePublic. All you have to do is go to npdigital.com to learn more, and we'll see you on the other side. So let me show you something, Neil. So this, I'm going to give you guys this prompt too. And so this prompt has been incredible for me. If you want a prompt that can basically handle a lot of the marketing copy for you. So this is a, it's like a 14-page document, right? What are the names here? Gary Halbert, Eugene Schwartz, Ogilvy, Kennedy. These are the old school, like OGs of marketing, right? Now what Neil's saying, I don't necessarily disagree with. Actually, his angle is in here. There is an angle in here where we send them the personalized ads with examples, right? And so your job here is not to just take this for what it is, like the offer, free LinkedIn ABM audit. Here's like the ads that we made for you. Here's the assets delivered.

Put this on Twitter for people.

Yeah, well, this, this we will. That's a good one.

You should put it on X.

But like, here's like the email sequences over here, but look, it's a 14-page thing and I did this with one prompt, right? And so it's a starting point because a lot of times when you're thinking about creating like our YouTube long form, A lot of the pain is in the ideation in my mind, right? What I would say here is the prompt is basically, hey, before you write it, I want you to take the wisdom from the 10 best copywriters and email people in the world, right? And then I want you to refine this and score each copy, each version you give me 0 through 100. Only return it to me when it's above 95 to 100. So it's going to do the work and score itself, right? And it's funny, when I made an SEO one, it actually, it put your name in there too. So it's such a good starting point when you can pull from all these, the wisdom of other people. But I think my real point here is the physics of business and marketing has changed. And for the majority of you listening to this, I think you're motivated, but just know that going back to high output management, you're going to have to be willing to be a lot more confrontational a lot faster because this is actually exposing the people who don't want to put the work in. Yes.

A few things. One, I'm going to go on, add to your copywriting thing. Eric sent me a version of my homepage using the template. I saw a little bit there. I've seen what you did with the Single Grain website based on what you put on X. And of course, these emails that you just mentioned here. This is just my personal take. I don't like using AI to write copy. I have never liked it. I've tried it so many times.

Yeah.

And I'm talking specifically for persuasive copywriting that convince people to try to convert, but I'm a little biased. I think I'm probably better than most people when it comes to writing persuasive copy, and I don't have a hard time coming up with ideas. So it's actually quicker for me to just write it out from scratch than it is to use AI and then modify it. And that's why I don't like using AI for persuasive copywriting, but I do like using it for other things.

Yeah.

Like for visuals and stuff like that, I love using it. Like Hey Parker's really cool. I don't know if you've played with that yet.

What's Hey Parker?

It's a tool that helps you create a ton of AI creatives for like Facebook ad campaigns.

Oh, cool.

Yeah, it's worth checking out. And then the second thing is, is the thing that I see with AI right now that's really irritating me, it's causing people to get lazier and lazier. And here's what I mean by this. Yes, AI should be used alongside a human and it should be making them more efficient and better. But I was interviewing someone the other day and this person was from Europe and they wanted to work for us. They, and I gave them an assignment. They ended up just doing the assignment and I could tell it was purely done through AI and they didn't check their work.

Yeah.

And the reason I know that is there were some big errors in there where if they checked it, with their quote unquote experience, they would've caught not one of them, not two of them. They would've caught at least 3 of them. And I just told them like, dude, this is not going to be a fit. There's 3 errors here. You didn't catch any of these major ones. Like, no, I didn't do AI. I spent a lot of time doing this. And I'm like, no, you didn't use AI. But I didn't care to argue with them.

What role was this for?

This role was for someone in data and analytics.

Okay. Like pretty senior.

I don't know if it was senior.

But important.

Important.

Yeah.

So that's why I want to be involved in there because I end up helping with a lot of our A/B tests and I'm very picky on how the data is presented to me. And a lot of people present data that I believe is off because I'll double-check it with things like daily signups versus traffic. And then when you eliminate bots, you can really see the conversion rates and they'll be like, no, look at these numbers. I'm like, okay. Did you block the bot signups for people creating an account?

Yeah.

Let's look at credit cards. Let's look at how many credit cards went through because some people could just be putting in credit card info to try to check if a credit card is real and do scams. So then I look at how much stuff is through and I'm like, so you ran this A/B test, you're saying it increased conversions 30%, something ridiculous like that. I'm like, where are 30% more new paid credit card signups? And they're like, well, that's what the test is showing. I'm like, The data's wrong. I don't see it in our backend of people actually swiping their credit cards. So I don't like data and analytics people who don't actually think things through just because the data shows you. Well, did you dig deeper? Is it off?

Mm-hmm.

Are you going all the way down to bottom of funnel? Are you looking at which cards are sticking around? Are you looking at, did your message convince more people to sign up, but it screwed up the churn during the trial or the cancellations or even churn in month 1 or month 2 because you overpromise? Because if you overpromise, you get more signups, but then people are going to be pissed off at you and it's going to hurt your brand. But the point I'm getting with this is I'm seeing more and more when I'm reviewing people for job interviews and there are other people on my team interview people first and then it comes to me and then others as well. But I'm just getting really irritated that people, at least in marketing type roles, are using a lot of AI for tasks and they don't just check their work. And it's just like, dude, I don't blame you for using AI, but at least double-check your work and make it better.

That's why, so it's interesting to me before I, come back to this topic. You know, the robotics company Figure, so this guy Brett Adcock, he's their founder CEO. So he posted to Twitter earlier this week or X, and he's like, guys, I'm paying $500 grand and I'll give you millions of dollars in stock options, okay? So here's what the job is basically, or the assignment. And it was like a 5-minute assignment and it's like a bunch of popups. Remember like when you're back in the '90s, you just have popups like scrolling through your screen.

Yeah.

And so it's like, I want you to use a general agent or use anything. You can use AI if you want. It needs to solve this in 5 minutes. The problem is most people don't know how to do that. If nowadays, so what's interesting is you don't want people to do it that way, right? But I'm just saying, and I'm not saying there's a right or wrong way here, but his side is like, I'll hire you for $500 grand if you know how to use this general use stuff. And so I did this for fun, not like I'm looking for the job, but I wanted to see, I was like, okay. So I inserted the exact problem set into Manus, which is Meta's now general AI thing, right? It literally solved it in 5 minutes and it sent me all the things like the formatting and all that. And I think that's what Brett, the CEO, is looking for. He's looking for people that know how to think like that. And he knows the majority of people are not going to do that, right? And he's filtering through them quickly. Now, I'm like, okay, if I combine that along with— I have a GitHub, an open GitHub right now that says Beat Claude. And so the whole idea here is for all the roles that we're hiring, they will click into like an assignment and then they need to fill out the written assignment. They can use AI, they can use all they want, but if they only match Claude, it's not good enough, right? And Claude itself will grade them on it. And so I don't even want to talk now until they maybe get through this because it's going to save you and I a lot of time. So yeah.

But the difference here is with the example you gave, and I look at them as 2 different scenarios. The scenario A that I'm talking about is marketers are starting to use AI to just do their work and not even check it.

Oh yeah.

No, I know what you're saying. And they're pushing out junk.

Yeah.

The example B that you gave, I look at that as smart. You got a problem and you're using AI to help you solve the problem.

Yep.

And then you can go and you can see if it's done right. It may or may not work, but you yourself are not a developer and you're just like, wait, I can actually use Mantis to help me solve this.

Manus.

Manus.

Yeah.

And use a specific prompt and let me see if it'll actually solve this problem. And you're trying to figure out what you can input it for to understand the issue and then solve it. You're using your brain to help you solve a problem combined with AI for a skillset you don't naturally have, which is development. The other one, the example I gave is someone who's a marketer doesn't want to do their job using AI and AI does a shit version of the job and they claim they did it all manually when we know they didn't spend more than 5 minutes on it.

Yeah. And so I think the point we're trying to get at here is just like how I believe that money amplifies who you are, I actually think when you have artificial intelligence, it also amplifies who you are.

Dude, I haven't heard anyone say it like that before. I think that's so true because we have a mutual friend. You know who I'm talking about? He worked for you for a bit.

Yeah.

He's using AI so much now and I'm like, this guy doesn't like working. Of course he's using AI because AI is making it so he can get stuff done and not work. And that's an example of someone who wants to collect a check and put in no time. Now, just being very blunt here, I've told this person this to their face, so I'm not trying to say anything that they don't already know.

Many times.

Many times. Yeah, you know that. But I do believe what you're saying, which AI just brings out people's quote unquote natural work ethic. More. So if you were lazy before, AI just is like, oh wow, let me collect a check and be more lazy. Those people in my mind are going to get replaced and fired from organizations.

Yeah. I think, look, I actually think this is so— actually, I'm not going to mention this guy because he's now in those files, but this one guy, tech guy, started one of the biggest companies in the world. He used to say that, you know, you want to work for lazy people, but lazy smart people.

Is this Reid Hoffman?

No, no, no. The other one. Yeah. Uh, Lee Reeds Hoffman was acquired by this company.

So, uh, Bill Gates.

Yeah, there you go. Yeah.

We don't know Bill Gates. Say whoever.

Who cares? I'm acting like he's a homie, but, uh, yeah.

Even like Melinda Gates.

Did you see her response? Oh yeah. Just talk to him about it.

Yeah. She's like, talk to him about my medication and potential STDs and whatever. And I was like, you know what?

I have a topic that's related to SEO and that in a second. But, um, so my, my, my point is I think it's okay to think lazy, but you, if, as long as you're smart about it, you're going to find ways around things, right? So the things that I'm showing you right Now, again, guys, I'm not saying, hey, use that prompt as gospel and use that 6, 7, 8 out of 10. It's not good enough, right? You have to refine it yourself. And so the copy that we actually have on the Single Grain site now and the ClickFlow site was refined by me, and it's going to continue to get better over time. Now, the last thing I'll show you, Neil, and then we can move on from this one, is, you know, this popup. So we used to— we had a popup on Single Grain site. It only pushed a carrot, right? I was like, okay, I'm going to have a popup. And then what is it? I think Cloud Code did this. Okay, so now the pop-up is 2 different adventures, right? Am I going to say that converts best? No, but it's better than just carrot, right? Because we have ClickFlow too. So it says 2 ways to fill your pipeline, pick yours. Trusted by, you know, these companies over here. ClickFlow is our AI content isn't converting, right? So ClickFlow's AI plans and writes production-grade content. So you don't need to get 10, you know, you don't need 10 more writers and editors. Early adopters average 27% more organic traffic in 6 months. Okay, that's one side, left side. Right side of the popup says Carrot, ABM and demand gen teams, right? That's for them. Our LinkedIn ads lack ABM personalization. Carrot generalizes personalized ads and landing pages for every target account in minutes, not weeks. Our team has closed— one team has closed 2 deals from just 15 accounts in under 2 weeks. So now my point is this is refined with a human being and this is at least better than what we had than with sitting there for months.

You want me to tell you a way that'll convert better?

No, hold it up.

No, I really want— you want to copy it? You can copy, you can jack it.

Go ahead.

Okay, so it's instead of 2 ways to fill your pipeline, I would've asked a question like, do you need more traffic? And they can click yes or no.

That's a standard Neil Patel pop-up.

Wait, wait, wait, check this out. If they say yes, then it would be like, okay, check out ClickFlow. It can help you with X, Y, and Z to get you more traffic, right? If they click no, I would be like, congrats on having massive traffic, but I bet you're not getting enough sales. Why? Because your landing pages, et cetera, aren't personalized. And then I'm like, with Karat, We can end up personalizing your ads, your landing pages. You can do it all within X amount of seconds. Fill out this information to learn more so that way you can get more conversions from the traffic you have.

See, that's good. See, we just made an idea baby with AI.

Sure. But I could have gave you that one beginning without AI.

I would've not had— I would've, I would've— it was, it was not this before. It was just carrot. So—

No, but I could have gave you that without AI.

But I wouldn't have, I wouldn't have thought about that to, to go to you about that. But anyway, my point is it's, Like also, okay, if we think about this, right? If you're to rate yourself as a copywriter or a CRO person percentile-wise in the world, what percentile do you think you are?

I don't want to say because it'll sound arrogant.

Yeah, but exactly, that's my point. So let me say it for you, right? I think it's in the top 1 percentile, like within like the top 1% of the 1%. That's what I would say. Okay? Because this is all the stuff we've nerded out for on a very long time frame.

Yeah, we've been doing this for 27 years.

So we are more biased and our standards are higher, but I am trying to make this more accessible for everyone. To just say that, hey, like, look, this is whether you're using Claude Cowork or Claude Code or whatever it is exactly, or Manus, you can use all these things. That's all we're saying.

I do agree. This is great work for average people. No, no, I'm not saying your team or anything, but I'm just saying like if—

I know you're not talking about my team.

I'm saying like if someone's in marketing and they want to use AI for this stuff, especially if you're new in your career, this will help bring you from a newbie beginner To intermediate.

Yep.

Which I think is great.

So can we talk about real quick, because we brought up his name and then we'll move over to the SaaSpocalypse. But so do you know why Jeffrey Epstein paid for SEO?

I didn't even know he paid for SEO, but I'm curious to hear.

Exactly. So, so they have discovery, right? There's all these leaked emails that came out. So literally Jeffrey Epstein was paying 10 grand a month for SEO services to hide his crimes. So what he was doing, guys, was Not necessarily. It's a subset of SEO. It's called reputation management. And he was trying to get the negative things people were saying about him pushed to the second or third page so people couldn't see it, right? So basically, among the emails that came out recently were some between Epstein and his SEO service providers. Basically, he was paying roughly $10,000 a month for search suppression and reputation control. The tactics outlined in those emails are uncomfortable to read, but they also reveal how search systems actually work. So Epstein's reps attempted to control Wikipedia because it sat at the top of the results. Okay, that's easier to manipulate. Create multiple standalone sites and manufacture alternate narratives, and then you build links to those to move them up. Okay. Push positive neutral coverage up with links.

Okay.

Yeah. So if you see examples already, you know, on the first page or maybe top of the second page, you build links to those and move on. That's reputation management. Then you flood the index with adjacent entities and similar names, manipulate images, headlines, and associations, starve negative pages of reinforcement while amplifying others. So Look, did he pay to do this? Yeah. But will this work long-term anymore? Probably not in the next couple years.

So I still think you can do a lot with reputation management. I got hit up in maybe late November, early December. I get hit up probably like 4 or 5 times a year for reputation management from people who have a lot of money. When I say a lot of money, I'm talking about people like on like Forbes list or they're not on the Forbes list, but they should be on the Forbes list if people really knew how much money they had. Remember you hit up that VC and he responded really Really quickly to you? Yeah, he was a billionaire. I think, I'm pretty sure he's still on the Forbes list.

Yeah.

So like one of the ones in, it was either November or December, came to me. It was a dude really, really well off. He would've paid me a solid 7 figures. My cost on it probably would've been $100,000, $200,000. And when I say solid 7 figures, I'm not talking about $1 million, substantially more than that. We didn't negotiate and fully get to the amount or anything like that.

Yeah.

Because I didn't care. But he did something to a woman or a girl. I don't know what their age was.

Yep.

And they settled and he wanted it all gone. And I didn't say anything to him because I don't want to create enemies. And I just say, you know, I first responded to it because they didn't tell me what it was or who it was for. And then eventually they ended up telling me who it was and what it was for and how much they would pay. And it's like, screw you. You did some shit to a girl, deal with it. You lay in the bed, you lay in the bed that you created. You said that? No, that's what I was thinking. So we just ignored him and it was like, screw you. I'm not going to help him. If you want to screw— it's like, if you want reputation management, because you're just like, I'm making this example up, but let's say you're Saudi Arabia. Saudi Arabia is trying to take over golf in the US and buy the PGA Tour. Who the heck cares? Just honestly, who really cares? I don't think 99.9% of America cares. You do something bad like that, you deal with it. That should stay on the internet. That's your problem. I just believe it's like some people have reputation issues for like silly things, you know, like, but others—

They should take responsibility for it.

Yes, they should take responsibility.

You know what's interesting? So I totally agree with that. The final thing I'll say before we move on from this topic is Jeffrey's a— he's a good negotiator here. So he's like, I was never told, never, that there was a $10K fee per month. You initially said the project would take $20K, then another $10K, then another $10K. So he's like negotiating it. He doesn't like the cost tied to it. But it's interesting to me too, because the person that posted this, 94,000 views, ton of bookmarks and things like that. And I think he actually sells SEO. Yeah, he sells SEO. So good for him for getting all those impressions. I hope it drove some business for him. Yeah.

Yeah. And then I saw so many mentions online, dude. The amount of texts I got from people that were in his list, you know, people were sending me all the documents for Reid Hoffman.

Yeah. I was shocked that Peter Attia was in it. The doctor.

Oh my, you sent me the Instagram videos. I was looking at the comments and then he left the protein bar company called David.

He did?

Yeah, he left. Of course.

Who's going to want to work with him?

Another one who's in there is Jason Calacanis, and he's like—

Oh, he's in there?

Hey pal, he turned off comments on X. Even this is after he got committed. He's like, hey pal.

Oh man, you know more than I do.

Like, oh dude, I was reading all this stuff and it was just like, you know what, why would people want to do business with someone who's screwing over other people and girls and stuff like that? Like, that's not cool.

Same with Bill Gates.

I don't care how rich you are, you want to do funky stuff. You know, people should just lock some of these people up.

I, I, they definitely should take responsibility.

Not just responsibility. Some of these people need to be locked up.

That's what I mean.

They need to be really committed.

In my mind, if you've, if you've done this, taking responsibility is getting locked up. Like, not just buying your way out of it.

Totally agree.

Okay, so this is interesting because you and I have SEO products right now and there's, there's, I think it's interesting for us to talk about and then, then we'll move into, you know, the SaaSpocalypse thing. But where do you think SEO products are going in the near future?

Automation. And I'll be very specific, automate fixing your on-page code, automating the creation of your content, automating the promotion of your content, automating generating links and brand mentions, right? Automating posting it on social media and driving brand awareness. And that's what I mean. I think it's going to automate more of the tasks that people do. You still need a human in the loop from what I can tell, Technology may change over the next 5 years, but I see a lot more efficiencies be created by these tools, but the market disagrees with me. And the reason the market disagrees, you've seen it, look how much SaaS companies are getting hit. It's like, oh, my company is HubSpot. I'm giving a hypothetical here, but let's say my company's HubSpot. Our growth rate has increased. Oh, my stock is now down 60% over the last year. Maybe HubSpot was overvalued before, but 60% hit, it's like, come on.

60% hit on 20% growth.

Yeah, so what? HubSpot is getting hit hard, Intuit's getting hit hard, ServiceNow I believe is getting hit hard.

Yeah, actually—

Adobe's getting hit hard. It's—

I went to X for this, and here, let me give you some numbers because we like numbers, right?

Yes.

So check this out. So ServiceNow, According to Grok, down 28% this week, Salesforce down 26%, Intuit down 11% in a single day. And then you see Snowflake down 26%, Adobe negative 35%, Salesforce down 35%. And sorry, this week ServiceNow dropped 28%, but overall it's down 44%. So it's happening across the board. And this is— we'll talk about the SaaSpocalypse, but let's focus on the SEO products first. So I actually think where this is going, because here's the deal, like these content— you have a content creator, I have a content creator as well. There's a bunch of content creators in the space and they all kind of— they're all a little different. But a lot of this stuff is becoming commoditized. And so I do think— I continue to believe that the way to succeed here is to upsell them into services, into managed services. And I do think the way we're going to do service in the future, Neil, you might or might not agree with me right now and it's fine, but I do think it's going to be like the Palantir model. And what I mean by that is Palantir has a base level of scaffolding for their main product, but then they have 4 deployed engineers that customize it for each customer, right? So right now when I, when I'll text some of our clients, they're like, hey, we need— can you make one of these bots that can do this for us? Like, these things are relatively simple to build, not easy. But if we can build the scaffolding for all that for Single Grain, which is what I'm doing right now in my little OpenClaw, like I have superheroes, I have Alfred, I have Flash, I have Oracle, I have Green Arrow, I have all the DC heroes, right? And I'm testing them to see if they'll constantly Be proactive, iterate. And the crazy thing is Oracle, which is my SEO bot, it actually uses ClickFlow. And as it uses ClickFlow, it gives feedback, product feedback on ClickFlow as well. So it's a self-iterating loop, right? And so that's where I think not just SEO products, I think all products are going this direction. Now, do I think SaaS is going to zero? No. I think what's happening right now is layoffs, by the way, was up, I think, 250% or 100% since compared to last month. Crypto is getting wiped out right now, like $67,000 for Bitcoin, right? And then you have this happening. So maybe there's a market pullback happening right now too. But I think what happened with SaaS specifically going into the SaaSpocalypse was everyone's like, oh, Anthropic came out with these skills and like you can just pay $20 a month for whatever. But what we talked about for weeks and weeks here was that for HubSpot, for Salesforce, you know that if you're a multibillion-dollar company running all those transactions through it, you need reliable, You need a reliable CRM, you need a reliable ERP. You can't just swap it out like that. There's a lot that goes into it. So I think a lot of these things are going to come back and maybe my sense is you and I are waiting for the market to pull back a little more.

Yeah, so I can buy more.

Yeah, exactly.

I'm like, oh, how much more will it go back?

Yeah.

That's not financial investing advice, but Eric, the way Eric looked at me is he knows me quite well. But yeah, dude, you've been using AI to code way more than I have. And let me ask you a question. I already know the answer, but just for everyone listening, how often are the products you're creating have bugs and errors and issues?

Oh, all the time.

Okay, so—

And you think you fix it and it comes out broken the next day. Yeah.

So how is anyone going to be like, yeah, I'm going to use AI to replace my HubSpot and it's okay if emails don't go out and my leads get lost and we don't close as much revenue.

Yeah.

And I'd rather do that instead of paying Or no, not even paying. Instead of using the free plan HubSpot has or paying $50 or whatever it is a month, people aren't going to make that decision. They're just going to pay the $50 or $100 a month or use a free plan than they are going to build that in-house because you're going to have too many bugs.

What do we say? We said AI amplifies your nature, which is also your intelligence too. And so when you hear people saying that, it's like, oh, you clearly haven't thought about this.

Yes.

Yeah. So, but here we, we, we, we, so I don't think a SaaSpocalypse, like sure, we're seeing everything drop right now. I think we're going to see a broader pullback. People are like, oh, Bitcoin might drop to $40,000. Or something like that. Maybe, maybe, right? Not financial advice, but you want to see here, let me just show you this before we move on here. I have, you know, Opus, right? The little clipping bot for social.

Opus.

Opus Clip.

Yeah.

Yeah.

So you're talking about the one that helps you find short clips from your long-form videos and so on. Has it improved? Because the last time I used it, it was eh.

I used it maybe a month or two ago. It was kind of the same. And so In my mind, right? Like even the stuff we're talking about today, like even Noah's unbiased, right? It's like we spent 80 minutes talking here. I know we have a bunch of good clips in here for your side and my side. And I know Neil has some longer form clips too, because Neil tends to talk like slower and like calmly and like it's longer, but I just know we have good stuff. So here's the thing. I use my little Flash. Flash is my social bot, right? I was like, hey, just see if you can replace Opus here. And it actually, I inserted an 80-minute podcast with you and me in here, right? And it found some good moments. Here's the headline over here from these moments, Cloud Code Will Replace Your Marketing Team, which we actually did use. 150 times faster than engineers. Okay, 1443 to 1535. Here's a hook over here, high agency people will never be replaced, right? So now am I saying these are amazing? Not necessarily, but I am saying these are better than Opus in my mind, right? And then our team can do something with this. Now the other thing I was testing out, because we had a prospect come out to come out to us, and I'm sure this is happening to you too. They're like, we want, you know, thousands of AI ads per month, right? And I was like, okay, well, typically if you're going to do that, you need to do the research on their ad library, their competitors' ad library, what the strategy is, research the company, understand how their marketing is, right? So my little bot over here basically came out with a bunch of concepts and did all the initial research. And I was just like, without much prompting, it used VO3 to just make like a simple ad here. Now, am I saying this ad's amazing? No, but it's a boxer that's like, punching and then like picks up the powder, right? And my point of showing you this is like, I didn't even prompt this that hard, but if I spent an hour, 2, 3 hours with this, if someone on my team did, right, it would be amazing. And I got the concept, I just wanted to see how it'd look. And so I can do this now on my whim. And when I have Neuralink connected to me, it's game over. I can just imagine things to life. Maybe that's going a little too far, but you get my point.

Yeah. Dude, the possibilities are endless.

Yeah. Yeah. For those that have imagination, which is all of you listening.

Yeah. I think there's certain people that are big believers that, and I'm a little bit biased for this because I have kids and they're in a gifted school right now. I'm a big believer that education is still really important. And I talk to a lot of people who are like, oh, with AI, you don't need to know these math formulas. Your kids don't need to know any of this. And I'm like, eh, if they actually know all this stuff and they have a strong foundation, actually some people we know, one of both our mutual friends in Orange County.

Are they really smart people? Yeah, I know who you're talking about.

Are you talking about book smart or business smart?

Just like street smart.

They're street smart.

Okay.

And I'm a big believer that no, if you give your kids amazing education and then you give them AI, they'll do much better than if you didn't give them any education.

But if they don't believe that, I don't think they're street smart. I know who you're talking about, but that, like, it's obvious. If you don't have a good foundation, how can you build on it?

Yes.

So how can you just say like, that is a stupid comment.

It's a very stupid comment.

Yeah. And so by the way, Jensen Huang, so even the Nvidia CEO, he was like, the market is just wrong about software stocks. So the same thing. I mean, he's probably said it. He said that. Yeah. So he said the notion that AI somehow is going to replace software companies is the most illogical thing in the world and time will prove itself. And he said that on February So just a couple days ago.

Look, I hope these stocks tank a lot more so I can buy in.

Yeah.

And then they go back up to where they were.

Which ones are you going to buy? Not financial advice.

As of today?

Yeah.

I would buy more Intuit, HubSpot, Klaviyo. Those 3 I would buy.

You would buy HubSpot because it's not— it's smaller than Salesforce, so more room to grow.

Uh-huh.

Yeah.

And HubSpot has gotten hit really hard.

Yeah.

I think too hard.

Dude, the funny thing is I saw a tweet the other day. So Figma, right? Imagine you wait 10 years for an IPO and then there's a 180-day lockup period and the stock's down 80 to 83%.

It's less than what Adobe offered the last time I checked. What is it now? Uh, yeah, $10 billion. Didn't Adobe offer $20 billion or something like that to buy?

I forgot. You'd have to cross-check.

Adobe offers billion Figma. I just do Google. It's easier.

Yeah.

$20 billion. Yeah.

So, okay. I think this one's related. That's, that's worth talking about.

Dude, wait, half the price. And do you remember that something Khan? I forgot her name, but she is the head of like the Is it FTC or something like that?

Lena Khan.

Lena Khan.

Oh, she's like this is a good deal. Like you know this is why it was good for me to get involved, right? Basically saying me us stopping that deal was good.

Former FTC Chair Lena Khan hailed Figma's August 2025 IPO as a major victory for antitrust enforcement, highlighting the blocking of Adobe's $20 billion acquisition allowed the firm to grow into $68 billion. Independent company. Boy, was she wrong.

Yeah. It's, it's funny how time proves itself out. Okay.

So, well, it's not even that we all knew that there was lockups. The moment lockups go up, a lot of times things go down. And when Figma first came out with their stock price, most people were like, this is overvalued. Even when Adobe offered $20 billion, most people were like, this is overvalued. This is why Figma said yes to the Adobe deal, cuz they didn't believe they're worth $20 billion.

Yep. Yep. Well, hey, we'll see what happens with them. But okay, because we're talking about Figma, because we're talking about the SaaSpocalypse here, let's talk about good versus bad software in the AI world. So I saw this tweet over here. This is from Boring Business. So he says, look, the real turmoil— go ahead.

Wait, once before we get into that, do you know what Figma's quarterly revenue is?

No, what is it?

$274 million. That was the last numbers that I can see. So roughly a billion a year because they were growing, right?

Wow, 68 on revenue.

Yeah, Yeah, no, companies aren't worth 68 times revenue, especially if they're not profitable.

Yeah.

It's just like, where do people get this math from?

Well, that's what I'm saying. Like, it's never been— even when you think about your kids, right? They just need to— you just need to be— to beat the average, like, it's not that hard, right?

No.

You just need to make sure they're above average.

That's it.

Yes. Yeah. Okay, so Boring Business says the real turmoil in software is so much the loss of existing customers. A massive Fortune 500 that already has their existing data on a Salesforce or Workday is not going to magically rip it out and replace with some vibe-coded internal tool. The real damage will be in the upsell of additional features that adds analytics insights or ease of use on top. So for example, if a CRM software upcharges your subscription to be able to analyze your data and provide customer insights, that feature becomes less and less valuable in a data-driven, AI-driven organization. As long as you can download the data, it's very easy for someone to use AI to draw out those insights themselves. Let me just scroll to the bottom. Here, the key point. The real test of a good versus bad software will come down to whether you are selling a feature or a platform. Okay, we should explain that. SaaS companies that are just selling plugin features with a monthly subscription on top will get easily replaced by AI. The companies that provide end-to-end platform-level solutions will continue to do well. In fact, the good software businesses might even see unit economics and margins improve because they will be cutting down their own costs using AI. So a platform would be like a Salesforce, for example. where there's actually a bunch of people building on it, right? And then a lot of the vibe-coded stuff that you're seeing pop up right now, it's just like kind of one-off features. And so those are the ones that I think there's not much of a moat.

Yeah, I agree with you. I don't think there's much of a moat there.

Yeah. So that, which is like when I think about software now, and maybe you're thinking about it this way too, when I think about whether it's Karat or ClickFlow or whatever, I think everything we build has to log into just, I bought the domain singlebrain.com where you have everything just kind of working together. Oh, that's cool. Yeah.

You're going to change your company name?

Maybe eventually, but I do think everything, I do think the way we're doing marketing for people is just going to be there. You have really smart, high agency people managing these, like these 4 deployed marketers managing these solutions. And I think your headcount's going to be a lot less. I can say that, you don't need to say anything. So my guess.

Yeah. Yeah. You should, yeah, I like single brain.

Yeah.

You should end up rolling it out. And we ended up getting offered or pitched by np.com. And I said—

To buy the domain?

To buy the domain. I said, sure. I said, how much you want? $100 grand, $200 grand?

What'd he say? $5 million?

No, I think they were at $1.5 last. And I was like, eh, I don't care to spend $1.5. But I would buy, I told the broker I would buy npd.com and they're like, oh, okay. Yeah, we're going to talk to him. We know him. And then they talked to him.

But funny enough, I'm an Well, don't give the numbers out in public here.

Follow up. But like, even then, I don't really care to own it that much.

You already have the main one, Neil Patel.

And I have NP Digital. Yeah, I don't really care, but I wouldn't, I wouldn't spend $1.5 million on np.com.

Yeah. If you're in B2B, here's the deal. LinkedIn account-based marketing sucks, but it doesn't have to. With Carrot, it allows you to create personalized ads towards your customers and also Personalized landing pages that align with the ads. So you're going to have higher conversions and influence pipeline reporting to prove that your ABM works. So if you want to learn more, just go to karrot.ai and you can join the waitlist. Again, it's karrot.ai and we'll see you on the other side. You want to know how to make $500,000 per podcast episode that we do?

How?

Okay, so, you know, TPBN, it was supposedly called Tech Bros Podcast Network, but it's actually called something else. It's like Tech Bro something, something programming, right? But they actually revealed the numbers on their podcast, okay? So they do a live every day for 3 hours. And they interview like tech people, and they'll also talk about what's happening. And they basically do 5,000 host-read ads a year, okay? And I think they do about, give or take, about $10 million a year. Now, remember I was talking about the Operators Podcast?

Yeah.

So the Operators Podcast, they charge $500,000 per year for a sponsorship. Okay, so this is— this should be interesting. There's money tied to this.

That's a lot.

So, okay, so to sponsor the Operators Podcast, the minimum buy-in is $500 grand a year for a show that reaches less than 100,000 people per episode. It's because just like TBP or Colin and Samir, the audience is incredibly valuable, niche, and hard to reach. Ridge has spent millions on podcast ads. The largest CPM we have ever paid is $40 because we don't need niche, we need reach. But if Fulfill started sponsoring Joe Rogan, the ads wouldn't work. Joe has 10 to 100x as a big audience, but he doesn't have people looking for a new e-com forward ERP. So you either need a lot of reach to charge $500,000 or you need a very valuable audience. And so when you think about Tech Bros Podcast Network, they might only have, you know, 1,000 people or 5,000 people at any given time, but a normal consumer, maybe their average, you know, dispensable spend a year is maybe 3 to 5 grand a year. If that, Okay, now when you look at, if we have VPs of marketing listening to this or founders listening to this, what is their, maybe they're spending $100 grand, $500 grand, $1 million a year, right? So the reality is, I mean, we do have a deal with HubSpot right now, okay? But the reality is when we're working with people, yeah, chances are we probably can charge, you know, and we have, we have charged at least $360 grand a year, remember? Yeah, we have. So I just think that—

We've charged more than $360 grand. What are you talking about? We charged $720 grand.

Oh, that's right. Oh, that's right. That's right. That's right. That's right. So it was $720 grand.

I forgot.

Yeah, thank you. I only thought about my share. I was being selfish. So we charge $720 grand a year and I think we need someone who's specifically on, and by the way, if you're focused on B2B podcasting, we are happy to hand it to you. We can probably offer 2 or 3 more slots and if you have those relationships, we'll happily offer it to you to find them for us. Yeah, because I mean, look, I, I, I, someone started messaging me. Uh, you didn't send me an email.

No, I know, because they started messaging me on pricing and I told them more.

So once they agree, I'll forward it to you. Oh, good, good. Yeah, but I think we should, like, you and I, like, especially when we think about influencer deals, by default now, um, I have an autoresponder when anybody reaches out to the Leveling Up sponsorship, uh, email. It just says, like, if they want to post, it's like, $25 grand, $25 grand, it just says $25 grand, right? And then at least $25 grand, right? And then most times I still don't want to do it, right?

Yeah.

But my point is with this podcast, there's a lot of people that offer to do it. I think it's just got to be like at least $250,000, $300,000, $500,000. Because operators, they do $5 million a year and they have a podcast network and they have 10 sponsors doing the $500 grand a year.

Yeah. This is a lot.

We just haven't had focus on it.

That's They don't, so they don't charge based on views or anything like that?

No, it's just like, we know our audience is really valuable and we know we can charge that. And just like Tech Bros Podcast Network, who's listening to that? It's tech bros or not tech bros, but tech founders. And we know this is a valuable audience.

It's a very valuable audience, expensive purchase. I look at it as like a mini Super Bowl ad.

Yeah, sure. Yeah. Mini Super Bowl ad, guys. It's just a pain in the butt to manage and we just, we have, we don't have someone dedicated on neither of us too. So.

No. Yeah. I look at a lot of this stuff as you just figure out what works for you and then you go from there. But like, I think like the ROSE CEO or founder who said how a Super Bowl ad is really a great value and was comparing it to companies spending a minimum of $200 million a year on ads. I was like, I don't know what reality you live in.

I listened to a podcast with the CMO of Wix. So he's been there for like 18, 19 years or so. Okay. So I think Wix, their marketing budget is $100 million a year. And he was saying, He was just saying that the Super Bowl is one of their best spends of the year. And he's like, you know, I've been marketing for a very long time. It's a different perspective, right? I'm not saying anybody's right or wrong here, but he's like on $100 million a year. He's like, that's one of our best things because everyone just remembers us, blah, blah, blah. So your mileage may vary. And Neil's just sharing data, client-side data that he has on his side. And then the ROSE CEO saying that it's his first time, but then you have the Wix side. So it's always interesting to me to hear different perspectives. Yeah.

Yes, I agree with you. This is always different, interesting to hear different perspectives.

Yeah.

But for majority of companies, because I see a lot of data related to Super Bowl and I see companies of all sizes running Super Bowl ads, majority of them are medium to big. Medium being at least hundreds and hundreds and hundreds of millions in revenue, big being in the billions. And we typically don't see companies generating ROI from Super Bowl ads. And this is whether Whether it's us or Publicis or WPP or Dentsu or Omnicom, you pick the agency, most people don't generate an ROI because you got to buy the Super Bowl ad, you got to buy like Olympic time slots, which tomorrow's opening ceremony, I can't wait for that. You got to spend millions on the creative and then the talent, and then you got to do campaigns around it and rally your people, and maybe you want to do stuff on YouTube at the same time.

Yeah, $20, $30 million.

It all adds up.

Let me ask you a question real quick. Real quick side note, because I was kind of talking about this with my team this week. So you just mentioned a couple of these holding companies, WPP, Omnicom, Publicis, whatever, and some of them are merged together. I lose track of this, right? Yeah.

Omnicom IPG is going through the merger.

So when you think about these big holding companies, agency holding companies, what do you think their differentiator is with each of them?

Publicis is more integrated approach. They've been doubling down more harder on influencer marketing than most of the other players. And they're probably the cream of the crop when it looks to quote unquote reputation. But from a tech perspective, my personal opinion and influencer perspective, I still think a lot of the small and medium shops do a better job. That's just my own perspective. Everyone can have their own opinion. And I'm also biased because I own an ad agency. So just putting that out there as well, but I don't think there's a differentiator outside of publicists. Publicists, when they pitch How it's all integrated, that story works well. The other big difference between Publicis and a lot of the other big holding companies, Publicis is very picky on what they buy and it's, they seem to be much more strategic. A lot of the other ones are just spray and pray.

Yeah. So that's exactly my point. So I was arguing with my team earlier this week and there, granted, there are some agencies that have differentiators, right? My whole thing was like, there isn't a lot of differentiation with a lot of these agencies. And you can actually, because the thing is like you basically have process power with people, And that was the old way of doing it. And so again, earlier in this episode, I talked about how the physics, the laws of physics as it ties in with business is changing rapidly. And I think it will force differentiators into businesses. But I do think like a lot of whether it's small business, small agencies or large agencies, I think a lot of them, it's just like, you know, they're helping people and like it's, they're easily, you know, interchangeable, right? For the most part. Now, one final thing, Neil, I want to wrap up a bow on the luxury creator economy piece. Okay, so TBPN. Okay, so TBPN, Check this out. So you like, here's a little P&L for you.

Okay.

So congrats, TBPN, $10 million plus in revenue. They started it last year as a podcast. Okay. $300 CPM. $300 CPM. Okay. 20 ads a day. True respect. So when you look at this one, 250 ads, 250 episodes, 20 ads per episode, $5,000 that you can make per episode, right? And sponsorship packages, 20 average sponsor deals, $500 grand, $10 million in revenue.

Wow.

Average sponsor deal over here, and they just do the math. Cost per ad, $2,000. Views per post, 6,000. CPM is $333. Main thing again is they know they don't have heavy, heavy volume. They might have like 2,000 to 5,000 listeners at a time, but they know it's really valuable.

So yes, because one of their listeners could listen to an ad and then sign a multimillion-dollar contract.

Yeah. With Salesforce or something like that.

Yeah.

Right.

I get why people spend it. If you're a big company too, you just have to end up spending money on brand awareness. And that's why a lot of companies do super lots because they just have so You've seen this, right?

As you scroll by. So the cool thing is like, and I know one of the co-founders at Ramp, I should reach out. Ramp is a presenting sponsor, so they put the sticker up there and then they have these like Formula 1 hats where it's like marketing school and you have all the logos that sponsor you. I'm like, that's pretty cool. Imagine we have like an orange one, like McLaren. That'd be pretty cool. See, I don't have time to manage this, nor do you, but it'd be nice if someone would manage it.

Yes.

Yeah. Anyway, you can pick another one.

All right, let's go with, Let's see. Goldman Sachs, AI agents will overtake SaaS quickly.

Ooh, did you see this graph? Here, let me show it to you.

Uh-uh.

Okay, so this, I keep calling it a graph, it's a chart. No, it's a graph.

Same thing.

Yeah, so okay, the profit pool in software is expected to shift towards AI agents. So this is a study from Goldman Sachs. So this shows the illustrative shift in total addressable market, TAM, for software as a service, SaaS and agents over time. So you see in 2025 over here, then it goes to 2026. Okay, 2025, you have SaaS TAM and agent TAM is relatively the same at $30 billion. Okay, then you go to 2026, you know, kind of the same. And then you go to 2027, and then agents start to move higher while SaaS starts to go downwards, right? You go all the way to 2030, where the agent TAM is $50 billion, SaaS TAM is down to $20 billion.

This is just so confusing because Because most of the SaaS companies I know—

I think it just becomes this.

Yeah, have agents and are creating agents for their customers. I'm like, this is Goldman Sachs, they're just creating a bunch of bullshit.

Yeah, so that's exactly what I was going to say, right? Like, wait, this doesn't just disappear, it kind of just becomes this.

Yeah. HubSpot's been big on agents. Didn't Dharmesh buy that agent web domain as well?

He bought a lot of— you know, he started an OpenClaw site this week too. Remember how I've been talking about OpenClaw and ClaudeBot? Like, he started that too. So What we've been saying earlier is that a lot of this stuff is going to shift towards outcomes. Like even SEO software, this is going to shift towards outcomes because people just, people ultimately don't want to do things. They just want things done for them.

Yes.

Yeah, that's how it is. And so let me just see. Oh, this OpenAI Frontier thing I got to check out. Did you see that today?

No, but I've been watching crypto prices all day.

You have? Oh, you're waiting. I still think there's more room to drop.

Oh, me too. Yeah. Not financial advice.

Yeah, not financial advice. But you called this a couple months ago. Okay, so let me just, this Goldman Sachs thing, We can move on. A recent Goldman Sachs research project projects AI agents are set to take over the profit pool in software. Agents are expected to account for 60% of software economics by 2030. I think whoever wrote this should be fired. That's what I think.

Yes.

Yeah.

Oh, dude, you want to hear something interesting? Because we were talking about crypto prices for a little bit.

Yeah.

So I texted someone earlier this morning because they had shares in a crypto company. And I said, happy you liquidated some. I'm like, you can't control prices. And why live in stress each day? They're in a place where their shares are worth so much, they could just buy a home free and clear, and that would be life-changing for 'em. And I told 'em—

Do I know this person?

Yeah, you know 'em. I'm like—

Tall person.

Yeah.

Yeah.

I'm like, if it goes to zero, you're screwed. If it triples, quadruples, yes, your life gets even better, but going to zero really screws you over.

You can't afford to go to zero. Yeah, you can't afford it.

I'm like, you should at least start liquidating. Eventually they started Yeah. And I told them like, no one can predict when it'll go up or down or any of it. It's all random. So then they text me back while we're recording this podcast. Yeah, it looks like it definitely has hit the bottom though. Have been holding above X price. They're talking about their stock, which is just above the value of their book or whatever it may be that they're claiming. Based on whatever's public information, right? You can Google this company and you can actually see all their holdings because they have to report it. And this person isn't exact where they know all the financials for quarterly earnings or anything like that. And then he straight up said over text, we've been one of the strongest stocks in the last few days. People are deleveraging, it appears, across the market. And I responded with, no one knows the bottom, sadly. It's all a big guess. Because the moment they say they think their stock has hit the bottom and then they're crossing their fingers, no one knows any of that.

Yep.

And then they respond with, sure, could definitely go lower and trade for free essentially, but also why I took some off the table. What do you think is going on? And no one really knows what's going on. Whatever happens, happens. You can never predict the bottom. All you can do is invest in companies you believe in. And I'm not saying their company is good or bad, but it goes back to the Warren Buffett philosophy. I wouldn't want to hold their stock for 10 years or 20 years, whether it looks like a buying opportunity or not. I'm not a buyer because I wouldn't hold it.

Mm-hmm.

Right? I'm not recommending people buy Bitcoin or cryptocurrencies, but certain things like Bitcoin, I'm a believer that I would still want to hold it 5 years from now. I don't know if the price is going to be worse 5 years from now or higher, But when you believe in something, it makes the pill much easier to swallow when things go wrong. And if you're not really, if you don't believe in the company where you're willing to hold it during its downtimes and just keep pushing through, then don't buy it in the first place.

What did Warren Buffett say again? If you're not going to hold it for 10 years, don't even hold it for 10 seconds, right?

Yes.

It's the same thing here. And so that's, I would say the psyche of maybe those of you that are listening to this are a little younger. You might think trading is the way. You see all these people on YouTube, oh, you look at this red candle over here and this, you know, I think the Downside is this and that. And then they draw like another line and goes like this.

Yes.

I'm just like, what is this? Oh dude.

And that same person used to be like, oh, after every Bitcoin halving it goes up like this. And this is what the stock's gonna do.

It's a 4-year cycle. Yeah.

And I'm like, dude, people are just talking out of their behind.

Yeah. They, it's like they're trying to just justify something from outta nowhere. So.

Yes. Just because something happened in the past doesn't mean it repeats itself again in the future.

What do they say? They say history rhymes, but it doesn't repeat. It rhymes, but it doesn't repeat. So keep that in mind.

I haven't heard that one before.

I like it. A lot of good quotes today. See, we need some good clips. Anyway, so by the way, I'll speak for Neil too. We're hiring good clippers. So if you're really good, by the way, don't send me a DM with like how you're good at it. Show me examples on the work that you did for us.

And yeah, he nailed it when he said for us. Don't take other random people and show us the clipping.

Show our stuff.

You don't have to show us a lot, just enough where we're like, sounds good, you're hired.

Some guy, before I show you this OpenAI Frontiers thing, this is some guy had like a lot of followers and I looked at his profile and he like, he has a lot of good clips. I was like, you know, if you really want to stand out, I would recommend that, you know, doing what he's doing. Well, you have a lot of good examples, whatever. I'm like, but yeah, you just look like everybody else. So good luck to you in your business. But let's go back to clipping for a minute.

Didn't you hire a company that was doing all these clips for you and experiments?

I'm still doing it.

Okay.

But what I've realized is, so they're good at it. I've realized, so people PB Launch, shout out to PB Launch. I've realized that when I'm like this, I enjoy making the videos and they tend to do better, right? The problem is when someone writes the scripts for me, I tend to just let it, I just follow it and I look at the performance and it's just not that good, right? Don't get me wrong, it's creative. It's like there's clone videos of me and all this cool stuff, but I like talking about the stuff we talk about.

So what I've found to be super effective And you can do this with your team. We go and interview the people who deal with specific things for clients. So someone may only deal with Reddit for clients. We'll have people who only deal with Instagram for clients or Meta ads for clients or LinkedIn ads, and you get the point, email marketing. And we will talk to them about not just what's new and fresh, but more importantly, because I already know most of that stuff, it's more so—

Yeah.

What are customers continually hitting you up for and what issues do they have and what problems are they trying to address? And we look for enough people saying those, giving them those signals. Then we create a script around those problems and then the solutions we provide, and then we integrate case studies and then we release those videos. You don't get as many views, but you get more customers. Works really well. I find it very efficient. You're writing notes to do this?

I'm going to do this with my little— because it's very doable. So you know what I've done recently, right? So because a lot of our calls are saved, I want to see over time kind of how people, how teams are adopting AI and what the AI score is relatively, or it can even score you on like, hey, Neil's like trending over time on core values. And it cites actually examples of quotes that you say. And so I tested this recently and it actually said like, oh, like Sean, CTO, for example, right? Like, you know, his AI fluency is like, you know, 8.8 out of 10. And it shows some examples of exactly what he said. I mean, it also shows some other people on the team where it's like, hey, maybe this person's like a little behind. Now I can literally take that framework and use it for what you just mentioned over here for content ideation. And it could just pop up like deal of the day, deal of the day.

Yes. But we don't like just using recorded calls. I understand that, especially if you get approval for calls. The reason we like to talk to the people on our team is a recorded call will tell you what people are asking, but it doesn't tell you what people are willing to pay for it. So what we find is when you talk to people on your team, there's certain issues that maybe not the majority say, but enough people say, and someone's willing to pay you $500,000 or $1 million to solve that problem.

Yeah.

While they could get a Reddit problem that a lot of people are asking you for, but people are only willing to spend 10 grand for.

Wait, I'm confused.

Why would your calls No, no, I'm saying let's say your calls are recorded, assuming you're getting permission. If you have recorded calls, potential sales calls, what they're willing to pay for their problems varies a lot.

Yeah, yeah.

Versus when you talk to someone, let's say if we're just focusing on Reddit, if you have 100 recorded calls of potential customers talking about their Reddit problems, you can create a video on that. AI can help you with that. But what we found to be more effective is because with the recorded calls, like if you use Gong and stuff like that, it's really easy to parse and get the data from. But what we found to be better is when we're talking to our sales reps and our team that works on, let's say, Reddit, for example, because we're using Reddit for this case, and they break down what our best customers, the ones who are paying us the most and sticking around the most, are having issues with Reddit. Reddit and how we solved it, and we look at all the commonalities and then create content around that versus what just the majority of the recorded calls say, we tend to pick up better customers because it's appealing to our ideal customer profile versus anyone who has a Reddit marketing issue.

Wait, but I'm still missing it. It's like, so how do you have that conversation if it's not recorded? Because we have Gong in all of our calls and it declares that it's recording.

Yes, but I'm talking about we will go and talk to, let's say Noah here, let's say he runs Reddit marketing for our corporation. We'll ask him the biggest issues for client-related calls. We don't record all our calls. So let's say he has a company, there's a company, Single Brain, and he's talking to you and you already signed up. The conversations are going to be different than what they are for a sales call.

Oh yeah, I think we're on the same page. So I think you might have the gone calls, but I think you're still talking to people too on the side, but you're trying to get whatever data you can. So I think this is kind of similar to what we were talking about earlier in the conversation is this is not the end-all be-all, you know what I mean? I mean, it's— you still need to have conversations.

Yes, but we're finding if we use data from recorded calls, the content quality isn't that good.

Yeah.

And the reason is, is we have too much volume and what people are looking for, it varies drastically. While if we just pull out what our ideal customers are looking for and base it off of that, it's much better content. Not necessarily from the views, but from a conversion from a watcher to becoming a customer.

But you can also segment those calls You can, yes.

Yeah, that's—

so my point is, I think a lot of this is— this does, again, amplify— AI amplifies your nature, right? So hopefully you're someone that thinks a little step, a couple steps, like, hey, like, don't just take the data for what it is, segment it, but also use that with actual conversations that you're having in person too. So the final thing to call out here is this OpenAI Frontier thing came out today.

Wait, I just have one business question for you.

Yeah.

So you record all calls, your clients will let you record calls after they become a customer.

Yeah.

Oh, we can't do that.

Yeah, you can't, but like ours, our thing keeps declaring there's a call recorder. I think some do ask us to remove it and we do, but—

Yeah, no, no, almost all our clients don't want it.

The vast majority of our call recorders are in there. Yeah.

Yeah, I would say if I have, if we have 10 calls with customers, uh, I'm lucky if one lets us record it.

Yeah, you probably have some type of, uh, it's probably these enterprise ones. Yeah. Yeah, so, okay, so speaking of enterprise-ish company, OpenAI today, so they just introduced today, this is breaking news, they introduced Frontier, which is a new platform that helps enterprises build, deploy, and manage AI agents that can do real work. Frontier gives agents the same skills people need to succeed at work, shared context, onboarding, hands-on learning with feedback, and clear permissions and boundaries. That's how teams move beyond isolated use cases to AI coworkers that work across the business. So— Clay actually just came out. So they're part of this, and I want to share this. Varun. So Clay is one of 6 companies to join OpenAI's New Frontier program. Tomorrow's best companies will run on AI across every function. Clay is already a creative tool powering growth teams at Cursor, Ramp, Figma, and OpenAI itself. Now we're making that same power available for enterprises through OpenAI's new platform. So here's some ideas of the benefits that we can talk about. You know, our reactions to this. So here are some ways Clay and OpenAI help GTM teams use AI to grow together. Number 1, RevOps and GTM engineering teams can turn their CRM into a living entity, one that automatically refreshes, dedupes, and surfaces accounts that are ready to buy. Okay, that's cool. Number 2, sales teams can score companies showing buying signals, pull verified contacts, generate personalized messaging, and sync everything to their CRM. Marketing teams can build targeted audiences and ship relevant creative campaigns from ads to landing pages to email. I think the main thing here is that you have all, imagine your sales team, your marketing team, they're all working within their own context. If you have shared context, you're going to be able to move a lot faster.

Yeah.

Yeah. So Clay's doing this, Clay's under Sierra, Harvey, which is the lawyer one.

Sierra is Brett Taylor, right?

Yeah. Brett Taylor, which she, I guess—

He's on the board of OpenAI.

They make bots, I guess. But I think the Clay one is actually—

I think a lot of Sierra stuff is manual too, right? Is what they're saying.

I think they use the Palantir model, which is is like they have the scaffolding and then they customize the chatbots for every client, which again, I think that's the future of software. So anyway, going back to this piece, if you're able, because here's my thing, right? I'm like, okay, I have so many ideas when it comes to sales, GTM, marketing, and you do too, but how do you make sure this stuff is happening? And I remember Dharmesh posted about vectors a long time ago where you want everyone kind of pointing in the same direction, because if I want to do marketing this way, you want to do marketing this way, we're actually neutral. We're not moving in the right direction. You want everyone moving together and you can actually see now you can onboard people the same way and all that. This just came out today. I'm excited about it and maybe I'll go try it when I go home.

On a side note, did you see the ads that—

Oh yeah. Did you see the one with the workout one?

Yeah.

The workout one was funny.

I don't know why I'm blanking on the company.

We were just talking about it. Anthropic versus— Anthropic made these ads, Super Bowl ads. Basically they weren't calling out OpenAI, but they were saying, hey, we'll never have ads in our products.

Yes. Yeah. It was hilarious.

Yeah. Like it was funny. So we'll just describe it real quick I highly recommend you watch. I think they have 4 ads that are coming out, but one guy's like, you know, I'm trying to— I want to get a 6-pack. And then there's this one ripped guy. He's like the AI, right? He's like, perfect. That sounds like a very reasonable plan. Here's how you do it. And the guy's like, amazing, right? And then the other guy's like, okay, well, in order to get started, sign up for this brand over here and we'll give you this discount. And the other guy's like, what? Anthropic will never have ads. Claude will never have ads.

Yeah.

And then Sam Altman responded.

Yo, he was pissed.

Yeah. So the funny thing about that response is someone who is a very popular VC Twitter, like he likes both Claude and OpenAI. He's like, dude, the Sam Altman response was, and I can say this, like he seemed like his butt hurt. He's like, you know, just to let you know that the ad was in good jest. I laughed at it. But just so you know, Texas has more ChatGPT users than all of Claude, right? There's like no reason to say that. And it's not like Anthropic was making— Anthropic just wants their users. It wasn't calling out OpenAI. So I thought he was a little butthurt about it.

Yeah, that was funny. Yeah, I watched him. I enjoyed him a lot. That's pretty much it.

I think we're done. So that's it for today, guys. Please don't forget to rate, review, and subscribe, and we will see you tomorrow.