Top 1.4% of Ad Accounts Are Responsible for 36% of All Meta Ads

There's 150,000 e-commerce stores, Neil. If you're to use the Pareto principle here, if you're to look at maybe the top 1% of ad accounts, how many ads do you think these top 1% of ad accounts are responsible for? Like 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%? Top 1%?

Yeah. I would go with somewhere around 50%, 60%. Okay, close. I would go the 50, like it's actually 36%. So Meta actually reported on this. So this comes from Brian Bubgarner shared this.

So the top 1.4% of ad accounts are responsible for 36% of all ad creative on Meta. So they use clay to scrape 150k e-commerce sites to answer one question. How many businesses are running ads right now? 47,000 of the 150,000 stores have one active ad.

And here's where it gets crazy, okay? So 9,000 accounts are running 100 plus active ads, so 19.3% of advertisers. 1,600 ad accounts are running 500 plus ads, so 3.5%. And then 679 accounts are running 1,000 plus active ads, that's 1.4%.

The median advertiser is running 23 different ads at any given time. The median North Beam customer, I think they're like an e-commerce company, platform for machine learning powered advertising attribution. Okay, great. Is running 150.

Oh, so this is not a meta study, but meta said it. Creative diversification is key, not just hooks, but completely new ad creative. So your marketing team should look like a content creation house. So there's this graph over here done by North Beam and you can see what's happening here.

So what I want to call out before we react to this is it's the whole idea of 80-20, right? You have this kind of top 20% or so. They are driving the 80% of the results ultimately because I think this is – if you add the results ultimately here, yeah, I think –

1.4%? No. Actually, no. They don't actually have that number here. But my point is 1.4% for 36% is a lot. The 1% in many industries account for a lot.

Taxes, ads. Yeah, that's true. That's true. By the way, did you watch the CNBC interview with Jeff Bezos? No. What happened? Yeah, yeah, yeah. He's like, bottom 50% shouldn't pay taxes, right?

Which I think is good. Remember, I've been saying for years, if you make under $100,000, you shouldn't pay taxes? Because a lot of the business owners, those people would spend more on rent, food, etc. And it would cause marketing to boom, right?

People would spend more money. It would actually just cause GDP and business to boom. That's at least my two cents. Yeah. Well, why don't you talk about some of your takeaways from that one?

Because I didn't watch the whole thing. I only watched that snippet, that clip. A lot of it was political. So let's avoid most of it. All right. But the big thing that he said that I think should be eye-opening for a lot of people, I think he used the example of a radiologist.

You know, a lot of people were being like, oh, AI is going to replace my job. He's like, I disagree. He's like, it's going to make you better. That radiologist is going to be so much better with this technology.

So his whole take is people are still needed in organizations. It's just more so they're going to have like a companion or a buddy or AI, whatever you want to end up calling it. That's just going to make them a lot better. I just – I don't know about you, Neil, but I just have more fun at work now because I can do more.

That like – It just makes me happier. And I think maybe naturally if you're an entrepreneur, you have an entrepreneurial mindset, you just want to create more, you want to do more. And I actually do think the majority of people do want to create something.

They do want to be useful. And this just enables it more. Yeah. But going back to the 80-20 thing, yeah, I think we've reacted to that one enough. So I want to come over to – you have one over here.

Wait, real quick. One interesting thing that I saw Bezos doing, you know, This is what all marketers end up doing. So everyone knows SpaceX is gonna IPO soon, right?

They've been talking about it. Numbers have been floated if they're gonna raise $50 billion and how much Goldman Sachs is gonna end up making from it. It's causing, you know, if one thing's doing well in a category,

extremely well, let's say SpaceX, it can create a whole wave, a whole trend. And in marketing, they call it, hey, when something's doing really well or catching on fire, ride the trend. So Bezos talks about how for the first time he's thinking about taking external investors. Because if SpaceX goes out and it gets a 1.5 or $2 trillion valuation or 1.7, whatever they're aiming for, he's riding the trend.

He's like, hey, let me just go, you know, get money while I can based on this trend being hot. I think that's smart. And also he's very credible. And also the space, like the real player right now is SpaceX.

And so if he can be a credible second player, I think there's more than enough room here. And I think it's very smart. He needs a lot of money to make this happen. It's very capital intensive.

It's very capital intensive. And in pretty much every industry, you need a number one and you need a number two, even if the number two is not that big. And he's primed to be the number two. Yeah, I think it's good to have competitors, right?

People like to say like, oh, you shouldn't... I think it's good to have healthy competition between like Anthropic and OpenAI, for example. I think it's good to have a Google versus Microsoft. I think these things are all good.

Yeah, the other thing that he did mention in his interview was... He will do more good from his for-profit companies than he will when he gives away the majority of his net worth. And he does give away a lot of his net worth. And he was talking about it like he provides a lot of shelters for homeless people.

He has a lot of projects that he's funding. But his impact on the world will be much greater from his for-profit entities than his non-profit. And I can see a similar side. Like you and I have created content and education for so many years.

You and I have also both donated money. We don't have anywhere near the money that some of these guys were talking about, but we've donated money where we can.

And no matter how much money we've donated, I believe will impact and help people do better financially for their families through the education we've created more than any of the money that we've given away. Yep, I agree with that. And the thing is, it's unfortunate that business gets painted in a negative light. Like the more success you have, the worse it is now today.

It shouldn't be like that because business ultimately is you trying to create a product or service that helps other people. And if people vote with their money, and if you do well, it means that you've created something very valuable. I don't know how it gets misconstrued into something else, but I feel like that's always been history. That's always been human nature.

It always has. Yeah. One thing I want to talk about, Neil, so I don't know if you saw, so Opus 4.8 came out yesterday, and there's one part of it that changes business completely. So I want to share this with everyone because I want you to, you know, this weekend's coming up, and hopefully you have time to play with it.

So you can see yesterday Anthropics, like, include... introducing Claude Opus 48, which I was like, okay, is it going to be like, you know, not that great, but I'm in all these AI chat, like CEO chat groups right now. And everyone's like, Oh, my teams are using this. It's crazy.

I'm like crazy in a good way. And they're like, yeah, crazy in a good way. Right. I'm like, okay, cool. So, um, You can see, obviously, when they market themselves on these benchmarks, this is one way of marketing your LLM, right?

You just show that you're beating everyone else. So whenever something new comes out, obviously you're going to be number one, right? So you can see Opus 4A is in the red. But what I want to call out here is that they launched this new thing called dynamic workflows, okay?

So what this means, Neil, is if... you want to do like a big migration on your website, you can just ask it to do it, okay? Now, I'm like, okay, well, Neil, you're not going to do a migration yourself. I'm not going to do it myself.

I'm like, how do I apply this to business, to making more money, right? And so I worked with my Hermes on this. I'm going to share a couple of prompts with you, but There's even a video down here, embrace long running tasks with Opus 4A and Cloud Code.

Basically, it will create all these workflows for you and you can have hundreds, if not thousands of these sub-agents working. And the main benefit here is that oftentimes when you have an agent work, they might not follow the process in terms of steps. They might go step four first, then step five, and then step 10, and then step one, which screws up everything. But dynamic workflows allow you to spin up a bunch of sub agents and they follow the process.

That's what it is. So any questions before I show you some of the flows, Neil? No. Okay. So check this out. I'm going to pull this up on my screen, on my Slack real quick.

Give me a second here. So here we go. Let me share my screen. Share this over here. Do you see this? I do. I can't read it, but I can see it. That's okay.

I will read it. Those of you watching right now. So I asked my Hermes, my Hermes over here. So you can see Hermes Paris. So highest leverage cloud code dynamic workflows.

I'm like, tell me what the highest leverage things we can do. Okay. Here's a cool one. I'm just going to read this one out. It's called revenue command center.

And you guys can screenshot what I have on my screen over here. So can you zoom in for people? So that way they can actually screenshot it. Or maybe it's just tiny on my end.

And it's actually probably bigger on. There you go. Oh, that's helpful. Thank you. How about this? How about this? Even better. Okay. This one's called Revenue Command Center.

Okay. So use a workflow, build a revenue command center for NP Digital. Inspect every available data source I provide. CRM exports, call transcripts, email threads, analytics proposals, client reports, SEO data, ad data, internal docs.

Create a unified map of where revenue is being created, delayed, leaked, or ignored. Then run specialist subagents in parallel for pipeline, retention, upsells, SEO, paid acquisition, outbound partnerships, operational bottlenecks. Return a list, a ranked list of the top 25 revenue opportunities with estimated upside, confidence, owner, next action, and required assets, and a seven-day execution plan. All right, so you got to check out this podcast.

It's called Create Like the Greats, hosted by Ross Simmons. It's brought to you by the HubSpot Podcast Network, the audio destination for business professionals. Each episode hosts an in-depth analysis of some of the greatest creations and creators of all time, along with deep dive conversations on the creative process that went into building companies, brands, stories, and more. He's got a recent episode where he is talking about the CMO playbook, so you've got to check it out.

So check it out again. Listen to Create Like the Great wherever you get your podcasts. then implement any safe automations, dashboard scripts, or draft assets needed to start capturing the top opportunities immediately. In fact, I want to hang up for this podcast right now and just run this on my cloud code and not talk to Neil.

And then this one over here, stalled revenue recovery, use a workflow, analyze all deals. My point of saying all this is client expansion engine, churn prevention and safe plan, So I'm going to see how these turn out. I would encourage you all to try something like this.

And I would encourage you to talk to whatever agent you're working with to figure out how you can make use of these dynamic workflows to make you more money because it's not just a programming thing for engineers. Yeah.

Dude, the technology is here. The tools are here. People have access to them. They're just not using them. Yep. That's real reality is what you said earlier.

Yeah. And I don't think that's going to change because, you know, if you look at traditional SaaS, it's been around for how long? I don't know, more than 20 years, right? Way more than 20 years.

Because SaaS was around even before the cloud. And when you look at SaaS, people still don't use all the features they pay for. And you and I are also guilty of it. There's no way that we use every single feature for all the products we pay for, right?

We do a better job because we're really picky and we're price sensitive, you know, being entrepreneurs and startups. But I think the bigger organization, you know, the bigger organization gets, the worse they get about using what they're paying for. Quick break. Look, I know what you're thinking.

Another AI content tool. Great. More garbage content on the internet. And I thought the same thing. That's why we spent years building ClickFlow differently.

Here's actual feedback from a user. It's sophisticated, grounded in real language, authoritative, but not academic. You've hit the sweet spot. That's not AI slop.

That's content that you would actually publish. ClickFlow also helps with things such as internal linking, building FAQs, and reporting on the content performance. If you're skeptical, you can just go to clickflow.com and try it for free for 14 days. And if it sucks, just cancel.

But I don't think you will. Back to the show. Yeah. Maybe a place where, because we're really talking about inertia, human inertia. And I've given this example already, but my dad said when the internet came out, it took companies five years to adopt email.

What I want to bring up is, you know, Neil and I have been fortunate enough to meet a lot of amazing people. Neil's been in business for 26 years or so. I think for me, it's – I want to say, I don't know, 13 years or so, 14 years or so.

But really, I get this question a lot. I don't know about you, Neil, but it's – Eric, how did you build your network? How do you find your peer group, right?

And how do you find AI peers or how do you find marketing peers in general? And so we both have different ways of doing it. I think Neil was very early to the content process. So he was able to build a big audience and naturally that carried a lot of people wanted to associate themselves with him.

But I think there's a lot of different ways to do it. I think there's a few practical things we can share here on how to build your peer network because I do think events and human-to-human connection is going to become more valuable. And I know networking has done a lot for me, and it's also done a lot for Neil. So, Neil, you want to go first?

You want me to go first? Go for it. Okay, so again, like I've been sharing like this, I'm in a couple of these AI founder groups, right? And I have friends that will message me on Instagram saying, hey, can I get into this group?

And I don't own any of these groups, right? And so I do think that let's say you don't have a network right now. I remember when I was 26 years old, I found a group of directors of marketing because that was kind of the niche I was in at the time. And I invited them over to my place and we just had meetups where we talk about our marketing challenges and things like that.

So as long as you're the curator, you can serve and you're saying, hey, I'm going to have I'm going to get a good group of people together. We're going to have a great discussion. Here's what we're going to talk about. And maybe even sometimes if you have some money, you can invite them to like a dinner, for example, or even it can be a happy hour or maybe you can have someone else sponsor it.

Those are all ways to get the right people in the room. And as you do that more and more, people are going to see you as a connector. And that's a way for you to start from scratch without having to quote unquote, know anybody. Yeah.

The main way I actually build my network is in-person events. So just meeting people, talk to them. I'm not talking about just ones I'm speaking at, just any in-person event. And the next one, and this is the bigger way, and you don't have to have a personal brand or be well known for this, is I just hit up people that I like and, you know, reading their content.

Like I was reading a piece of content from Search Engine Land by this guy named Adam Ganus. I'm probably mispronouncing his last name. Sorry, Adam. I did message him.

He may not see it, you know, depending on if he logs onto the platform that I message him because you can do LinkedIn, you can do email, you can pick whatever, you know, platform LinkedIn, I mean, Instagram and direct message.

Um, and that's worked out really well. And, you know, I just read a lot of content. Eric reads a lot of content. I know a lot of you guys consume a lot of information as well.

So it's, I think it's really a good idea to just message people. That's how Eric and I got to know each other back in the day. Um, and I, people message me all the time and I message a lot of random people who have no clue who I am and it works out. And I built many, uh, long lasting friendships from this method.

Honestly, you might think it's too late to create content today, but it's not because we talk about like Gary V likes to talk about, it's no longer social media, it's interest media. So if you, if you post something that's really good, um, people are going to follow you and they're going to want to meet with you. So I'll give you an example here. Um, Neil, remember how I just showed you those stocks, the one that jumped like 8,000% in a year.

That's from a guy. I just saw his post on X and he just started posting. He does a lot of research on, um, all the AI bottleneck stuff. And he literally has a website called AI bottlenecks.app or something.

Um, And he just started posting and he has some of the best investors in the world following him now because he publishes interesting stuff. To him, it wasn't that interesting before, but he lives his life all the time. But when he started making it public, he started, you know, people like to say increasing your service area of luck.

And so even if you might not be doing really interesting research right now, even if you're an intern right now or you're in college right now, share your experience of learning whatever it is that you're working on, build in public, and then people are naturally going to want to associate themselves with you. So for example, my best friend, Mark Andreessen, or my best friend, Gary Tan, for example, from YC or Andreessen, like that's because I've created content. Neil, the way I got acquainted with Neil was because he used to have a blog called Quick Sprout. And I really enjoyed his content around how he spent X amount of dollars on clothes or how he spent on like this and that and how it generated, you know, Y ROI.

But being more serious here, like he wrote a lot of stuff around SEO. And then one time I thought he was making something up. So I emailed him. And then that's how our relationship started.

And it started on Skype, right? But – and here's the other thing. As you create more and more content, you're producing things. You're kind of – you're building in public.

Maybe as an example, Neil, like let's say you're speaking at a conference in Mexico or I'm speaking at a conference in Germany. One time I spoke at a conference in Germany and one guy hit me up and he's like, hey, I've been a fan of your work for a very long time. Let's meet up. We met up, we had, we had great German food, right?

And it was a, that was cool, right? You never know what happens with these things. So the content piece is great. I think putting events together is great too.

Going to events, hanging out with people too. I think eventually you make your own mastermind too. You get a group of like-minded people. We had the same, we've had the same mastermind since 2018, Neil.

I put this group together and I can see this is a group that I'll probably grow old with. So, All that to say is, you know, be helpful, be out there creating content, and then you'll increase your surface area of luck. That's how you build this peer group.

And that's how you get into these AI chats because people are aware of who you are and what you stand for. Dude, totally. And speaking of Adam that I mentioned, who I messaged, and Adam, congrats to you. Love the article.

He wrote an interesting article that breaks down the average time on site for traditional SEO traffic versus LLM traffic. You want to hear some of the stats? It's quite interesting. Yeah, go for it.

So for tool and demo pages, organic average traffic for those tool and demo pages from, let's say traditional Google traffic was 101 seconds. LLM traffic, 146 seconds. So roughly a 46, 45, 46% increase. Homepage traffic from traditional search, 36 seconds.

Homepage traffic from LLM, 82 seconds. So more than double. Service product, 69 seconds for organic. LLM traffic, this was a decline of 63 seconds. And article slash content, organic, traditional 56 seconds.

LLM traffic was 40 seconds. So there was quite a bit of a drop there. Cool. He has a table popped up here. So service pages, punch above their weight, class for LLM traffic.

So service product, I don't know if you mentioned this already, but LLM sessions for 1,000 organic, 30. And then 23 for article, that's number two. And then FAQ is number 14. Home page is 5.6.

This is pretty cool. What company does he work at? I have no idea. So by the way, as Neil pulls that up, tool on demo, I think this is where you called out?

I think this is what you called out. Yeah, he works at Saltbox Solutions. Okay, sounds like an agency. So I would say if you look at this table I have pulled up over here, this is what Neil was just calling out.

So in the middle, you can see organic average time and then LLM average time. And in most cases, LLM time is higher, except for service and product pages, organic time is just a little longer.

Dwell time is mostly longer on using LLMs, to what you said. Yeah, it's kind of crazy. Cool. This is awesome. Well, I think that will wrap us for today.

Hope you enjoyed this, and we will catch you tomorrow.