Ep. 3004: ChatGPT Quietly Switches From Bing to Google

Eric, did you see ChatGPT has quietly replaced Bing?

I do see— you picked the worst one. I was complaining about this topic that you picked. I was like, God, Neil picks the worst topics. But no, it is a big deal. I actually didn't see this, so please explain.

Yeah, so there's this French blogger, alexseo.com. He has a newsletter and he did an analysis with certain queries. When you look at the search results in ChatGPT for some of the ones that he pulled, I don't read French, you can't translate the image as easily as the text, but roughly 30% of the results were from Bing, but on the flip side, 90% were from Google.

You know, this is interesting to me because aside from this, you look at the Microsoft OpenAI partnership, it's getting shakier and shakier, and then you've seen the commentary on X where people have said, Sam Altman's lost so many executives, right? Like before, like the CEO—

Facebook and everyone.

Recently, yeah. But even before he lost like Mira, he lost all these other people resigned, right? And then there's so many people that have said that Sam Altman, and I can't verify this, but this is just what I've seen on X saying that, you know, he's not like a moral upstanding guy. Like he has problems with Elon too, whatever. And then I've been reading through about the Microsoft partnership, how they've been trying to negotiate this deal and like get out of like certain splits that they have to deal with from like a revenue sharing standpoint, right? And so this is happening over here. ChatGPT is like, okay, screw you, Microsoft, we're just going to move to Google. So I think it's just going to get more and more complicated. But I don't know, like on one side it's like Sam Altman has created one of the most important technologies of our time. But on the other side, you just start to hear so many of these things and it's like, you know, it's— it gets a little shaky, I think, you know, when you look at it from like a company standpoint.

And I don't see Microsoft being flexible on their deal. They gave too many billions of dollars. The company wouldn't be where it is if Microsoft didn't pour all that cash in early on, which was a big risky bet. Now, yeah, Yeah, you can go look at it being like, ah, it's not that risky, but back then it was. They poured what, supposedly $10+ billion into that thing and they deserve their revenue cut.

And I doubt Satya or anyone on the legal team is going to be like— No, it would make them look bad.

Yeah, they're like, sure, we'll give you— give us less revenue. They're like, screw you.

So I actually listened to an interview yesterday with one of OpenAI's co-founders, Greg Brockman, I think is his name, and with one of the Collison brothers, right? And Greg was like, dude, so many times in OpenAI's early days, they almost died. They didn't know what to do. They were like building like gaming stuff and like they didn't know that it was going to be this chat thing. I think they were trying to figure out how to sell these APIs in the very beginning and somehow they figured it out, right? But without Microsoft's war chest, I don't think they would have figured it out. And now it's like, it kind of seems like from the outside, us looking in, it's like they're kind of biting the hand that used to feed them. And, you know, I don't know. That kind of sucks. And Microsoft shouldn't let up. I don't think they should.

No, they shouldn't. But speaking of doubling down, I think that is with most businesses and startups. Even when we run marketing campaigns, we don't know always what's going to resonate and convert. And, you know, I would love to hear how you guys have pivoted over the years and potentially died. At NP Digital, you know, we started as a consulting company, then had software. We've had some rocky paths in between, but the company grew through a decent economic time. So the scariest moment was probably COVID. Yeah.

Right?

Because we didn't know what was going to happen.

Even the very beginning.

The very beginning of COVID But, you know, we've had a decent tracker, I'm knocking on wood, and we've had some scary moments similar to you and other agencies, you know, where people are like, oh, AI is going to kill marketing agencies. AI is going to kill SEO. Now, you and I didn't believe either of those 2 things, but I have to say, and I know you agree with this, we're also biased because we're in that industry. And we also think the Sam Almans of the world are biased because they're building the platforms and yes, they want the biggest TAM, so they're going to go with the pitch of we're going to be able to do everything that people do and you don't need to pay these people. But even like I look at right now in my company, you know, internally we were focused quite a bit on Amazon over the past 3 to 6 months and we still do quite a bit of Amazon stuff, but I've really shifted the team's focus to LLMs and I'm like, why not try to be the agency that really helps people with everything LLM related? And double down on that because I think that's a much bigger pie for companies like yours and ours versus, you know, focusing on Amazon where the 3rd largest seller is only doing $800-ish million in revenue, which is a lot, but it just shows it's fragmented with a ton of SMBs.

Yeah. So I don't think we've necessarily changed our opinion because you're the one that came up with search everywhere optimization. And, you know, companies call it different things. I, and I just call it what you call it. And so I do think that that naturally falls into— and here's the interesting thing from my side. You've seen so many of these VCs back these like Profound or whatever. I actually talked to another company this week. They're building one of those LLM, you know, surface trackers. You're doing the same thing. You're adjusting Ubersuggest now, right? There's so many of these coming up now, and it seems to just be moving in that direction. And it's— I think in this case, it's more so this is just where the tides are going to, the wave is flowing.

Yes.

So why push against it? And I think it's the ones that usually have a different angle in the beginning. Those are ones that stand out because you have a point of view.

Yeah, but I think as a company, when we're trying to come up with our unique value proposition, right? How do we adjust our copy and our positioning to stand out in a crowded place where everyone looks alike? You know, we differentiate through technology and we're going to end up showing unlike other agencies, we actually have Alum-based tracking and our own proprietary technology. But I've also seen you pivot as well. If you look at your history, Single Grain is now what, 10+ years old?

Single Grain started in 2009. I took it over in 2014.

All right, so it's 10+ years old of just your ownership.

Yeah.

More than 10 years old when you look at total.

Yeah.

But when you look at Single Grain, it used to just be a marketing agency that really focused on SEO.

It was just an SEO shop, yeah.

Yeah.

Yeah.

And then you added in the other service, social, paid management, email conversions, a lot, yada, yada, yada. But if I look at you now, you've had a huge shift into ABM and focusing heavily, whether you say it or not, B2B because of the ABM, right? It's mainly B2B companies that are using it.

I'll give you the angle. I mean, so for us, we're focused, we call it B2B OS with Carrot and everything and Singularity. It's really all encompassing. So on our end, we're just like, let's just focus on B2B because B2B is often so far behind B2C, like 5 years behind. And I think we can do a lot better if we just— our focus is on making B2B, this B2B operating system easier for everyone. That's already a wide enough TAM because B2B is what, $20 trillion a year or something like that?

And it's massive and it's actually stickier. The B2C has much more ups and downs and higher churn from what we see than B2B.

And so we're just going to keep building technology around it and hiring more people around it because I think the more you focus about it, just like if you're going to say, okay, we're going to be the LLM agency, great. If we're going to be B2B OS, great, right? I think you need to have some type of positioning here in a world, to your point, where everyone looks the same and You and I, I think it's, I think the problem with a lot of agency owners is they, they don't adapt quickly enough. And that's why when we talk to people from recruiting right now, a lot of the people that we're pulling over, they're like, the agency is just not adapting. They don't want to do this AI thing. They don't want to skate to where the puck is going. And so it's, it's like, kind of like, I don't know, stealing kids from a, stealing candy from a kid when you're talking to people right now, because all you need to do is paint the vision around AI and what you're doing and instantly like their eyes light up, right?

Yeah.

Um, so I don't know about you guys, but recruiting has gotten a little easier for us because you paint that vision now, so.

I actually think recruiting's gotten easier because there's not that many agencies that are hiring. The whole agency space, according to my CEO, I don't know what article he read, has contracted 4% over the last 12 months.

Oh wow, wow. Have you been paying attention to the holding companies?

I have, I have.

Okay, so how's that going?

Publicis is probably doing the best outta all of them and they made some really good bets back in the day and they've been paying off. Their most recent bets have been around influencer marketing and they've been doubling down. They bought multiple companies in the influencer space, so they're trying to go for reach and just own the whole space. And they're saying that is a big future. The Omnicom-IPG, I believe, merger, it looks like it's getting—

It's going through. Yeah.

No, it looks like it's getting regulatory pushback.

Really? I thought I read a couple weeks ago that it was going through.

It could still be going through, but I read also that they're getting regulatory pushback. It doesn't mean that they won't find ways around it, just like Activision, Microsoft, right? They figured out how to compromise and make the regulators happy. I think Dentsu, they've been struggling. The rumor is that they've been, they're planning on selling. They've been liquidating assets like Mute6 and others that don't fit the story. So my guess is they will. WPP has been struggling. Martin Sorrell's S4 Capital, which bought MediaMonks and a few other companies, they've been struggling as well. I'm not trying to talk Crap, a lot of agencies just have been struggling. These ones are public, so you can just look up their numbers. Yeah, I think just the overall space for the holding companies, what's happening is like if the Omnicom and IPG merger go through, you're going to have a bigger organization that moves slower, run by accountants, cutting more costs.

Yep.

From all the articles I'm reading and the trends we're seeing, I see a shift happening over time to more people pushing towards independence.

Dude, so you just hit the nail on the head. In my opinion, it's bureaucracy that holds a lot of these holding companies back these agency holding companies, it takes too long to get things done. It's slow as molasses. And then when you have people at the top where it becomes more of a math equation, and I'm not saying math's not important, finances are important, financial models are important, but when it becomes all about that, you're actually losing sight from focusing on customers, right? And I'm reading that book right now that I told you about, the—

The billion-dollar one, right?

Yeah, Company of Giants, right? And when you read about like Steve Case of AOL or Scott Cook from Intuit, right? These guys were just like, dude, you just have to, it's all like, one of the quotes is, it's all about the customer, stupid. It's not about this financial model, but the moment things start to really grow, and this was in the Steve Jobs interview in there, when you have people at the top, like, I think it was, God, I remember Steve Jobs hired the Pepsi CEO and then the Pepsi CEO, all they did at that point was they focused on private jets flying to nice islands and things like that. And it was all about making sure they have nice company cars and all that. It was no longer about the customers. And so I think for you, because you're founder-led and you own the lion's share of the company, you get to call the shots and where the— you understand, again, fundamentally your job is to help these companies grow and you have to understand where the puck is going. And that's all it is.

Yeah.

It's focus on the customers, That's all.

I agree with that. But the hard part is, even though we're founder-led, like my co-founder and I were texting back and forth when I was flying back from Hong Kong, and he actually said something to me that was really interesting. He's like, dude, business was so much easier when we were smaller, like $30 million or less in revenue.

Mm-hmm.

Because right now we're trying to implement some AI technology. I've talked to you about this where AI calls all of our leads and We have some other things in the works, so it's not just as simple as them just calling our leads. There's some other cool things that we're trying to do that I don't want to go into. Talking to my co-founder or texting him back and forth when I was on the flight, we've been trying to do this for like a few months now and we can't do it.

Yep.

Because we're bound by laws and laws are different in EU versus the UK versus here versus Brazil and Mexico and Canada, and the list goes on and on. And there's a lot of privacy and regulatory issues or legal issues because a lot of these software solutions will train on your calls and use that data for other things. And we're not allowed to do that based off of certain countries and laws and how we collect leads. So I can't implement it and test it out. And my co-founder and I were texting, we're like, man, if we were a startup—

Oh, instantly.

People just do it because you don't have much to lose. But at our size, I'm not saying we're massive.

But you also have bureaucracy holding you down.

We don't have too much bureaucracy. When him and I get involved, we can just get through it.

But the hard part— When I say bureaucracy, I mean the regulation from like the EU or whatever.

Yes. Yeah. We have bureaucracy that's holding us down. But as a startup, this is going to sound bad, but this is reality. Most startups just ignore bureaucracy and the laws. They're not going to blatantly break the laws like selling drugs, but they're going to ignore it like a ChatGPT scraping a New York Times and learning from it, right? I'm not saying that's what happened or didn't, but I think that's what the case is all about. And you're skirting around laws and no one cares because you don't have much to lose. But when you're larger, you have a crap ton to lose. We don't have anything to lose like ChatGPT, but still it's a lot for us.

You're still a target.

Yeah, we're still a target. And that just, that really does slow us down and it just makes us not grow as fast. And then this is like, all right, if we were sub $30 million, dollars a lot smaller, you know, back then, I don't know how many people we had, maybe 200 employees, right? We were like 1/5 the employee count, maybe even less than that, maybe like 100 and something employees and 1,000 plus right now to give you guys perspective. It's just like, all right, who's going to really sue us? Just go for it. They're going to be like, oh, we're small.

Let me give you the counterfactual. Elon does a lot that he wants to do. For example, I'm going to start a new party. I'm going to start the American Party, which is like, He should be like, I think you and I could both say like, he probably shouldn't do that.

Correct. And Eric and I are not talking about politically what he should or should not do. We're talking about when you look at it from a shareholder perspective, a lot of the stuff he's doing is just causing people to get pissed off. And it's also causing his Tesla shares to go down, not just because of the political side, but people are also boycotting Tesla, which everyone's seen on the internet. And it's hurting their sales.

Yep. If you're in B2B, here's the deal. LinkedIn account-based marketing sucks, but it doesn't have to with Karret. 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 karret.ai and you can join the waitlist. Again, it's Karret, K-A-R-R-E-T. And we'll see you on the other side. Your point around this was that LLMs are now going to drive a lot of agency revenue. And now we're kind of talking about, you have this other piece over here on how relationships matter more than ever. So I feel like all of your topics today are centering around one thread. What is that thread?

Well, let's actually go back to the one that you just mentioned on how LLMs is going to drive agencies quite a bit of revenue. We're starting to see the ChatGPTs, the Perplexities, and other solutions. recommend agencies for LLM optimization. We're starting to get referrals from the actual LLMs.

LLMs. We are too. Yeah.

And I'm like, it's a great channel. And when I say referrals, I'm not talking about someone going to ChatGPT, uh, typing something in. I'm talking about an employee at like a Perplexity or ChatGPT is saying, hey, you should talk to NP Digital. Yeah, we're starting to get business from that.

Yeah, that's awesome.

When, when you look at traditional agencies, most people don't know this, but Google and Facebook drive a lot of revenue. When you look at Mute6, what caused Mute6 to grow was Meta.

Meta.

Yes.

Yeah.

The people who at Meta who were in charge of managing their top advertisers would recommend Mute6 when those clients or the accounts that they were managing would say, hey, do you know a good shop that I should talk to for managing my paid ads on Meta?

Yeah.

And Google and Meta drive a lot of revenue to agencies if you get on their good graces.

How did you get people working at ChatGPT and Perplexity to refer? Because it's like, usually that's not a normal thing because like, at least with Meta and Google, you have that relationship. But how is that happening with ChatGPT and Perplexity?

We're building relationships with them.

Oh, okay. You're reaching out, building relationships, events, things like that.

Yeah.

Yeah, that makes sense. Okay. And Continue. Let's go to the next piece on relationships matter more than ever because this is actually related. You're building these relationships.

Yeah, because in a world of AI, you can get recommendations from a platform like ChatGPT, and that's great. And I think for small things, inexpensive things, new toothbrush, a camera, a television, I do believe people are going to use ChatGPT because it could be more analytical. But when it comes to like big expensive items, like someone signing up for Single Grain or NP Digital, our contracts are not cheap.

Mm-hmm.

Right? I don't know what your pricing is, but I'm guessing your pricing is 6 figures plus for majority of your contracts. Because it's such a big decision, they're not going to just do research on ChatGPT and say, ChatGPT told me so I'm going to sign up. You'll get a small percentage, but I genuinely believe Just like most service-based businesses, the majority of your revenue is still going to come from referrals and relationships. I don't know how it is right now on Single Grain, but I'm guessing because you've been around for more than 10 years, a large chunk of your revenue comes from referrals and relationships of just knowing other YPOers, EOers, and then referring business to you. Would you agree with that statement?

Yeah, I would. And, you know, actually I was just telling Noah here, my past 2 days' sleep has been bad because, you know, I was at a— I hosted like an AI founders dinner around here. And then yesterday was AI potluck. It was an AI like founders potluck. Outlook, right? And this is, it's no different than joining a peer group or like an EOYP or Hampton or something like that, or even us hanging out or like, you know, us having mutual friends where we hang out like once a year. And so it's in a world that's really driven by AI, people are going to long for more human connection and who actually controls the power at the end of the day, it's actually the relationships.

Yeah.

It's sure you can be number one listed and that is important, but nothing beats relationships more because Everything is people to people and people like doing business with people, not necessarily a ranking. Yes. The ranking is usually like when you're not sure what to look for, but when you have like a word of mouth thing, oh, Neil, who do you go to for a dentist? Oh dude, I go to this dentist over here and you get to watch TV and you wear these glasses over here. Like you're telling, you're selling it to me, telling a story, right? And then my friend's like, oh, you got to go to this eye doctor over here. Like they'll do this and that. Like sold, done. I don't need to think anymore.

Yep.

Because I trust the person.

When with your agency, with your revenue, You do a lot more than a million bucks, but let's just use million dollars as a hypothetical.

Yeah.

If you looked at a million dollars, what percentage of the revenue would you say comes from referrals and relationships?

I'd say right now it's probably like 30 to 35%.

Oh, wow. Yeah, that's good.

What is yours?

70-ish percent. Yeah.

Yeah.

It's, it's— For enterprise, for SMB, it's probably 20-ish percent referrals, 80%, um, inbound and other forms of marketing, but from enterprise, which is majority of our revenue, you're looking at 70 plus percent.

And that makes sense because people move around. Yeah.

Right.

And that's most of the 70% is coming from people moving around that you did a good job for.

Yes.

Yeah. And the thing is, by the way, we'll call this out too, as an agency and even a software company, you're not going to do the best job 100% of the time.

No.

Maybe like, what do you think the percentage of time you're doing a good job?

It depends what you consider a good job. Like if you're talking about positive, producing positive ROI, positive NPS, CSAT or something. Yeah. It's majority of the time by far.

80%.

Yeah, 80+% I would say, but there's going to be like 5%, 10% of the time where things just don't align and you can't always control it. A great example of this is if you're doing marketing for Nike, okay, I'm giving you a real hypothetical here. We'll see what happens in the next few months. The US just passed a deal with Vietnam where it costs 20% more. I believe they implemented a 20% tariff. I don't know what it was before, but let's just go with that.

Yeah.

It costs roughly 20% more now.

Yeah.

That means Nike, who produces a lot of products in Vietnam, they have to, they don't have to, but I'm guessing over time, because they usually say the business owners pass that cost over to the consumer because everyone has to end up paying it. It's going to hurt the people who are doing marketing. So if you're doing marketing for Nike and you're producing amazing ROI, but just hypothetically imagine they increase their prices 20%. You're not going to perform as well and it could be a loss. And that was a campaign that didn't go well. But is that your fault? Is it external factors? The reality is it doesn't matter. It didn't work whether it was your fault or because of external factors or because of the company paying you. They had a shift in people and they never got anything implemented. You're never going to have 100% success rate, but when you fail, it doesn't matter if you were in the right or wrong, it still hurts you. And when you do good work and people are happy, it doesn't matter if you're in the right or wrong. You do some campaigns are messed up, but overall you made them a lot of money. You'll keep getting referrals.

What matters is ultimately, if we were to simplify this, is that you're learning and then you're just improving and then you're just being consistent for a long time. And that's really what, how all of this works. So we're going to end the episode here. See you guys later. Please don't forget to rate, review, subscribe.