75,000 Person Agency Is Doomed

The penalty for average has never been so severe. So what do you mean by this? Sure.

Okay, so let me tell you what I mean. So Wall Street Journal tweeted this out, not tweeted this out, they wrote this. It says, why AI will widen the gap between superstars and everybody else. Workplace tensions and resentment will rise if top performers benefit more than everyone else from artificial intelligence tools. But there are things companies can do to level the playing field. So this person, Signal, tweeted, the penalty for being average has never been so severe, but the payout for being extraordinary has never been higher. And when you look at some of the things that you're talking about right now, okay, how do you reduce that rate from 60% error rate to maybe like 10, 15% or so? You're probably leveraging some technology to help you with that, right?

Yes.

I look at the speed in which we can, you know, the fact that with ClickFlow, our internal SEO tool initially was saving our team 20 hours individually per month, right? Now it's saving 55 hours per month per individual. And then now if we're going to add in these other workflows, it's going to save an additional 30 to 40 hours a month. And so that means our people, that means gross margins are going to go up. And that means that people that know how to leverage these tools or these products, even our internal product, they're becoming way more efficient. But there's also people on the team that aren't using this stuff, right? So that's why the penalty for being average has never been so severe. But the payout for being extraordinary has never been higher because we as business owners would be stupid not to reward these people if they're becoming 10x or 100x of themselves.

Yeah. Do you want me to— I agree with you on this. Do you want me to get you the lead breakdown?

Do it.

Do you want me to go back to 6/25? 6/25 is what, June?

June.

June was 8,345. July was 7,549. These are just MQLs. 8 is August. So August is 8,018. September is 8,802.

Okay.

October is 3,703. The month isn't over yet, but it says pacing for forecasts because it predicts how much we'll get this month. It says 6,551.

Okay, so it's down.

It's down.

Yeah. So 8,800 minus 6,600, is that what it is?

8,802.

Uh-huh.

Uh, it says L8W average. What's L8W? Uh, L8W, L8W averages 7,600. Let me find out what L8W is.

That's down 25, 30%.

So let's see, what is L8W average? What does that mean? Yeah, I don't know what L8W average is because he says our L8W average is, oh, last 8 weeks. I don't know why it keeps saying 7,600. That must be broken because it's been saying 7,600 since June. So I would ignore that one. But yeah, it's Yeah.

So I guess we're seeing this trend. And by the way, speaking of agency trends, so we've been interviewing more people from— dude, by the way, Juicebox, juicebox.ai, guys, I'm not getting paid to say this. When you have these agents recruiting for you, looking for the very specific criteria that we look for. So been talking to a lot of these holding company people that have been working at these companies. So like Dentsu would be an example. So I spoke to this guy at Dentsu. Dentsu's about 75,000 employees, right? Something like that.

Yeah.

Yeah, so he believes that, so I was speaking to him and then he believes that they're doomed. I was like, why do you think you guys are doomed? And he's like, well, we're just, we say we're doing this AI stuff, but we're just like a big tanker. We're not moving quickly enough and it's taking us a long time. And sure, we have all these like internal tools and all that, but it's kind of like a joke, right?

Yeah.

And he's like, you know, to be frank, like I'm speaking to you right now, but I'm also like WPP and Publicis, which are other big holding companies, have tried to poach me, but he sees the same outcome there. And he's like, it's also not a good signal when like Odensee is trying to sell 'Cause that doesn't show like a lot of confidence, right? And so I thought that was interesting. And then later today, I'm grabbing dinner with a guy who's a candidate, and he actually works at WPP right now. I was like, why are you leaving? And he showed me all the stuff that the, you know, their, you know, stuff that he just talked about, the stuff that they're working on. And he's like, yeah, it's the same deal. I don't feel like I'm growing. And it's like, it's moving too slowly right now. And so the question is, Already 75,000-person agency is doomed. I don't think they're going to die immediately. I think there's going to be a little bit of pain, but I think that's reality, right? And it's always been like this, like startups, like smaller companies are going to move faster. They're more nimble. And like these companies too, they have to protect— it's more about wealth preservation for them instead of trying to produce, like trying to move very quickly and break things all the time. So I thought that was interesting. I mean, we've been having more and more of these conversations, but what's your take?

I don't know if they're going to break. There's not too many options for global help.

Yeah, I don't think they're going to break, but I think it's becoming challenging, more challenging.

I think it's been challenging because the economy's gone, but okay, let me ask you this. You've been using AI. I've been using AI. Have we really seen AI grow agencies like crazy and change our business? Like honestly, I'm not talking about your workflows. I'm talking about from a new business standpoint and client retention standpoint. I haven't really seen it change agencies yet and marketing yet. I'm not saying that we're not using this technology to do better, cooler stuff, save time, but what I'm getting at is when I go to events, whether it's HubSpot or South by Southwest that I was just at, and no matter what country I'm in, and when I ask people have they seen an increase in revenue from the usage of AI in marketing, very few people ever raise their hand.

Yeah, I don't think we're seeing— so here, this actually, this will align with what you're saying right now, and then I'll kind of address it. So McKinsey dropped a stat that should terrify every CEO. They just reported that 80% of companies now use AI, but only 1% are doing it well, meaning that only 21% redesigned core workflows for GenAI. Just 28% have their CEO personally leading governance. And the last thing here is those who did. So those CEOs who are personally leading governance, and I would count your— you're not a CEO, but like I would count yourself as you're a founder pushing it. Those who did are already reporting higher EBIT from GenAI. Yeah. And so my take on it is like, there's pockets of this, right? But like, you see, I do see reports of like, oh, this smaller agency is getting 80, 85% gross margins from most of the AI work that they're doing. That those numbers maybe need to come down a little bit. But, you know, I think for us, like you look at your engineers, my engineers, like, yeah, if they're even 3 times faster, 5 times faster doing that amount of work, I mean, that flows into our gross margins. Yeah.

So on a side note, they gave me US lead data. Do you want US lead data? Yeah, yeah, yeah, yeah, yeah, yeah. This is not MQLs. So before I give you MQLs, this is just leads. So the numbers are going to be higher, but keep in mind, majority are not qualified. So I only have data going back to 7/1/2025, because that's the screenshot he showed me. So 7/1 is July. It's 20,670. August is 17,944. September is 18,111. And October is 8,088. Again, the month's not over. It looks like from my calculations, it's pacing for around 14,700.

Mm-hmm.

He claims it's pacing for at least 15,000 because they look at weekdays versus weekends and when the month started. But he actually thinks it's going to be closer to around 16,000 this month for leads, not MQLs. Keep in mind, MQLs go through more of a tougher criteria.

Yep. Look, so I think all these things kind of go together. Like the penalty for being average has never been so severe. So yes, start to learn this stuff. And then also the majority of companies are not using AI well. And I think the majority of companies are failing to report on the benefits of it. But you saw the thing I shared on my Instagram stories yesterday about like, The new cancer breakthroughs from Google, right? Like, I actually think cancer's not gonna be— it's gonna be a non-issue, probably could be 10, 15, 20 years out, like knock on wood.

I hope sooner than 10, but yes.

Yeah, but you see where this is like, I think the— you and I can see the long-term benefits of— I think we're starting to see the revenue benefits from it, but if you were to ask for this maybe like 2 years ago, I think we'd be hard-pressed. I think we're seeing a little more now for sure.

I'm curious what the AI companies are gonna do with this, 'cause From what I've seen with all this cancer research stuff, they're working with the pharmaceutical companies and they're helping 'em do all the research. Correct me if I'm wrong.

I don't know.

And then from my understanding, a lot of them are paying for this technology. They're in labs, they're doing experiments. But, um, it would be cool if someone other than a pharmaceutical company, uh, released it when there is a cure, hopefully sooner than later, released it. So then that way they don't charge like $100,000 for the drug.

Yeah. You, you're talking Because at the end of the day, the pharmaceutical companies, they have their incentives, right? The reason they're able to charge so much and all that. And like, you know, I think we will see different studies. I think some of them are working with them, some of them aren't. And, you know, maybe they're gonna wanna release this data where, I don't know, it benefits the rest of humanity and it's not just the pharmaceutical companies. I don't know. But here's one thing.

I don't think so.

Let me come back to one that I was looking for this in my studies. This is the beauty of me reading X, right? Ethan Mollick tweeted this. So one of the first randomized controlled tests, so it's a controlled trial testing whether GenAI boosts revenue, not just productivity. So this is a controlled study on whether AI drives revenue, okay? It says it does. A large mature e-commerce platform using older GenAI models, GenAI tools found most of them, from customer service to marketing tools led to large significant revenue boost. So this is one of those like these large abstracts, right? When people do these studies, like you can see who did it. Yeah. What I'll call out here is across the 4 GenAI applications with positive effects, the implied annual incremental value is approximately $5 per consumer, an economically meaningful impact given the retailer scale and the early stage of GenAI adoption. So when you look at this over here, Neil, you can see these are the 4 different business workflows here. So one's a chatbot and it shows the incremental value per customer. So 0.274, that's like 27 cents or so. That's the upper bound, right? Lower, let's just call it like 25 cents incremental. Search query refinement adds about like 6 pennies or so, and then product description adds about 1 penny, and then marketing push messages add like 0.0004, right? So there's like a time multiplier, all these things, but when you add it up, it's about total linear additivity is about $5 per consumer.

That's not bad.

I mean, I don't know what platform this is, but if it says large significant retailer, I'm like, who could that be?

I don't know who it is.

But that's, this is like, we're going to need to see more studies like this. I guarantee you guys would do like really well if you just do a bunch of studies like these.

Yeah. Those studies that you're showing on your screen, some of you guys can see, some of you can't. Those are just really expensive and time-consuming to do. What we do is we focus on stuff that gets included in presentations.

Dude. So this one's from Cornell University. Like, how do you pay these universities to do these studies and then attribute it to you?

You could.

You go do it first. Let me know how it goes.

I don't want to spend the money.

I know.

I'd rather just do stuff that people share internally.

Yeah.

Because that just keeps us top of mind when they're doing RFPs. Like we were talking about how you're saying for you and me, based on the lead count we've both been showing, it's been tougher periods, right?

Yeah.

This is probably going to be our highest month in new sales closed.

Yeah.

Now again, a lot of the deals are older, But for us, you get an RFP and it comes in and I'm just making up a number. Like we're pitching an RFP right now. It would be $7 to $10 million in fees that we would just collect. You close a deal like that, it doesn't matter how many leads you get from a website or not. Like you close it, you having a good month.

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 the difference between a $10,000 a month contract versus $100,000 a month contract?

$100,000 a month?

Yeah, $100,000 a month.

What's the difference?

Okay, so literally I'm listening to this call.

Okay.

All I had done on these discovery calls, so 7, 8 years ago, I was on these discovery calls. Literally, I would just say, hey Neil, tell me what your goals are exactly. Okay, so what metrics do we need to hit exactly to be successful? Right? And you see where I'm going with this. And then I'm like, okay, Neil, what's the value of meeting that metric?

Oh, okay.

$10 million. Well, Neil, let's just say hypothetically, we can guarantee that we can hit the $10 million, 100%. Is it fair for you to pay us $1 million? Why, yes, Eric, if you can guarantee it, right? Nobody, like what idiot is going to say no to that, right? I'm going to, I'm like, I'm listening to this segment of the call and literally we priced it on value. We said, okay, we, if we think 50% success rate for us, we're going to take that 50%, multiply it by that 10%, right? Which is $1 million.

Yeah.

Then that, the high end of the contract is going to be $500K, then $250K, then like $125K, okay? Instead of just saying, we're just going to charge you $3,000 a month, which is what SMB people are used to selling. So long story short, I listened to that segment. We had priced this thing on value, okay? And basically one of the deals was, the guy was like, oh yeah, no, we don't do the smaller ones. We're going to do the $100,000, okay? But like that was just reframing a little bit, asking a little more, asking about the value. And you can't do that for everyone, by the way, just keep that in mind. But the more you can frame around value-based pricing, the more successful you're going to be. I've literally, I've like, we've done this, like we have like a couple calls recorded where I just did this and we just, for whatever reason, we stopped doing it. And you can't do this for enterprise deals, by the way, as much as you'd like.

Yeah, no, no, I was like, this won't work for us.

Yeah, but you can do this for, we're talking like smaller or mid-market or so, especially if it's founder-driven, they're just like, no, no, no, especially when there's ego sometimes too, if they raise a lot of money. Oh yeah, yeah, no, we're just gonna go for the big one.

Yeah, Gijon from Spiralize, they're based out of Atlanta. They do this really well. Their agency does a lot of performance-based stuff. But you're right, this kind of stuff works for SMBs. We don't do it. We've thought about doing it years and years ago. It just doesn't work. Like you go to a company that makes $10 billion a year in profit, forget revenue.

They have their forecasts.

Yeah, they have their forecasts. Their CFO's like, oh, this is what we're telling the market. Like they're not going to do anything that can screw up their stock price.

Yeah. So it's just, I think our thing here is like, it depends.

It depends, but I think you should start implementing this at Single Grain again.

I'm just like, dude, like there's so much game tape and like the, my blood boils as I listen to that thing. I'm just like, dude, it's really not that hard. I think it's, you know what it is? I'll just say for those of you that are listening to this that, that have the ability to implement this, I think it's, it's for you, your job is to kind of go there and disarm your sales team and say, hey, look, it's okay. Like I know you're used to selling it this way, but I'm telling you, you're, if would you just frame it first, hey, would you like to make more money? Would you like to make 10 times more money, right? Maybe just frame it that way. And then like you can have a coaching conversation because, and oftentimes if you're the founder listening to this too, maybe they're a little scared to listen to you. So just disarm them and just say you're here to help them. And that's how you make it easier.

Dude, one thing that people get wrong with sales and marketing, and I hear this all the time, because a lot of people ask me, how do I go upstream? Even when I was in Singapore, not the most recent trip, but the one before, I did an interview with Tim from Ahrefs and he was talking about how to go more upstream.

Guess who set that up?

Yeah, you said that.

That's right.

So I appreciate it, by the way. And when I was with Tim, he was asking me questions like how to go more upstream.

I actually listened to that one.

Yeah.

It was a long interview.

It was a long interview. And I get that question a lot. And people believe that enterprise contracts are for a lot of money and they're sticky and they don't churn as much and you can make really great money. And a lot of those things are true, but what they forget is most enterprise contracts, even if they're RFP, sometimes they do, but majority of the time they don't start out at $2 million, $5 million, $10 million, $100 million. They start out really tiny where they test you out with something that's $100,000, $200,000, maybe even smaller, sometimes a little bit larger. And if you do well, you keep getting more contracts, more divisions, more regions, because these enterprise brands typically have a lot of different product divisions because they bought a lot of companies and they're in a lot of different countries. So you can grow your contract by expanding by regions or by division and selling multiple products for them or services. And then it adds up to multiple millions of dollars over time, but it's just a long game. And people just believe that, oh, I'm going to end up getting these enterprise contracts and it's going to be great. It's just like, no, you're going to close a contract and you're not going to get what you want in the first 12 months.

And you might not even be profitable. You're going to take it on the chin, but you're playing the long game. It's, it's, and by the way, when you're in, you're in.

Like, dude, we just had one enterprise company that we work with in one region introduce us to another region. And they said, okay, now that we introduced you to another region, we want a discount.

Mm-hmm.

And I was like, wait, wait, wait, we haven't even closed the other region. And they're like, yeah, yeah, but we want a discount. And there's a certain point, and I was telling them like, we already have terrible margins with this, with your business. And I just shared the results with them. Like I literally was very transparent. I'm like, here you go, here's all the profitability on your account.

Mm-hmm.

I'm like, if I do what you're asking me for, I will lose money. It doesn't even make sense for me to keep you on as a customer, but it's not that easy to deal with depending on the region because each region, their procurement departments have different cultures. And some regions, their goal is to just continually squeeze and squeeze and squeeze. And there's a point where you have to say no. Some companies are great though. Like we had one company ask us to go above and beyond, be there for a lot of meetings, Do things that were out of scope, help us on other regions that we didn't even have a contract for. And if we didn't do those things, they would get upset and they would rate us very poorly when we were doing the reviews. So we sat down with this gentleman, older gentleman, I think he's in his 60s or something like that, nice guy. And I'm like, look, I know you're not happy because we're not doing these things. Here's our contract. Here's what you're asking us for. Here's our profitability on your account. We're losing money and I know you're not happy. We're discussing on how we move forward. We're open to just ending the agreement or we can readjust the agreement. And they're like, oh, I didn't realize you were losing more money. And this gentleman was really kind and they're like, look, it's in our company culture to just keep grinding our vendors. And I'm like, I get it. You should do whatever's best for you, but I don't want to lose $300,000, $400,000 on an account a year. That's a lot of money to lose. And they're just like, okay, no problem. We'll fix this. We'll stay within bounds. And, you know, if we want extra work, we'll add additional scope. And I was like, cool. But sometimes you just have to speak up and let people know.

By the way, when I, when I said, um, once you're in, you're in, what I really mean by that is once you're in with the enterprise company, procurement sometimes takes a very long time and you might be taking on a chin for a year or two, But once you're in and you're doing a decent job, they're going to try to expand you because it's easier because they don't want to go through procurement again.

No, they don't, but they usually have to for expansion. It still goes through procurement, but it's easier because you're in your system, you're already a vendor, they can just easily add you. And for certain things, they don't have to go to RFP again because they have different rules depending on the company. If they onboard new vendors, you know, sometimes they would be like, okay, you want to do this? You want to try to find new vendors? All right, let's issue an RFP. They don't want to waste 6 months to a year of their time. They're just like, let me just onboard someone quickly and just get it done.

It's interesting for us because a lot of the agency deals we have are, I would say, probably more mid-market, but on the Carrot side, we're going through procurement quite a bit. Like it constantly gets thrown in front of us, but you can pick the next one.

I look at, before we pick the next one, I look at your Single Grain and Carrot business as 2 different types of businesses because it's different customers.

Yeah, I look at it as the same because Single Grain's ultimate mission, and my team will and this, but the ultimate mission is focusing on B2B. And so it's the marketing partner that focuses on B2B customers. Sorry.

Yeah.

Yeah.

No, that makes sense.

Yeah.

You have one in here on, is SEO jobs dead?

Yeah. So are SEO jobs dead? Are SEO jobs dead? So this is interesting because at the Ahrefs conference, you remember Ryan Law who wrote, he used to be at Animals and now he works for Ahrefs, a great copywriter. Writes great blog posts. So his talk was actually on, I didn't get to see his talk, but we talked for a little bit the day before, but there is a general fear, like I could sense it in the audience too, on like, what's going to happen with my career? And my talk actually became more inspirational, right? And I was like, guys, like we as SEOs have taken it on the chin. Panda update, Penguin update, Donkey update, right? Like all these updates, right?

Yeah.

And so SEO jobs, like we've had to adapt the most. And so in my mind, it's like, I don't think SEO jobs are— I think if you just focus on traditional SEO and you don't adapt at all, like, yeah, it's probably just like how paid media jobs are, as they, as we know it, are probably dead, right? But my whole communication was like, look at all the things that are happening right now. We're talking about search everywhere optimization. ChatGPT just came out with apps, you know, ChatKit, like all these different things, right? So in my mind, we've talked about this, like, I'm excited about it. And then people came up to me afterwards, like, oh man, that was really inspirational, man. Like, I didn't intend it to be that way. But I kind of sensed the fear in the audience, like, oh, where do I go with our career? Because that was one of the things that Tim brought up like the day before, like he put it on Twitter. It's like, yeah, like where do we go with all our SEO skills? And I'm like, well, you've learned to adapt and take it on the chin. So like, just keep doing that.

Yeah, you don't really have to do anything. I agree. I don't really see them going anywhere. I think they're still going to be valuable as long as you adapt and they have adapted over time, so they should be fine.

So that's it for today, guys. Please don't forget to rate, review, subscribe, and we'll see you tomorrow.