Why Website Traffic Is Almost Dead Today

Neil and Eric break down how AI search is crushing publisher traffic, why informational content is losing, and where transactional SEO still wins. They unpack search everywhere optimization, local AI model economics, Mac Studio RAM needs, and Intercom’s Fin turnaround from slowing SaaS growth to a major AI pivot. A sharp episode on adapting fast, measuring revenue over traffic, and using creative destruction to stay ahead in AI marketing. Key takeaways AI search is killing informational traffic Transactional SEO still drives revenue Intercom’s AI pivot shows how brands survive Chapters (00:00) AI traffic collapse (01:01) Informational vs transactional SEO (03:31) ClickFlow AI SEO break (04:04) Search everywhere adaptation (05:16) Local AI model costs (06:50) Mac Studio RAM for AI (08:27) Intercom AI pivot (13:26) Creative destruction strategy (20:13) AI hiring and staying ahead

• 0:00Speaker 00

  Finding customer information shouldn't mean digging through emails, call notes, and spreadsheets. HubSpot Starter Plan brings all of those details together, so it's easier to track what's happening with every deal and what to do next. It pulls information from calls and emails. Then it gets to work, logging and summarizing them. Right after your call ends, HubSpot pulls next steps from your call notes and drafts suggested follow-ups. And it stays up to date for you, so you can build demand, win more deals, and spend less time on repetitive work. Your records, insights, and next steps are ready for you to act on. And you don't have to miss a beat searching for any of it. Try the Starter Plan for $10 a month. Learn more at hubspot.com. So Neil, let me show you this, this table over here. You actually, um, you know, you hit on one of the topics I have here. So I checked this out.

• 0:39Speaker 00

  So this guy over here, Danny Critchon, Critchton, Critchton, Danny Critchton, posted this. So he says, "No discussion of tech media can get past this basic traffic fact. In the AI world, Google on social no longer refers to traffic, which means that the vast majority of readers just never find you in the first place." Here's the analysis over here, right? So you see digital trends, peak traffic, 8.5 million in 2024, March 2024, January 2026, 264,000.

• 1:06Speaker 01

  Wow.

• 1:07Speaker 00

  97% decline, right? Remember ZDNet? 7.6 million visits per month, February 2024, 768,000, minus 90%, right? So like wired.com, 7.7 million, November 2024, drops to 2.9 million, negative 62%. Dude, so I think, by the way, like what Neil said, he's not, he's not wrong, right? But I think it's just really interesting to look at these big publications from the past. How they, some of them have adapted, but I don't think a lot of them have adapted.

• 1:35Speaker 01

  The bigger problem with publications is they're creating informational content. Informational content is the stuff that's getting hit the hardest. The content that's more transactional, navigational, that's the stuff where you can still get a ton of traffic and more importantly, your revenue. Because if someone wants to talk about, let's say with CNET, they talk about computer-related stuff. Or How To Geek, I don't, I've never been to How To Geek, but I'm assuming it's computer-related stuff. If you go to CNET and you're reading an article on how to optimize the memory in your computer and make it more efficient. So it's your PC is more quicker. Cause memory is a big issue in AI, right? Not a lot of people are talking about that, but this is why companies like Dell are booming. Um, and with that article.

• 2:19Speaker 01

  Google in AI overviews or AI mode or Gemini or chat GPT or perplexor Claude. They're just pulling in what to look for and how to fix it. So that way you don't have to go to CNET. Reading that article on CNET produces them revenue, but it produces them revenue typically on the ad front. On the flip side, transactional keywords, navigational keywords, these are more bottom of funnel. They drive revenue. People still go to websites for it. Even if they don't go to your website and an AI agent just helps you finish the transaction on their website. That works too. At least you're capturing revenue. Who cares if you get the visitor? You just want the revenue. That's really the end game. And if you're like, no, I want the visitor. You're looking at the wrong metrics in your marketing and you have been for years.

• 3:03Speaker 01

  Yes, I get have more visitors in theory relate to more revenue, but even three years ago, five years ago, before all this stuff. Your number one metric that you should have been looking at is revenue, LTV profitability, even more than traffic or followers or likes. But it's the type of content that people are uh creating. And the informational ones are getting hit the hardest. And a lot of businesses haven't adapted from just shifting away to, we're just going to cover everything related to uh marketing. Like what is SEO? Well, that's cool and all, but there's already a million articles. Now, even if you don't get traffic, you do want to make sure you're covering your topic thoroughly on your website.

• 3:47Speaker 01

  Because you still want AI and Google to pull from you, even if you don't get the traffic. Cause you're building up the authority and the brand visibility and they're more likely to cite you. Um, but you just don't want to target all informational keywords like what is marketing? Well, that's been beaten to death and a million sites hypothetically have covered that.

• 4:05Speaker 00

  I think a lot of this just comes down to adaptability. At the end of the day, so we've talked about search everywhere optimization to death over these last few years. Um, and so. A lot of these like digital trends, for example, they just didn't adapt their business models and they were they're not quick enough, right? Versus the ones who have, um, let's use Andreessen Horowitz as an example, right? I was actually listening to a podcast where Ben, Ben Horowitz and Mark Andreessen and and we're talking about the new media model that they have where they get to control the narrative and they have they have multiple podcasts going on. They have like multiple YouTube channels. They've really kind of mastered the media game, not just for themselves, but also for their portfolio companies as well. But that's a good example of a company adapting because, you know, the reason they, part of the reason why they did it was because they were getting attacked all the time by like the Wall Street Journal or the New York Times.

• 4:50Speaker 00

  Uh, but now the internet, they've realized that the internet has changed and you want to be able to own the conversation. Um, and that's what they have done, right? And so, um, I think that that's that's the name of the game here, like, don't get me wrong, like Neil and I first started in SEO. And when we look at this, it's like, yeah, it sucks and but like our traffic has declined. I don't know if we've declined ninety seven percent, we definitely haven't. Um, but we've also adapted to kind of appear on on other channels. You want to know something interesting, Neil? So yesterday I was talking to Apple and I was like, okay, we're ready to pull the trigger on some of these Mac ultras, right? I was like, but we need the five twelve, five hundred twelve gigabyte unified RAM. And they're like, we don't have those anymore. I'm like, why don't you have them? He's like, because we're sold out. I'm like, oh, damn it.

• 5:31Speaker 00

  So it's like, you you do need, if you're going to run local models, you do need the five twelve at minimum. Um, you can't have two fifty six because like a model like a Kimi two point five, which is an open source model, runs on like two fifty, two forty or something like that. So you don't have much wiggle room there. So I'm like, okay, well, you know, my strategy now is like, okay, if I'm going to buy these Mac ultras, I was going to buy like one old one because it's a Mac, uh, it's on the M three Ultra chip. But the M five is about to come out. I thought it was going to come out this week. It didn't come out this week. It's going to come out in like two months or so. So my point of saying all this is that I think a lot of people are buying, um, you know, on prem, um, at least for themselves and run local models because Neil, I spent five grand just on my own anthropic API key in the last thirty days or so. And that stuff's going to keep adding up.

• 6:15Speaker 00

  I'm like, god damn it, if if, sorry, pardon my language, damn it, okay? If I released my open claw to my team and they're all using it like crazy, you're going to town. I need to have local models if I want to save the monies, so.

• 6:27Speaker 01

  Yeah. Dude, yeah, RAM is super important. People forget that. They all think like, you know, these GPUs or CPUs, I forgot whatever they call it. I think they're called GPUs. You know, like Nvidia chips, AMD chips and all that kind of stuff, the TPUs. But they forget that you also need tons and tons of RAM. And without that, it doesn't work.

• 6:50Speaker 00

  You know, the research I did on this yesterday, I, what, what, what was I using? Was I using Claude? Was I using OpenAI? I was using none of those. I was using Gemini AI mode for this. Okay. And I was like, okay, so give me a crawl, walk, run for this. Like, what do I need to do? So my crawl, walk, run, just if anyone's interested for my local on-prem setup, is I'm gonna have multiple Mac Studios, okay? And each one of those is like 10 to 15 grand, by the way. So you have a few of those, okay? One might be running the local models, or the, the open source models locally, but you might have some for clients. You might have some for like your team's agents and all that. But the other thing too is you actually will need an Nvidia, like one of those CUDA, chips as well, because if you want one of those, that's gonna decrease the, the time it takes to get these tokens out, right?

• 7:35Speaker 00

  So if your team starts complaining about, hey, it's taking too long to get answers. Well, that's when you start to need to include like one of these like 10 to 15 to $20,000 Nvidia CUDA chips. So my guess, Neil, is you're, you're probably be buying from Nvidia soon as well. So we'll see.

• 7:49Speaker 01

  We could be. I have no idea what the heck my team buys half the time. I just see the card bills and I stopped looking at them. Cause I used to go line item by line item. It's just so much stuff to go through line item by line item now.

• 8:02Speaker 00

  Neil cashes the line items for our podcast costs sooner than I do.

• 8:08Speaker 01

  I, I have a tendency to nitpick every expense and it's like.

• 8:12Speaker 00

  What is this? It shouldn't be this. It shouldn't be this. I'm just, and at the same time, I'm like, I don't know.

• 8:18Speaker 00

  So, so yeah, but the good news is HubSpot, which is our, our sponsor, those checks should start to come in soon at some point. So we will see. Yeah. By the way, Neil, did you see how Intercom, they talked about reinventing themselves into doing 400 million in ARR?

• 8:34Speaker 01

  I did not see that. I'm really curious about this. Do you wanna catch me?

• 8:37Speaker 00

  Okay, yeah. I'll, I'll share my screen that like, then I'll get your reaction first to this. Okay, so.

• 8:44Speaker 00

  I don't know how you pronounce the name. I don't know if it's like Owen. So he's the CEO of Intercom, okay? So originally, Intercom was like a like a help desk software, like a Zen Desk, or like a little chatbot you have in the bottom right of your screen. Neil, did I miss anything there?

• 9:02Speaker 01

  The original version of Intercom, think of a chat that pops up in your bottom right corner. And that chat was human powered, right? And it was mainly humans using it, although I think you could have some pre-filled tags and you could input a database.

• 9:16Speaker 00

  But those were never that great.

• 9:18Speaker 01

  It was never that great.

• 9:19Speaker 00

  Yeah, okay, so Intercom, I think they were maybe at some point in the last couple of years or so, they're maybe doing a hundred million in ARR. Now today, they're doing four hundred million, okay? So, um, so the CEO and founder, he left for actually two to three years or so, okay? And then it was only when ChatGPT came out where that was existential kind of code red moment for them. Um, and they started building Finn. Finn is actually their customer service agent. That I actually interacted with one yesterday where it actually answers and solves questions. It's basically your automated customer service team. Did I miss anything there?

• 9:51Speaker 01

  No, and it's really good. And to be clear, when Intercom made this pivot, instead of saying, hey, intercom.com, you can get all this stuff. I believe it was like fin.ai or they created a separate brand, separate product, so that way there wasn't confusion. To get more adoption in the market.

• 10:09Speaker 00

  Yep, so I'll kind of give you the TLDR on this post. I highly recommend you read it because it's it's got eight hundred seventy thousand views on on Twitter. But so Owen says, look, um, you know, I'm going to offer Intercom, the company I run, as a case study to help me explain how SaaS companies can be saved and share the things we did starting three years ago to find relevance in this new world. We found ourselves being referred to as the poster boys for success as a late-stage SaaS company in AI. And I'd like to first share some numbers and background context to help you understand why that is. So. Three years ago, our future was looking pretty bleak. I jumped back into the CEO role after a two-year hiatus while sick, while we were heading quickly towards negative growth. All SaaS growth rates fell as the ZERP days waned, zero interest rate policy, and ours fell extra hard due to unfocused business strategy that led to poor product-market fit.

• 10:54Speaker 00

  A month later, ChatGPT was announced, and we all begin our collective gradual and still ongoing journey to try to understand what that all means. I'll give you a simplified meme version of the story first. Step one, AI. Step two, a lot of question marks. Step three, profit. Because I'll be explaining how the how shortly after. But in summary, we launched our new service, Finn, in the summer of 2023. So now it's we're talking two and a half years or so, and our business began to violently, violently recover thereafter. Like, by the way, look at this chart. So Neil, I'll let you kind of explain this chart.

• 11:24Speaker 01

  It's too small for me to see.

• 11:26Speaker 00

  Oh, you can't see. Okay, so basically, what you have here is you can see in 2021. Okay, 2020 happens, right? So everything starts moving up. SaaS intercom growth rate is like a high of 37%, okay? But 2021, okay, it drops to 30, 31% growth rate, then 30%, then 24%, then 18%, then 12%. At 12%, when it's like it's coming down a cliff right now, that's when the CEO change happens and he comes back in, okay? So CEO, like, by the way, CEO change happens. It drops from intercom growth rate drops from 12%, 7%, 5%, 4%. When it's at 4% at their lowest, that's when they launch Finn. And right when that launch happens, 4%, 6%, 7%, 9%. Frictionless Growth Marketing hosted by Sonya Thompson

• 12:11Speaker 00

  is brought to you by the HubSpot Podcast Network, the audio destination for business professionals. Frictionless Growth Marketing hosted by Sonya Thompson shows how modern brands grow by removing the friction in their marketing and customer experience that pushes customers away every single day, often without leaders even realizing it. Each episode delivers practical insights, real brand examples, and conversations with marketing leaders, showing brands how to win with the new rules of growth, identity-driven marketing, cultural relevance, and frictionless customer experiences. Listen to Frictionless Growth Marketing wherever you get your podcasts. All the way back up, my camera's blocking the way, but it goes up to like 30.

• 12:47Speaker 01

  37%.

• 12:48Speaker 00

  37%. All the way back up to 37%. And they're crushing the average, like, you see the black line over here? That's the average SaaS growth rate. They are now well above that. And that's, that's the recovery story, right? So they're now 400 million ARR overall, with Finn about to pass 100 million. Go ahead.

• 13:05Speaker 01

  Yeah, but when Eric's showing thirty seven percent growth, their growth rate is higher with a higher base. To grow up thirty seven percent when you're doing four hundred million dollars, that means you're adding a hundred and forty eight million dollars in revenue that's recurring each and every single year. But here's the kicker, right? So if they're at four hundred right now and they grow thirty seven percent, next year they're at five hundred and forty eight hypothetically, assuming the growth rate stays flat, but their growth rate is actually increasing. So let's say if they just maintain thirty seven percent growth. It's five hundred and forty eight because they added a hundred and forty eight million. And then when you look at the five forty eight,

• 13:48Speaker 01

  thirty seven percent growth on that is two hundred and two. So it keeps compounding. So then there'll be over seven hundred million dollars and I bet you at that point they'll go public.

• 13:59Speaker 00

  Cool. So let's talk about how they did it, right? So then the main gist of his article here without reading the entire thing is this. At Intercom, we destroyed so many things, big and small. We ripped up our old values and created new disruptive ones, which we then hired and fired against to drive the behavior we needed in this new world. We rewrote our mission to focus on our agent goals and we changed our targets to Finn revenue. We switched up our board, removing mature and experienced leaders for startup founders. We moved our R and D focus to be nearly eighty percent on Finn, while that business was still a single digit percent of our revenue. We created a whole new brand and a separate site with a separate one million dollar dot AI domain. Ouch. And drove one hundred percent of our paid traffic to it. Every bit of marketing said Finn, Finn, Finn, not Intercom. And it was risky.

• 14:43Speaker 00

  Uh, big picture, brand and positioning, marketing to hopefully pay off later versus incremental performance, uh, marketing to definitely pay off, um, that quarter. So again, while Finn was a single digit percent of our revenue, all because we needed a market to give us a new look for a new category. We aggressively promoted and compensated our AI team above others. We grew it from six people to sixty people in three years with real PhD level AI scientists and researchers. We built out our, our own AI and trained our foundational models more than that soon. So the main thing he's calling out here is this. We did not hold back on the creative destruction. That really is the only idea I'm trying to sell here. We deserted our past to make way for our future. So you could, what I just mentioned earlier, he's like, you know, we, we got rid of our entire board for screw experience. We bet all in on the single digit thing that wasn't even that the main thing.

• 15:28Speaker 00

  Um, and they, they just ripped everything apart and they were willing to destroy everything for the future. And I think not many people are willing to do that. And so he's saying, look, this shit is insanely hard. The shift from on prem to cloud wreck many companies, but the change was not nearly as fast. And while it proved quite difficult, right? I think this shift is, is, um, he's like, look, it's hard to do this, this shift, but little old left for dead fifteen year old intercom did it. You can do it too. You have the brand, the customer base, the cash flows and the access to debt. So what do you think?

• 15:57Speaker 01

  Dude, I think spot on the, the big thing here that I think a lot of people need to take away, at least from marketing and a business lesson, is sometimes people try to adapt their old products or create old products. But there's brands and people perceive a brand for a specific thing. It's really hard to reinvent a brand and say, hey, we used to do this, but now we're this. Sometimes people just have a pigeonhole on what you do. So sometimes the best approach to move forward is to not try to convince everyone your brands change. It's to launch a new brand that's owned by the main parent company that focuses on this new thing and you get easier adoption in many cases. And you start seeing a quicker uptick. That's at least what we've noticed.

• 16:43Speaker 01

  Because it's hard to please the old customers and the new customers at the same time. So in many cases, it's easier to launch a new brand.

• 16:49Speaker 00

  So you know what, what I want to call out here is, um, I'll call out this last line that he has here. So it says, look, I want you to survive it and win. So this, this change being this AI change, right? All it will take is destroying everything you love. So here's what's happening right now, at least for me, the experience shares from my side. Like supplement companies are growing very quickly. Crypto companies are growing very quickly. They're coming and they're seeing the content that I'm putting out and they're saying, hey, we want to do what you do. Um, and then internally, the struggle that we're having right now is that you can't just be button pushers for paid media. You can't just say, I'm going to do keyword research and do traditional SEO. Doesn't work anymore, right? So what we're doing from a creative destruction standpoint is,

• 17:29Speaker 00

  My CTO leveraging OpenClaw last week, we are actually building something that's actually promoting a fin-like product, right? For another company, but we're building all the scaffolding for the cold outbound, the voice, the voice phone calls, right? With voice, uh, voice, um, voicemails as well from, from the, from the AI. And also we're building the agents to kind of fulfill everything from end to end, right? Now, everything from like the software stuff, we're just like, let's just give that all away. Um, or at least do freemium for all of that and then have those be lead managers that lead up to kind of the main things over here. And then the, the dashboard that you saw earlier, that's all going to plug in together, right? Um, so that's how we see all this doing, but without this, this coming together, but without an open claw and a claw code, we wouldn't be able to do this as quickly.

• 18:14Speaker 00

  Um, and so I think it's, it's easier for us because we're, we're, we're smaller to, to be able to kind of move quickly like this. But I think if you're a larger company like a, like a intercom, for example, it's more difficult. And he had to make that bet to just say, you know, F it all in and it paid off for him at least so far.

• 18:31Speaker 01

  Yeah. And when you said that internally, you guys are working on creative destruction, maybe I heard that incorrectly.

• 18:37Speaker 00

  It's correct.

• 18:38Speaker 01

  What do you mean by that?

• 18:39Speaker 00

  It's actually, he used this term over here. We did not hold back on the creative destruction. So, um, this actually came from the, the Cole brothers, right? So, um, I think it's the Cole brothers.

• 18:48Speaker 01

  The Cole brothers? Is that the Stripe? No, that's not Stripe.

• 18:50Speaker 00

  No, that's the Colson brothers. Um, you know, the, uh, oh, Coke, Coke, Coke, Coke brothers. K O C H. Yeah. So, um, so, you know, they're, they're a conglomerate, just for those that don't know. Um, they have a bunch of privately held companies. I think maybe the largest, largest privately held conglomerate, maybe. No?

• 19:09Speaker 01

  No. Cargill's larger, 160 billion in revenue a year.

• 19:12Speaker 00

  Um, so anyway, but they're pretty big, right?

• 19:15Speaker 01

  Passed away. Although his wife just bought, I think it was last year, she bought part of the New York Giants. She bought, I think 10% of the Brooklyn Nets. She's been making some moves.

• 19:24Speaker 00

  I love how you're on top of it. Anyway, so the Koch brothers, I read one of their books, they do talk a lot about creative destruction. You have to be willing to destroy what you currently have for what's new. And that's what Intercom did. The way I see it is that whether we like it, at least for us, whether we like it or not, a lot of the our original services in its current form is being destroyed, right? So we have to evolve into the the new world. And, you know, it's it's it's hard for people to adapt, but I'll tell you, the ones that are, it's like, like, e- every week, I'm almost, I stole this question from one of our mutual friends. It's like, hey, Neil, what have you automated this week? What have you automated this week? It's like, OK, the ad production now is going from like, you know, maybe two to four hundred, because people are now asking about more AI creatives, right?

• 20:08Speaker 00

  And they they want like four to five thousand a month or so, or ten thousand a month or so. And, you know, the team has gradually been building up to that, right? So that's an example of, hey, you can't do creative like necessarily the old way. You still need a lot of the process, but you need to learn this new way. But major creative destruction is what some of the products that we're building. That's major creative destruction. Because the way we see the future is like, we need to hire probably a lot of strong marketing strategists too. So that doesn't go anywhere, because that's what we do. That's what you guys do too. But we need to hire a lot of AI automation engineers to to leverage the open clause in the cloud codes and then partner up with the marketing strategist. And then you also need to have a leadership layer. So at least for me, that's very different than what I was doing, you know, even two years ago.

• 20:48Speaker 01

  Yeah. Dude, it it's just so funny how I'm now starting to see big corporations finally encourage AI, at least in marketing. And what I mean encourage it, there's been a shift. If I had to say last year, even towards the end of last year, not, which isn't that much time ago, right? You're talking about like six months ago. I was seeing a a shift in organizations of all sizes. Hey, you should get way more done with AI. Why don't, why aren't you able to fire a lot of people? Why do you need all these people? Can't AI just do a ton of this stuff for you?

• 21:31Speaker 01

  That was a narrative, call it, in Q four of twenty twenty five that I was seeing from a lot of companies. And that was the internal pressure they were getting from their CFOs, CEOs, etc.

• 21:43Speaker 01

  Now internally, I'm actually seeing organizations say something very different. They're not pushing the narrative, why do you need all these people? Shouldn't you get more done with AI? I see the narrative of, hey, everyone's moving so fast because of AI. What resources do you need to make sure that we don't get left behind? And we can stay ahead of our competitors? Who do you need to hire? What training do we need? I'm not really hearing too much of the narrative of firing. I hear more of the narrative is there's so much going on and the executives understand this. There's so many LLMs, so much technology, so many platforms, it's hard to keep up. There's like, how do we stay ahead? Who do we need to hire?

• 22:28Speaker 01

  Cause many of these large organizations have so much cash on their balance sheets, right? They're not focused right now as much on cost cutting. Their focus, and I think this is actually more important. How do we stay ahead and make sure we don't lose? And if that means you're paying extra in the short run or spending more, I think that is an investment that's worth it. But the key right now, if you want to win in this AI race, is not about cutting costs. It's about how can you make sure you're adapting to this technology? And implementing and integrating it and using it correctly faster than the competition.

• 23:03Speaker 00

  All right, guys. Well, I hope you enjoyed this. I need to go run to a webinar. Neil's gonna go to bed, so I hope you all enjoyed it and we'll catch you next time.