Speaker 1There is zero chance that this bubble doesn't pop. We're so far over our skis in terms of valuations. It's crazy. That's Jeffrey Moore. His book, Crossing the Chasm, has sold over a million copies, and he's advised the CEOs of Microsoft, Salesforce, and Cisco. When new technology rewrites a market, he's the strategist companies call to figure out who wins and who gets left behind.
Speaker 2It's not that SaaS would be taken out of existence, but it could easily get marginalized and its economic model could deteriorate. In today's episode, Jeffrey shares which
Speaker 1software companies are most at risk, whether he'd buy the anthropic IPO at $2 trillion, and whether AI has to eat
Speaker 2your headcount to pay for itself. Where do I think the big returns from AI are coming from? I think they're coming from.
Speaker 1What makes Jeffrey so valuable is that he doesn't stop at whether AI is a bubble. He gets into what AI disruption means for your company, the frameworks to think it through, and what to actually do about it.
Speaker 3Welcome to Top Line. Hey, everybody. It's Sam Jacobs. Welcome to Top Line. As always, I'm joined by my amazing co-hosts, AJ Bruno, the CEO of Quotapath, and Asad Zaman, the CEO of Sales Talent Agency. Another great show. We've got a very, very special guest, Jeff Moore, Jeffrey Moore, to those not in the know, the author of Crossing the Chasm, among many, many other books. Jeff's been. How many books have you written, Jeff?
Speaker 2Well, seven business books and then one on philosophy.
Speaker 3Seven business books, one philosophy books. Crossing the Chasm, as many know, is canonical in the startup land for thinking about how. Brands and products can reach the larger population of adopters and consumers rather than just the folks that are early adopters. And it's shaped a lot of how Silicon Valley has thought about marketing and sales and go-to-market over the last 20, 30 years. You've written many other books. You're also an investor. You're an advisor. You were a speaker. You gave a keynote talk at GTM Pavilion's GTM conference three weeks ago. And we're really, really excited to have you as a guest on Top Line for the very first time. Well, thank you. Good to be here. Let's dive in with our first question. But again, I'm sure that Asit and AJ will have plenty to talk about. But we want to talk about this concept of Crossing the Chasm. Jeff, you've said agentic AI is clearly in the early market but hasn't crossed the chasm. But 2026 is the year it has to. We're in the last quarter. It is October, the middle of October, almost the end of October when we're recording this. So has agentic AI crossed the chasm? What does it need to do?
Speaker 2Yeah. So crossing the chasm in that sense means at least in one industry for one class of use cases, it's become the de facto standard. It is now the normal thing that any pragmatic person would do. I think cloud code is clearly an example of crossing the chasm. What you would like to say then is the idea behind crossing the chasm was you'd start with one use case and one vertical, but then it would sort of migrate out. So what's happened with cloud code is software development, particularly in the software industry, it's become the de facto standard. It's kind of what people do. But I'm not sure it's having what we call the bowling pin effect that we'd like, which is that it would knock over other use cases and other use cases. So if I said, well, what about reengineering normal workflows, be they in sales or marketing or customer service or whatever? And I think we're on the verge of crossing the chasm, but I don't know that we've nailed it yet. For example, if you called up a customer contact center and you never got another IVR message, never, that it was always an agent answering, you say, okay, then it would cross the chasm in contact centers. We're not quite there. The voice stuff isn't quite right. It's getting there. It's very close, but it's still teetering on the edge. So I think we may have some work to do at the beginning of 2027 to get this thing really institutionalized as agents. By the way, co-pilots are everywhere. Co-pilots are like inside the tornado, but autonomous agents, I think, are still a little bit struggling on the other side of the chasm.
Speaker 3Setting the stage, let's take a temperature check before we go into more detail. How optimistic are you? You know, we talk a lot on this show about AI. I think the three of us are very, I think we are AI bulls. But I also think we're concerned, sort of exactly to the point about crossing the chasm, we're concerned about the rate of adoption relative to valuation and I think relative to financing. This build-out is not financed by, you know, cash flows from, you know, from the hyperscalers. It's financed by debt. And so there's a timeline that's ticking that relates to, to your point, how quickly it becomes adopted, how quickly it shows up in GDP growth, which is a way of saying productivity improvements across, you know, American enterprise and global enterprise. So there's optimism, but it's guarded optimism. Where do you sit on that spectrum from pessimist to optimist?
Speaker 2I mean, look, every time we've done one of these things, there's been a bubble and the bubble does pop. It's crazy. And it's particularly the frontier models with these amazing, you know, these incredible trillion dollar build-out kind of things. Here's what's going to happen. The winners are actually going to win. In other words, for the people that actually become the number one or number two, those investors will say, well, it was a scary moment, but we got through it and we're really in good shape. But three, four, five, six, seven, eight, nine, ten? No. So I think the valuation issue is something that there's a reckoning that has to come. And I don't know when it's going to come, but the math doesn't work.
Speaker 4The counterargument you hear from people right now, and I worry about this spend a lot these days, because I think for one of my fears is that for this to not pop, revenue growth has to keep up with the spend. That's one of the things that has to happen. And for revenue growth to keep up, you need to keep unlocking more and more use cases. So we have code, we have chat. But I haven't seen a lot more in the last few months in terms of real use cases, real ROI-based use cases that are getting unlocked. And so it does give me some anxiety. The counterargument you hear, and I'm curious to get your thoughts, is that it can be that it's not one big pop of the bubble, but it's more so multiple little corrections over a long period of time. Every few quarters, you know, 5% drawdown. And so that's this other side of the argument, what are your thoughts on that? Like, how unlikely is it that that is enough? And at some point, you know, maybe it's the macroeconomic conditions, but it does pop and then we do get like a desert period.
Speaker 2Investors are herd animals. You know, once the stampede starts, I'm sorry, you're not going to stop the stampede. It's not going to stop. I mean, 2008, do we remember 2008? Do you remember 1987? I mean, I remember when the Dow went down 22%. I think it was in 1987 that day. Okay, so look. By the way, is it recoverable? Let's hope so and presume so. But you look at the circular financing that's going on. And the way, I mean, this, you know, the co-sponsorship and whatever, it reminds me a lot of the dot-com thing and how the dot-com bubble developed. Having said all that, do I think AI is going to change the world for the better? Yes, I do. Do I think it's going to improve productivity? Yes, I do. But by the way, the way it's going to improve productivity, it's going to commoditize. So they get cheaper and cheaper and cheaper, which will let it become more and more. It's going to become more and more ubiquitous. But again, that works for the market leader and the second player. But after that, then there's not enough margin to pay for the build-out. And by the way, the other thing that's happening is the digital transformation we went through the last 20 years has created huge cash flows we've never seen before. So there's a huge amount of money sitting on the sidelines wanting to put itself to work. And so people are saying, well, we can put more money into here. And that's kind of what I think has happened.
Speaker 4And if it bursts, is it necessary? So in the last couple of moments where things burst, we had a long period where things were quite quiet then, right? Like after 99, it took a while for the market to come back. After 2008, not as much time, but it took a minute. Is it necessary that when it bursts, it's going to be a year or two of just like silence and you're in a desert? Or it can burst, but then it can recover quickly? Is there any reason to believe that?
Speaker 2Yeah, I mean, I think that people, people that have implemented AI are not going to be that effective. I really don't. I mean, in other words, if you have an AI implementation and you've been doing it on top of a system of record, or you've been using one of the frontier models, and what, they're going to go bankrupt? I don't think that's the issue. I think the financial markets, this reminds me more of 2008. In 2008, it was all around housing credit. It was just, at some point, it was just really, really embarrassingly poor decision-making. I think we're doing a version of embarrassingly poor financial decision-making.
Speaker 1Jeff, typically when you see these type of bubble bursts and we go through this recession, you'll see companies like Google or Amazon that were created during these times of durability pop out, but we were unfamiliar with them. It feels like we have a lot of too-big-to-fail type of companies that we're talking about. And does that give, is there an opportunity to give way to companies that are being founded today in this? I think the answer is going to be an obvious yes, but give, those companies to cross the chasm, so to speak, during a time of recession. And do you have any examples of when companies in this period were able to do that?
Speaker 2Yeah, in fact, I wrote an article about it called During a Downturn, Provoke Your Customer. It was in the Harvard Business Review. But the idea behind that was, look, in a downturn, people only want to spend money on things that are broken. They just want to fix whatever's broken. And the truth is, those markets tend not to be very large. So the established big players, players can't really serve those markets. I mean, it's not that they couldn't, but they don't get enough bang for their buck. And so they're looking elsewhere. So startups can actually do quite well in a recession because they can afford to narrow their focus down to something that's big enough to matter for them, but it's small enough for them to lead. And by the way, it's a good fit with what we would call their crown jewels. So in other words, that was the formula we used at that time, big enough to matter. If you're a $10 million AR company, $100 million market is huge, right? It's 10 times bigger than you are. So big enough to matter. But of course, to Salesforce or IBM or Cisco or Microsoft or Google, it's like, it's grounding it. So basically, big enough to matter, small enough to lead. And it's got to be niche-y enough and gnarly enough that people don't really want to do the last mile. So that's why you get to lead it as a startup. And then it's got to be a good fit with your technology. You've got to be able to really break the back of the problem. But that's what next-generation technologies can do.
Speaker 3Do you talk about that use of the word provoke? That's very interesting. Is that intentional when you talk about provoking your customer?
Speaker 2Yeah, we actually called it provocation-based selling. I think challenger selling and provocation-based selling are almost complete overlaps. But the idea was, as opposed to trying to be, well, what keeps you up at night? What would you like to help with? Blah, blah, blah. It was more like, you know, we've been spending time with. The way you do provocations is, we've been spending a lot of time with your industry. And, you know, there's a part of the operating model here that really just seems to cause people grief all the time. Here's what we think the problem is, and here's what we think the industry's stuck. Are you having that problem? And if they are, then you say, okay, well, then let me tell you about how we think we can take that off the table. The other thing that can happen is you get to that point, they say, well, actually, no, we're not. You go, wrong one. But then what they will say inevitably is, let me tell you what we are struggling. So that's why I think either way, it's kind of a good way to have the conversation.
Speaker 3We had a great conversation at GTM a couple of weeks ago, and you walked through the zone to win framework. And I thought that was really interesting. So walk us through that framework, that two by two matrix, so that people understand what it is. But also because, to the point we were talking about before we started recording, there's so many AI companies that are in, they think they're in the transformation zone, you know, they think that they are, and they might even take their ARR or their revenue for one day and multiply it by 300 and say, you know, based on today's growth rates, we're going to wear a $1 billion ARR business. But they're not in the transformation zone, they're in the innovation zone. And then there's the other part of it, which is like the performance and like, how do we make sure that we are part of the ongoing operating machinery of the business? Walk us through the framework, because not everybody's read the book yet. And then talk about how people can migrate between the two. And then talk about zones in an effective way.
Speaker 2By the way, I'm looking at a two by two matrix right now with four pictures in it. Are the people seeing this podcast also looking at a two by two, are they going to keep the four windows together? Yes. Okay, perfect, perfect. Because we're each going to be in a zone. We're going to play zone to win like Hollywood squares. Though Sam is in the performance zone. The performance zone is the zone that delivers the number this quarter, makes this year, pays for all the other zones, takes credit for everything. And by the way, delivers value to the world. This is the zone that engages with the world and fulfills our mission. So in that sense, the other three zones, Saad, AJ, and myself are in service descent. Okay. Agreed.
Speaker 3This is all accurate so far.
Speaker 2It's amazing. It's amazing. This is like
Speaker 4Sam journaling every morning.
Speaker 1But also I can drag my window across and just replace Sam right now for the record.
Speaker 2Okay. So it's Saad's in the productivity zone. So the productivity zone is all the shared space. So it's all the shared space. So it's all the services that make Sam look like a hero. So it's everything behind the service. It's marketing, it's customer success, it's professional services, it's finance, it's HR, it's IT, it's cybersecurity, it's purchasing, it's legal, it's all that kind of stuff. And typically, there are more people in the productivity zone than in the performance zone because that's what it takes to scale globally. And those two zones are 90 plus percent, 95% of the headcount in any global established enterprise. And all the annual budgeting is organized around those two zones, which is fine until a disruptive innovation comes in. The disruptors are Jeffrey and AJ. So Jeffrey is in the incubation zone. He's the one who's saying, no, no, no, you guys, you're doing it all. You're all old school. It's a whole new way of doing things. You guys are so clueless. Let me explain to you how incredibly smart I am and how incredibly amazing this technology is. But the incubation zone is basically designed to be a counter to the other two zones. It's supposed to be a challenge. It's supposed to disrupt innovation. It's supposed to disrupt them. And so as a result, you can't manage it in either of the other two zones because it's just an irritant. So the incubation zone and the way we used to manage it originally like 40 years ago was we had corporate labs. That was what the incubation zone was. It didn't work. Corporate labs could make the technology, but they couldn't find product market fit because they had no way to get to market. So basically, we've now realized, no, it needs to be organized more like a Y Combinator startup garage kind of thing. You need to fund independent operating units that look like entrepreneurial leaders. You need to give them sales resources and their own sales resources, their own marketing resources. You need to let them kind of run their own game. Or you acquire them. You kind of tuck in technology acquisitions to populate the incubation zone. And this is the fail fast zone. And it's all that lean startup kind of stuff, which is great for that zone. But if Sam fails fast, we're going to fire him. And if Assad fails fast, we might have to put him in jail. So, I mean, fast failure is only for the incubation zone, but there it's a key idea. And then what Sam's going to ask me about in a minute is how the hell do you get out of this thing? Because that's what we found in AI for the last couple of years. People were piloting like crazy and having great success in my zone, but they couldn't get out of my zone. They couldn't get it to scale anyplace else. So how do you do that? Well, there's actually a number of paths, but the zone to win talks about the first path is the one that AJ's in, which we call the transformation zone. That's where you go all in really big, really big on a truly disruptive thing. And we've been writing about this for 20 years. This is Steve Jobs going from Macintoshes to music. Really? And from music to phones. You're kidding. Visa is going from selling books to retail? Going to cloud computing? What are you talking about? These are these amazing transformational things where you really move massive amounts of assets out of an established category into a whole new category. You're taking significant category risk, enterprise risk, but they're the game-changing plays. And we have SpaceX and Tesla and all those kinds of things to play with. So that's the transformation zone. And the time when a normal enterprise would do it is not voluntarily, because only the Bezoses and the Musks and the Gateses and the Ellisons can do that. But for most of us, we could end up in the transformation zone if we're under existential threat. I would say that Gerstner coming into IBM in 1992, IBM was under existential threat. He had to change their model. Dramatically. And he went from essentially a hardware-centric to a software and services-centric business model. And it was extraordinarily challenging. But frankly, his back was against the wall. And so when your back's against the wall, and we watch Walmart dealing with Amazon, their back gets against the wall. They're going to take transformational risk. But most people, most of the time, are not under existential threat. And so then the question is, well, how now would you get the pilot out of incubation? And the path we've now, I think, settled on says, first of all, we're going to have to get the pilot out of incubation. So let Jeffrey get to product market fit to say, okay, this looks like this is something we really ought to do. Then Assad, you get to be customer zero. We're going to take this thing and we're going to do it inside our company. We're going to re-engineer our internal productivity zone using these technologies, using these agents or whatever. And we're going to, first of all, we're going to get the benefit of that, both in, not just in terms of cost reduction, which frankly doesn't really matter, but service improvement, velocity improvement, latency reduction, quality improvement, the really important stuff. We're going to make that happen. And then once we're comfortable with that, then we'll go into Sam's area, but we'll go into it use case by use case. Rather than try to say, we're going to just re-engineer the entire offering, we'll go either vertical market by vertical or use case by use case. And within any given use case, we're going to say to Sam, okay, Sam, you got to make, you got to develop the market here. It's going to be hard to make quota inside this envelope because you're creating a market. You've got to make a market. You're creating budget. You're developing new relationships and you have to make your quota. So you can't do it everywhere, but, but pick some spots. And we expect, you know, X percent of your portfolio to be on the new, on the new technology doing as best it can in terms of sales velocity and whatever. And then the rest of the thing, you're going to sell it yourself. And that actually, that's a kind of a use case management approach. And those are how the four zones I think are going to play out.
Speaker 1Jeff, the thing that you said that was interesting is that back against the wall, existential threats that doesn't happen very frequently, but I would argue that every company that was founded prior to 2023, at least in software is under that existential threat today. And so the question is like, we're all kind of thinking about that transformational zone right now. What is the likelihood that the majority of us make it out of this? Or what's the likelihood that the majority of us are going to fail?
Speaker 2Well, I think I can't, I, if, if you, if you wait long enough, your only choice is transformation and transformations are very high risk. This zone to win starts with 54 iconic companies that were market leaders that don't exist. So, I mean, the track record was not a happy one. Okay. Now, having said that, I think what you, let's take, let's suppose you're a SaaS company right now. You say, well, okay, that was the SaaS apocalypse. Okay. The existential threat. AI is going to put you out of business. Well, I think we've kind of figured out that probably is not true. That AI needs systems of record and then they need the underlying infrastructure. So it's not that SaaS would be taken out of existence, but it could easily get marginalized and its economic model could deteriorate, et cetera, et cetera. So I think what we're seeing right now with SaaS companies is, can we begin to onboard a consumption-oriented model, something that's more oriented toward the new economy, without abandoning our seats model entirely? And I think that's, I think that is a classic thing of where you would go to Assad first and say, ask Assad, do we have the internal systems to even bill for consumption? Do we know how to manage our token spend? Can we do that? How can we pass that along to our customers? And then we go to Sam and say, okay, Sam, we've now got a new set of authors. We're going to give you flex credits. They can use them for seats. They can use them for consumption. They can consume this or they can consume that. You know, it's going to settle down eventually. And eventually there'll be some sort of stabilized pricing where we can build the industry out with some confidence. Right now, we don't know what the pricing model really ought to be. And so, and until frankly, AI commoditizes a great deal more, I don't think it will settle into a consumption model, but I think eventually it will just the way it does. I mean, if you look at AWS and Azure and Google Cloud, Bill, I think we'll have some version of that eventually.
Speaker 4I remember reading Zone to Win. It was one of my favorite books. And I remember you had spoken at the time about how, you know, different companies try to have these second and third acts. And it really just, the gist comes down to how do they treat these acts? You know, I think you gave examples of companies that were successful at this, put the new teams with different key cards and different offices. I remember that really stuck out to me. And because you would hear a lot of people say, well, I don't know. You know, big companies trying to be innovative and doing it the other way and it wouldn't work. And I've been thinking a lot about actually that book as this shift has been happening with AI. And it seems like we're all downstream of model improvements to some extent. The model keeps improving. We learn how it improves and it has new capabilities. It's jagged, but there's these new like things that pop out that it can do for us. And that has huge implications internally in terms of how we're going to do it. How we operate our businesses and in terms of what we do for the market, what our product can do, what we can do for our customers. And so it feels like every three to six months, there's a pretty significant shift. And so I wonder if the companies that have been formed right now, instead of having this like bifurcated way of thinking about the main act and the second act, you almost have to develop a different sort of like capability of always being able to do the same thing. And so I wonder if that's something that you've been able to evolve at a, because the pace is changed so much that it has to be more ingrained rather than like, we'll separate it and then we'll bring it in. It's like, this is just going to be part of our business and part of how we operate. I wonder if like that is necessary now or the pace of change around us and its implications on us doesn't really impact that.
Speaker 2I look, I think about starting a company, whether you're starting an AI first company or whatever is, but do that when you're starting out, you have no place else to put your money. I mean, you, there's no competing, nobody, you don't have a legacy business that's competing for budget, right? And you, and by the way, you're not funding your, your, your enterprise out of cashflow. You're funding it out of venture capital of some, some form of venture, venture investment. So, so there's no conflict, which is why a Sierra, you know, it looks like they can run rings around an established enterprise because Brett, you know, has left Salesforce. He's now the CEO at Sierra. And it's like, yeah, I, I, I could do outcome-based pricing. I can, I can do anything I want because basically I have no, I don't have the money. I have no, I have nobody competing for the alternative. The problem with that is you also have no customers and no footprint. So, so it's like, I mean, understand that the established enterprise has global scale and hasn't installed base. So I, I think the game has got to be, no, we're going to play this. And I think the way you play it is you played zone defense rather than zone offense. So if you're a disruptor, you play zone offense. You are looking for the, the latest model comes out. You want to figure out what can I do with this that I couldn't do before is I want to be even more disruptive. If I'm an established enterprise, I'm playing zone defense, which is first of all, do I really have to adopt this one? Second of all, where in my install base is the trapped value that we want to try to use technology to release? Will this help me be a better trap value releaser for my customers? If not, I'm not going to do it. Because then where you're working backward from the ROI instead of forward from the need to innovate, we don't need to innovate. We need to deliver ROI. As long as our, as our customers are getting ROI, they're not going to defect. They just aren't. But I think when they, when they start seeing their ROI decline, like it come out and when we're not showing new ROI, okay, now, now it's like, okay, at some point, doesn't that
Speaker 4feel like in a market that is, there's so much noise in this market. You probably, if you take Salesforce on it, as an example, you're probably traded company. There's X, there's all these commentators and they have basically made it so that they're going to be like, Oh, we're not going to do this. We're not going to do this. We're going to do this. have just been pushed to the side and are trading at less than what, like a fraction of Anthropic, whereas you have almost the same amount of revenue, if not a little bit more now. And it feels like if you're them, you feel this need to try to keep up with the Joneses, right? Like, I don't wanna feel like IBM. I wanna feel like I'm at the edge. I'm big and I'm at the edge. And so if there is a new model capability, look at how we've implemented this. So there's like this marketing communications perception issue that drives some of this beyond the practical. This is how we should think about value in our business.
Speaker 2So I think you do have to control your narratives. And if you watch the last earnings call that Mark was on, he really called on the saspocalypse. And I think actually, if you looked at their stock price after that, the market said, you know what? Yeah, the market loved that call. So he kind of went, but having said that, that only gets him back to sort of like medium. That doesn't put him in the anthropic link at all. What is interesting, though, is that, and I spent a lot of time with Salesforce, so I don't think I'm talking out of turn, but this notion about headless. So all of a sudden you say, well, what do you mean headless? Well, you can use our underlying data clouds and data infrastructure with your own applications. You don't have to come through our applications. And all of a sudden you start beginning to use the word platform instead of product. If you can make the transition from product, a ubiquitous product like the IBM PC, to a platform like Microsoft OS or like the Intel x86 architecture, you've changed the world. And SAP really wanted to do that, by the way, in the 90s. I remember working with Shai Agassi. And SAP had Ubiquity. It was the ubiquitous ERP program. And they had ABAP. And they thought, well, if we could just get everybody to build their applications on top of SAP, we could be a platform, which was true. But it was just architecturally impossible. It wasn't going to work.
Speaker 1Joe, just a question on that, but it sounds like you're saying that Salesforce isn't a platform, or I'm mishearing that, because I've always thought of Salesforce as a platform people build on.
Speaker 2Yeah. Salesforce has been a platform for the Salesforce ecosystem. But if you said it was headless, you could say, no, no, no. You could put ERP stuff. You could put cyber. You could put anything on top. This is the MuleSoft informatic data wrangling architecture underneath. You know, because Databricks is out there and these other guys, well, come on, we can play that same game too. And it's a system of record. And I think the thing that the AI community really undervalued was how foundational a system of record really is. You can't run the world economy without systems of record. And yes, they're kind of stodgy and intentionally stodgy, but it's the source of the data. And we all know that the data feeds the AI, so come on. Having said that, they can't move very fast. And the systems of record, they do not open themselves up. Like you have in vertical industries, like you have Epic. So Epic is like as closed a data system as you can find in American healthcare. Okay? You know, that is successful until, frankly, it goes away. Because at some point, the industry is just going to say no.
Speaker 5Most customer churn isn't a product problem. It's an education problem. Customers who never get to value, never renew. This episode is brought to you by Thinkific, the leading enterprise learning platform where courses and community come together to power business growth. Thinkific helps customer success and revenue teams build learning programs that lower churn, accelerate product adoption, and scale without adding headcount. Automated onboarding paths, role-based access for complex organizations, AI tools that dramatically cut admin time, your learning platform should drive growth, not just track it. See what's possible. Book a demo at thinkific.com. That's T-H-I-N-K-I-F-I-C dot com.
Speaker 3This episode of Top Line is brought to you by Pavilion. I want to tell you about Pavilion Gold. It's so exciting. Pavilion Gold is an invitation-only community for senior go-to-market leaders at many of the world's leading B2B events. companies, CROs, CMOs, presidents, and executives navigating scale growth and transformation in real time. Pavilion Gold's about a year and a half old, but our first group of 24 Pavilion Gold members came up for renewal on October 1st. I want to ask Asit, how many of those 24 people renewed their Pavilion Gold membership? Which, by the way, is not insignificant. It is an investment. I think I did my talk right
Speaker 4before, and so I would imagine close to 100% after this on me on stage. Your talk? I was also there, Asit. Your talk. AJ was around. AJ was around. I would imagine, just from the conversations I've actually had with people that are in it, I would imagine this has one of the highest renewal rates you've ever seen in your life.
Speaker 3100%. 24. It is 100%. 24 out of 24 people. 100%. 100%. And then there's a few people that came up for renewal through October and into November. Again, so far, the renewal rate is 100%. That's crazy. And in fact, my favorite, so here's one other thing. Matt Thompson was the president of this cyber company called SoCure, and he got recruited away to be CEO of a new company called Transmit Security. We got an email from him that says, may every day be day one. That's the subject. On his first day as CEO of this new company, we don't normally let CEOs join, but we're not going to kick you out if you were previously an operator and then you get promoted to CEO. And so the first thing he did, literally, this is not a joke, the first thing he did was re-sign up for Pavilion Gold, and he gave his old membership to his CMO at the departing company. Sam, let's talk about
Speaker 1this for a second, because this is no BS. I think Pavilion Gold is awesome, like legitimately the best conversations I've ever had as a part of a community member at Pavilion have happened at Pavilion Gold events. The dinner in New York was amazing. When are you going to let CEOs in, and why can't we be in? Seriously. I'm honestly frustrated about this.
Speaker 3AJ, you are well-served, sir. You are well-served. I know. You got Pablo. I know. You got Insight. I know. You got HubBot Ventures. You're being wined. You're being dined. I'm asking for the world, though. I'm asking for the world's sake. You were an early investor in SpaceX. You're doing just fine. You don't need us. You know who needs us? Operators. Operators need us. So the answer is, we tried a CEO group, and sadly, besides AJ and Asad, who I love, besides those two people, I can't say that it was the top brass of CEOs that we were getting. We were getting a lot of RevOps consulting firms trying to sell into the go-to-market community, and that's because, you know, we're not pavilions for operators. Yeah, we're a go-to-market
Speaker 4community, so I really liked that in the event that I attended, I even felt for myself that we were able to be far more open because you trust, it was a smaller room of people that had come there. The bar is higher. Yeah, the bar is much higher. Trusting that we're not going to let this information out. We're going to use it for ourselves, but we're not just going to like, you know, quote this person on social media, and so it actually allows real conversations, and that's very rare. It's very hard to have real conversations.
Speaker 3It's our highest-performing program. It's really, really powerful. If you're interested, please reach out to me. You can email me, sam at joinpavilion.com. You have to report to the CEO, and you have to be either super fast-growing AI or at 100 million plus, but otherwise, we'd love to talk to you. Thanks, everybody. Where do you think the moats are? You know, that's another huge topic of conversation is where are the moats. I was just talking in a, just prior to this, I was talking to Dan Lee, the CEO of this company, Nooks. He's talking about the pace of feature development and software development. It feels like companies can build new products and features more quickly than ever. Software is easier and cheaper to build. Now they've got software factories where agents are building the software. Agents are QAing the software. So where is the long-term sustainable competitive advantage that companies should be investing in if every feature, every digital action is ubiquitous and commoditized?
Speaker 2Yeah, I think, so I'm a big fan of this discipline I'm calling trapped value analysis. So trapped value analysis says, look at an industry's operating model and ask yourself, why can't this industry be twice as productive as it is today? I mean, they'll be a very, very good. I mean, it's not like people are lazy, right? There'll be, there'll be gating items fundamental to the structure of the operating model.
Speaker 3What's a, give us a great example of trapped value analysis. Like what's the number one example in your head?
Speaker 2Well, the one that always came to mind was the taxi business. Where was the trapped value in the taxi business? I mean, cause if you've ever stood in a street corner and you were all old enough to, I think, have done that. And then all of a sudden trying to get a flag down a cab or trying to get a cab to come to your home. And then all of a sudden Travis comes along and does Uber. And you realize. Well, first of all, there was trapped value in the fact you had to own cabs. And second of all, there was trapped value in a dispatch system that didn't dispatch. And so there was all, it's like, come on. And so every industry, retail, how does, so why is Amazon prime? Well, you have to go to a store and by the way, the store may or may not have what you want. Cause like, no, you don't, you just, you sit on your butt and you go to prime and it comes to you. Okay. So is it trapped value in re-tip? And we said. I mean, healthcare, there's still a ton of trap value in healthcare that we haven't re-engineered out. There's the trap value in education is tragically, you know, huge amount. If you look at what kids learn, as opposed to how much time that's invested in trying to teach them. And it's not that kids don't want to learn. It's that the system really traps value extraordinarily, you know, it's, it's okay. And then in South cons working on that and God bless them. I hope Khan Academy and Khan Amigo, you know, make it all the way. No, it's, it's good stuff. So anyway, the game would be. If I commit myself to be in service to a class of customers, that I'm going to become a specialist in the trap value in those, in that industry or in that or workplace, it could be, I'm going to be a sales specialist or a service specialist or a marketing specialist or an HR specialist, but I've got to pick a domain where I'm going to be a specialist in the trap, in the current state of trap value with the understanding that every technology innovation that re-engineers that, that industry is going to move the trap value. Trap value never goes away. It just goes to a different place. It's like, it's, it's like traffic. You, you, you, you widen the freeway in one part of the city and it just creates a traffic jam in another part of the city. So there's always the next generation of trap value. And by the way, it's not the traffic doesn't get better, but it, but it's never perfect. It's never the third, the second law of thermodynamics, right? You're not going to get a perpetual motion machine. So there's always going to be trap value. And if you, as long as you do that, as long as you genuinely take on the next. Challenge and trap value. I think that is the most.
Speaker 3So is that, does that translate to verticalization in a way?
Speaker 2Yeah, that's the easiest way. But to be fair, you know, horizontal place, like how much trap value did Google release? Perfectly horizontal. But you know, there was enormous amount of trap value in search. Okay. So it doesn't have to be vertical, but it has to be, it has to be some kind of domain where, where your investments amortize in a way. And they don't commoditize. It's not, you're not boiling the ocean or whatever.
Speaker 4We talk about moat so much. And, you know, you see all companies at these different sizes and stages talk about like the moats that they have. Like the crazy example would be like a seed or series a company talking about the moats that they have or will have. Um, even companies that I think have a couple of billion dollars are like, this is our moat right now. And I think of it across the following lens, which is that for it to be a real moat, like for more to be interesting, it should give you. The ability to thrive across at least multiple platform shifts. If the minute a platform shift happens, you're dead in the water. Like, do you have a moat? Like just cause you were doing well for like a certain amount of years, as the world evolved, you hit a wall. And if I look at it from that perspective, I can't think of many companies that had any real moat, but as soon as this platform shift happened, they all like, you know, kind of like struggle city right now. And you saw the same thing when it happened. From perpetual to SAS. And so you could argue that maybe one of the, maybe Google has moats, right? Great. Like, and Apple has some moats, you know, supply chain capability. Like maybe there's like 10 companies in the world that have real moats and maybe everyone else never really had a moat. Cause if you're new, you're not going to have a moat, right? Like by nature.
Speaker 2You might have portable moats. I think for the, I think the thing that I would wish for the normal company is who's not the Google, who's not the Apple. But I would say, okay, first of all, we have floating moats. So they're a moat, but, but that was not, this is not a castle. We're not on the war. Maybe we're on a ship that we can, a battleship where we can move, but we have to move. Our moat has to be able to migrate. I do think that, I mean, if you look at the value disciplines of market leaders, there's a book written. Gosh, 30 years ago, 40 years ago, Tracy, Michael Tracy and Fred Worsman. Anyway, he said, look. It's either operational excellence, customer intimacy, or product leadership. Those were the three moat making disciplines. So we've been, so what you're pointing out is if you say it's product leadership and tech, you're kidding yourself. Because this, the product leadership is just, it's too.
Speaker 4I think it has to show up somewhere is the point I'm making, which is that if you have a moat, you have to have, you have to have a certain revenue growth rate over a long period of time for me to say, that's a moat. Yeah. Otherwise, it's not a moat.
Speaker 3I think mode is a dynamic state. And to your point, Asit, what it really is, is what is the rate of decay? Like how durable are the cash flows associated with your strategy? And the more durable, which is like the lower the rate of decay, the better the mode.
Speaker 4I'm arguing against that, Sam, in a way that I'm saying this obsession with it being cash flow related is probably the wrong way to look at it. Because what does the market value? The market values your ability to be much bigger tomorrow than you are today. The market doesn't really value profitability all that much. Otherwise, Salesforce would be trading at a much higher multiple than it trades today.
Speaker 3I don't mean profitability necessarily. I just, when I say cash flow, I mean it as proxy for revenue too. My point to your, I mean, I think we're all. In agreement, which is these things are all fluid, right? No company has a permanent competitive, maybe the U.S. government, right? The U.S. government has a permanent advantage in the market of issuing currency and, you know, killing people on behalf of American citizens and things like that. That's where they have a permanent monopoly. But everybody else is looking for where can I try to compound advantage such that that revenue is more likely to be there tomorrow? You know, it is the highest likelihood possible that it is either there tomorrow. Or growing. But it is a dynamic state, right? Nothing is permanent. And maybe the moats of today are the, like, if you can move more quickly than it is both the speed of execution and how quickly you can interconnect brand plus product, you know, plus product, plus distribution in some kind of flywheel that at least for some time period compounds. The rate of change, the problem with exactly what we're all talking about, the problem with the rate of change generally in society is that what it will do is. Depress all moats. And the problem with that is that the should be problem. Maybe it's not true in markets. The should be problem is that what it should do is impact, you know, the terminal value of all these enterprises. Like if everybody can be copied instantly and nobody has a secure place in the market, then all companies should be worth less.
Speaker 4I think this is why the better framing for companies to think about is this idea of a comparative advantage, because like the moat almost feels like, you know, I remember. My mom used to tell me when I was in high school, I was a bad student and she was like, this is the time to study. You can have fun later. And like, I feel like it's that version of business. Like it's we're going to do all this hard stuff and then we'll have moats and it'll be chill. And it's like there's no businesses ever chill. Like ask Google. It's really hard to be Google, right? It's really hard to run Google as well. Like business doesn't necessarily get easier. It just gets different. But the hardness stays high. And so a comparative advantage is an interesting framing. It's like what comparative advantage is to you? And how do they evolve as things grow to give us an edge in the market that we're competing in? But it's going to be violence, you know, competition is violent and it's going to be that and we're going to have some advantages and they'll evolve.
Speaker 2So this goes back actually to the book right prior to Zone to Win called Escape Velocity. And basically what I was trying to say in that book, because it relates to what Asad said about why is Anthropic worth a gazillion dollars and Salesforce not? And by the way, the dot-com era, why is, you know, the dot-com guy and IBM not kind of thing. So what we realized in that book was he said, look, investors care about performance, but they also care about power. And the idea was the more powerful you are, the more I can project future performance and therefore the more, the higher the PE ratio I will give you. And so we said, well, we need a better vocabulary for power. So we went into this framework called the hierarchy of powers. And it said the hierarchy of powers has five powers. Category power, company power, market power, offer power, execution power. Those are the five powers. And the idea was that is also the priority order. So the number one variable in the performance of any company is how well is the category it's in do it. So Dell and Lenovo and HP are in the PC category. PC category is not so exciting. OK, meh. OK, you know, Oracle and SAP and ServiceNow and Salesforce. And SaaS category, meh, meh, meh, better than PC category, but meh. And then, you know, then we have the hypercategories. Like we've got, you know, NVIDIA and we've got AMD now kind of doing stuff there. And now some other chip guys are coming in. Whoa, category power up the union. So that's where the bubble comes from. People are really enthralled with category power. But one of the things you realize if you go down one level is as the category establishes itself, there's a pecking order around companies. There's a pecking order around companies. And the market leader has the number one power. And the number two has about half of the power of number one. And number three has about half of the power of number two. And then it's almost no power at all. So all of a sudden, company power becomes incredibly important once category power settles down. That's where the moat idea, it starts there, I think. You say, well, and my moat, if I'm number one or number two, is I have an entire ecosystem that brings me business, that makes a living. And I have a lot of people that are on top of implementing my stuff. Oracle databases aren't going anywhere, people. And God bless Sybase, and God bless relational technologies, and God bless Informix, but it was Oracle. So then you go down to the next level. What do you do if you're Sybase or Informix or Oracle? I can't win that battle. So now I'm going to go down to market power. I'm Apple. I'm going to go into education. I'm going to go into art. I'm going to go into graphics. I'm going to go into advertising. I will find markets where I can kick IBM out. And I will have a Macintosh. I will have a Macintosh PC model. So that's market power. And I would say I spend most of my life dealing with tier two companies trying to figure out, make their piece. They're not the gorilla, but they need to have a future. And so how can you do market power? And if you establish a strong left position in market power, like the people in CAD or the people in specialized chips, okay, that'll work. And then there's offer power. And right now we're inundated with this. Very, very rapid permutation evolution and offer power in AI. And it's making the whole category kind of.
Speaker 4I like the word power for this, because I think in our, as we grow up, we learn so much about power through schooling, right? Like you read history. And the one lesson I think we all take away is like power is transience. Like, you know, you gain it, you lose it. Like it's not, it's not permanent. It's very, I can't think of permanent power. Whereas when you think of moats, you think of them as like these really long, long lasting things. And that's like my gripe with it. It just feels long lasting. It does. It's not really in the grand scheme of things.
Speaker 2Well, okay. But let me give you some long lasting moats. The combustion engine. How long has the combustion engine had a moat? A hundred years. A hundred years. Yeah. And by the way, do you think it's going away?
Speaker 4But has any company within, I think a piece of innovation can have a moat. You could say the internet has a massive moat. Yeah. But. Yeah. Any company within that space, I can think of maybe one or two companies that you could say have real moats within internet as a category.
Speaker 2That's not. Mercedes, Porsche, General Motors, Ford, Toyota. I mean, I can name 10 custom companies that have a 50 year moat because they have distribution because they have, I mean, I mean, look, the key is it's got to be a slow moving industry. You know, it's like mining, you know, I don't know who the mining companies are, but I'll bet the mining company of today is, was the mining company of 50 years ago. The reason we're so sensitive to it is in tech, your argument holds a hundred percent. But we all live in tech, right? We all live in this, in this world that, I mean, I mean, IBM, when I, when I entered tech in 1978, it was unimaginable that IBM would not dominate tech for the rest of my
Speaker 3life. And how, how long was IBM's run was from the fifties, maybe like late fifties, early sixties through like the mid eighties?
Speaker 2Not really. I would say that their run really started in the seventies.
Speaker 3Okay. So that's still only like 20 years. Cause I'm thinking, I think if we're reflecting on it, I think we have to give credit to Google. Like that had, what an incredible run that company is at.
Speaker 4It's wild. Yeah. But even on the examples, Jeff, that you were giving, if you look at like Mercedes or you look at these companies and you look at their stock performance over the last 10 years, you wouldn't say if we just traded a bastard of the world's most performant companies, they wouldn't be in your top 20.
Speaker 2No question. Because they, because they don't, they're not in growth categories. Remember, this is maybe the way I would try to make the point that I think you're making. It's an important point. I think that the stability of the moat and the growth rate of the category are inversely
Speaker 4correlated. Oh, yeah. I see what you mean.
Speaker 2Yeah. And so therefore the valuation is going to be.
Speaker 4So if you have like a slow moving category, you can have a moat that like is, yeah, it just might not look great on your stock performance.
Speaker 2By the way, if you are a value investor, that's all you invested.
Speaker 1You love that. So, so Jeff, this, given this category of frontier models, just all over the place, we know it's just a skyrocket. You use the word settled down yet. And you look at the company power of open AI and anthropic in three years. Are they still top leaders in this category? That's a good question.
Speaker 2I actually don't think they have as much company power. They have category power up the in yet. I think it's still category. I think because if you look at like, I mean, you contrast Sam and Dario, I think those are two very different leaders. One of whom I think is extremely competent. One of whom I always have questions about. So we won't say which one is, I won't say which one, but my point is right. Now, and by the way, I mean, I'm the worst display, but I thought Google guys were just crazy. I mean, I thought they were nuts. You know, so, I mean, come on. I mean, I'm not that good. But the point is, I don't think that has settled in because, whereas I think to your point about Google, what was amazing about Google is it has, and Apple, I would say, is the other one. The fact that Apple has been able to re-embed itself so many times. And Amazon, when Bezos was there, I don't know how it's going to happen under Jassy, but under Bezos, it continually did it. But that is a miraculous performance. I mean, those are like, you know, yeah, that's amazing.
Speaker 3To your point, Jeff, like the, what you were saying about the intelligence eventually will become a commodity, which will be good for everybody, but not good for the first purveyors of it or the first manufacturers of intelligence, which would be anthropic and open AI. And to your point, like, it's hard to think of a commodity, if you're equating it to electricity, or you're equating it to some raw material like steel or something like that. It's hard to think of a company that has anything like the multiples that open AI and anthropic. That produces a good that is effectively a commodity input to all of the other goods in the economy. You see it in cars?
Speaker 2Look, I would say my example of a successful commoditized industry was the Wintel franchise. Now, think about it, because the cash flow at Microsoft, it was incredibly good for a very, very, very long time. And the same thing with Intel. You could, the number one player can actually make quite a good. If you're living in a commoditized environment, the number three player cannot.
Speaker 4And then you've got the Toyotas, right? They make as many as Ferrari has made in its life, they make in a week. And so with intelligence, the way I think of it is you've got frontier intelligence, which might be really high margin. This is your bunch of Harvard PhDs in a data center type of intelligence. And then you've got your like, you know, community college type of intelligence in a data center. And that's, you know, low margin, super commodity, like electricity pricing, something like that. That's probably the best way. I think so. And by the way, I
Speaker 2think it's a trickle down. I mean, I think that eventually that Harvard stuff does get to the junior college, but it's going to get there in a way where it has building guardrails. It's going to be, it's going to be much more modular. It's going to be much more kind of like Lego assembled to order kind of stuff. It won't be as miraculous, but it will, because it's so horizontally ubiquitous, it will raise the GDP. I mean, there's. That is what's going to raise GDP.
Speaker 4Can we click into this a little bit? So you're kind of the master of thinking through how something goes through a market. And so here's this challenge, and you understand enterprises really well. So here's this current challenge, which is that for all of these investments to make sense, AI must eat labor budget. It cannot just take the technology budget. It has to eat some of the labor budget in the economy. Yeah. It cannot just make every person slightly better at their job. It has to start doing some of those jobs or do them at a level where you need far fewer people than you need today. So if you're doing insurance claims, you need like two instead of 2,000 or two against 200 people, something like that. And that's where the problem of accuracy comes into play, that when you talk to these large enterprises today that are testing AI across all these different use cases, they're not going to be able to do that. They're like, the error rate of even frontier intelligence is too high for us to give it such important things. And the argument many are making, the naysayers are making is, this is not, this is built into it. This is part of the technology. It's probabilistic and will make 3% to 5% errors. So then you just can never give it mortgage applications and insurance claims and payroll where you have zero risk tolerance to imitate. Right. In that case, we will hit a wall in terms of the ROI generated versus the spend.
Speaker 2Okay. I would push back against that argument in two places. The first one is the initial assumption was the ROI from AI will come from some portion of labor reduction. Okay. I'm going to say that that is possible, but not necessary. Okay. And the reason I'm going to say it is, no, the ROI is going to come from trap value being released. We're using people to release that trap value today. Okay. If we can use AI agents to release the trap value, that's where the ROI is going to come from. By the way, today, people are doing a bunch of jobs where the people are trapping the value. It's like when ATM machines came out, right? You used to have to go to a bank to cash a check. You had to walk in and cash a check in a bank. It was the only way you could do it. Then the ATM machine came out. Does anybody go and cash a check in a bank anymore? No. Are bank tellers' jobs go away? Yes, they did. Firemen on trains, did they go away? Yes, they did. Okay. So it's not the job. Okay. Okay. Well, it'd be job elimination. The question then is going to be, but so what? I think they're coming from quality improvement and latency reduction, not from labor reduction. To me, that you are fundamentally changing the dynamic of the transaction. Now, to do that, you have this issue of, okay, I've got probabilistic AI, but a transaction has to be deterministic. So you can't say, well, I don't know, it was either $10 or $15. Why don't you put $12 in this transaction? It's a system of record, and we'll round it out at the end of the quarter, right? No. Okay. So we know that the transaction must end in a deterministic state. We also know that no transaction begins, or very few, in a deterministic state. It begins in a probabilistic state. Well, Mr. Moore, what kind of shoes would you like to buy today? Uh-huh. And then at some point, I buy the pair of shoes. And we've always said, well, humans do the probabilistic stuff, and cash registers and PC. Point-of-sale systems do the deterministic stuff. Now we're saying, no, no, hang on. We can put an agent in front. Well, okay, will the agent make mistakes? If they're doing probabilistic, yes, but so do people. Okay? So then you go, but the problem is I can't hold a software accountable. I can't hold a person account. Uh-huh. So what we're doing right now is we're putting human right in the middle. Probabilistic, probabilistic human, just check. And you don't have to check everyone, but stay on monitor, stay on top of it, blah, blah, blah. And then we'll go to deterministic. That hybrid model for the next 10 years, that's going to be the model. I mean, we're going to, whether AI enable the front end as much as we can, we'll still keep humans in the loop where we think we need to. Because the humans are responsible. AI can be accountable, but it cannot be responsible. Humans are the only thing, only people on the planet can be responsible.
Speaker 4Is it possible to pay for the intelligence plus the humans in the loop?
Speaker 2In my way of saying that, why would GDP go up at all? And the question would be because we released more trap value than we ever have in the past. I mean, the notion that we could release that much more trap value, I'm excited about that opportunity, but that's me being very, very optimistic.
Speaker 3Therefore, economy is zero-sum. But we all know that the economy is not zero-sum. The actual size of the economy is, the potential size is indeterminate. Because it is, what is the maximal appetite of all of the consumer, consuming entities in the economy? And my experience is that we will do more stuff, and we will do stuff that we haven't even thought because it was trap value. And we will do more stuff, and we will do stuff that we haven't even thought because it was trap value. So we're building a dedicated ABM microsite for every company in their TAM. They have a list of 500. We're going to build microsites for every 500, all 500 of those companies. We're going to scrape the internet. We're going to figure out what those companies' problems are, what those challenges are. We're going to use AI to figure out how this particular product solves that challenge. We're going to send out a bunch of emails. That's never before been possible. I said, well, we've got a universe of like 10,000 pavilion gold members. Could I build a custom microsite for every company in their TAM? Every potential pavilion gold member. So I say, Asad, you've been nominated for pavilion gold. Let me show you the microsite that we built for you. And it says, here's Asad and pavilion gold. Here's the challenges of SDA. Here's how we're going to solve them. That was a thing. That's not, that budget doesn't exist, you know, like in the sense that that capability was not possible before. And then I'll give you another example. For you to pay for that, Sam,
Speaker 4it has to come from your limited P&L today, right? So that's the other side of it is you will say, oh, this is a brand new thing for me to spend on. In terms of all the things I spend on, here's a new item that I'm adding in. But at some point you're limited by how many new items you can add in and you might have to then look at-
Speaker 3Unless they drive ROI. Unless I can sell to vastly more people because of this and because they find value in vastly new ways. And to your point- But nothing is certain.
Speaker 4It is still a bet that you would make. Oh, for sure. So when the time of making it-
Speaker 2Do you remember, by the way, you remember that Google search started with no business model? Remember? There was no business model for Google search when it started. It was ubiquity now, revenue later. Yeah. And that's what they did. But anyway, okay, Sam, you had your other one. You had another example.
Speaker 3Well, this one is not going to meet Asa's standards. Because it is coming from limited budget. But it's very specific and arcane. I used my dot, my chat GPT agent. So I buy vinyl records. And the place where you buy vinyl records is discogs.com. But discogs.com is a pain in the tush to use. So really, it takes 20, 30, not 20 minutes. It takes about 10 minutes to find the specific thing that you want. So you only buy a couple at a time because it's a pain. Well, I had my agent. I built out a list of 100 different records. And I had my agent fill up my shopping cart. While I was doing something else watching a Netflix show. And it added 75 of the records that I want that all were from Spain, shipping to Spain, original pressing, and single LP. I can tell you that I spent an ungodly amount of money today clearing that shopping cart in a way, like my average purchase was 10x what it normally is on discogs because of the existence of agents. Now, I don't have unlimited money. So that's money that didn't go to somewhere else. But that is an example of a consumption case. It's a capability that was not before possible that is now possible due to the existence of AI.
Speaker 2And this is also an example of the trapped value in their infrastructure model.
Speaker 1Yeah. Jeff, something you said that was interesting. You said, in the probabilistic termistic, you said for the next 10 years that we wouldn't, there was something there around 10 years. Why 10 years? And what, I mean, it feels like things are changing so fast that we get there more than 10 years at a time.
Speaker 2Because, well, the thing is, and this is another line I've been dining out on, the speed of innovation has never, ever been faster. Right. But the speed of adoption has not changed. Interesting. And so that's why it's 10 years. It's the speed of adoption, not the speed of innovation.
Speaker 1Are we the early adopters, like the 10 to 15%?
Speaker 2Yeah. I mean, the software industry is, I think, fundamentally wired to adopt this stuff quickly. But if you're in, if you're in any, if you're in government services, if you're in construction, if you're in lining, if you're in, you know, the military, no. And so I think, and then you think about how much liability that could be, at stake in health care and in education and whatever. And so you start thinking, whoa, so I think it's just going to take, and by the way, it doesn't need to take, it's not like it needs to be hurrying. At some point, what will happen is, in fact, it's already happened. We've innovated way past the speed of adoption, which means that we're not going to get the return on this stuff because by the time it gets adopted, it's just, it's going to be downstream.
Speaker 1Well, here's a contrarian question I have for you, because I have never considered myself an early adopter, actually. It's funny being in tech. Me neither, by the way.
Speaker 2I'm not, I'm not for sure.
Speaker 1So both of us probably didn't get the iPhone when it immediately came out. We were in the, yeah, the late adopters. Blackberry forever. But I guarantee both of us are pretty AI proficient today. So we are in the early adopters. So that has, like my, I've changed and become an early adopter in that way. But you said the adoption rate hasn't changed. Is that just ubiquitous across the thing?
Speaker 2To be fair. So the reason, I can, by the way, speaking out of both sides of my mouth is a gift that consultants have. But if I could state for a second, the amount of risks that you and I take when we use generative AI is very, very low. So it's not like we have an adoption challenge because, you know, we're just asking co-pilot to do stuff. Whereas if you say, I'm going to re-engineer the workflow of a corporation or a governmental entity or any regulated industry or whatever, now that's where you start getting the adoption, early adopter problem. There's no, there's no chasm to cross if the idea is just, well, I don't know, hit enter and see what happens. And that's kind of where I think, I think we are.
Speaker 3Jeff, it's been amazing having you on Topline. We'd love to invite you back sometime if you'd ever be open to returning.
Speaker 2Oh, well, this is, this is a lot of fun. It's like Hollywood Squares. I'm enjoying it. Fools vs. Bears.
Speaker 3We, before we go, sometimes we do a segment called Bulls vs. Bears where we ask if you're bullish. I'm going to ask, it's a version of that, which is simply just buy or sell. Would you buy the Anthropic IPO at a $2 trillion valuation?
Speaker 2I'm not going to, so no.
Speaker 1I think we figured out which CEO he's talking about when it says Dario or Stan. No, no, by the way,
Speaker 2ask that open eye.
Speaker 1Why is that? You just feel like we're over our schemes
Speaker 3in terms of,
Speaker 2would you buy the OpenAI IPO?
Speaker 4He said no.
Speaker 2Yeah, I would buy, I would not buy either IPO. SpaceX, I would not buy the SpaceX IPO. No. No, interesting.
Speaker 4Fair enough. Would you have bought the Google IPO when it happened?
Speaker 2Never. Are you seeing a trend? Ask me about my holdings in bonds.
Speaker 3Jeff, give us something to sell or promote that people can learn more about you if they want to.
Speaker 2I think that, I think right now for most of the people in your community, Zone to Win is the book to read.
Speaker 4Awesome. Because everybody,
Speaker 2first of all, to be in your community, I have to have a certain amount of income, which means I probably work for a stable company or a company of some size. Not always. I could be a founding startup. But if you're a founding startup, it'd be crossing the chasm. But if you're, if you're in an established enterprise, it's Zone to Win. And those are the only two business books that really matter, I think, right now.
Speaker 4There's very few books, I think, over the years down the line. Zone to One has stayed in my brain in a way that very few other business books have.
Speaker 3Thank you for being our guest.
Speaker 4Well, thanks a lot. It was a pleasure.
Speaker 3Yeah, it was great.
Speaker 5Thank you. You're welcome. Have a delightful day, everybody.
PodcastTranscript AI