E431: Marty Kausas on AI, Trillion Dollar Startups & Founder Psychology

How I Invest with David Weisburd

Marty Kausas, CEO of Pylon, shares his unconventional approach to building a high-ambition startup centered on 'fun and adventure' rather th

Key takeaways

• Founders should be honest about their motivations—Marty cites 'fun and adventure' as Pylon's core driver, not world-changing missions.
• Market size is the dominant factor in scaling; even weak teams can succeed in large markets, while great teams fail in small ones.

Main topics

• Founder motivation and psychology
• Market size vs. team quality

Notable quotes

Our reason is we just love the game. It's fun and adventure is the reason we run Pylon.

If you take the best founders and put them in a small market, the market wins.

Conclusion

Marty Kausas demonstrates how combining authentic motivation with a clear, ambitious

Transcript preview

Speaker 1 (0:00) Marty, you run one of the hottest companies in Silicon Valley, Pylon, which is backed by Andreessen Horowitz and General Catalyst. And last time we chatted, you said that the reason you started the company isn't to change the world. Why is that? Speaker 2 (0:12) This is interesting. My guess is a lot of founders are not honest with why they start companies, and we are. So if you think about the reasons why someone starts a company, it could be ego, it could be money, it could be to solve a problem that they encountered at a previous company. Our reason is we just love the game. It's fun and adventure is the reason we run Pylon. And if you think about the other reasons, like money, there are much easier ways to make money. If you think about you had this problem at a previous job and now you want to solve it and you're so passionate about it, are you going to spend the next 10 years of your life grinding away to make it happen? I don't know. It just feels a little disingenuous to me. Especially if you're building, for example, we're in the customer support category. Like, I love what we're building, but I wouldn't say I got into it because I'm so passionate about customer support that, like, I must spend all of my time doing it. And Speaker 1 (1:02) you started by looking for... a very specific market, what were you looking for? Speaker 2 (1:07) So this goes into like the foundation of the company. So we have our why and our motivation, which is fun and adventure. And then there's the what. And we think about the company like a giant board game where the win conditions of this board game are get to a billion dollars of revenue in less than 10 years. That was the initial goal. Three years ago, now it feels like that goalpost has shifted and we'll talk about that. But when we thought about, okay, how do we get to a billion dollars in revenue? The question we asked ourselves, well, how have other people? gotten to that revenue scale because that's a lot of money. So we looked at public at the time, SaaS companies who were big and worth over at that time, $10 billion or $10 billion in market cap. And the surprise to us was that there were only 26, $10 billion public SaaS companies at the time. And that meant that, oh shit, not that many people can get to really big revenue scale. What's unique about them? And the unique variable was primarily market. Market was by far the big differentiator. And what we realized was if you take the best founders and the best team and put them in a small market, the market wins. You just can't grow out of that small market. If you take the worst founders and put them in a great market, the market still wins. Those founders and that team will probably do well. And you take a great team and great market, and then that's where you get generational company. Speaker 1 (2:20) And the intuition on that is pretty intuitive, which if you have a billion-dollar total market and let's say you capture 20 % of the market, which is a massive amount for a startup, you still have a $200 million a year business. If you have a $10 billion market and you capture 20%, now you have a $2 billion a year business. And Speaker 2 (2:37) I actually think some people might say, oh, it's hard to predict certain markets and the size of them. I always like to think about Bitcoin and Coinbase. And if you think about the crypto market, in the beginning, you would have been like, oh, crypto is such a small market. How big can it be? But really, it's the financial systems market. And it's like money movement and that market is very big, but there's now some sort of new way to operate within that market, or there's some new trend within it. We basically were looking for, if our goal to win this board game is get to a billion dollars in revenue at the time, then we need to find markets that are already big. We were looking at, okay, within B2B software, where is most, what are the biggest companies? And we looked at Salesforce at the time. Speaker 2 (3:19) And their biggest revenue driver, interestingly, is not their CRM, which is what they're known for, but it's actually their customer support platform, which is called Service Cloud. They make $10 billion in revenue off that per year. And so our lazy way of thinking, okay, what market is big, was just saying, okay, let's go to the top, look at their biggest revenue driver, and try to eat up that market. That's how we picked customer support as our starting category. And then within that, you have to find some sort of new thing that's happening, right? Because you can't go build the same thing, sell it and expect to win. The incumbents have too much of an advantage. And so we went and LinkedIn DM'd a bunch of people in customer support asking them for what's new in the space, like what's happening? What do you care about? What metrics matter? And there's one pattern that we started to see emerge, which was a lot of people were starting to offer customer support over Slack. Which is different from what we're doing now, but it was this like omni-channel communication that was starting to happen specifically for B2B companies. So if I'm a company, I might open up a shared Slack channel or shared Microsoft Teams chat with one of my customers. And suddenly the communication moves away from email to these new conversational platforms. And so that was our initial, we saw that none of the existing business systems could integrate. away from email into these chat platforms. And we said, Hey, there's like an omni-channel moment happening in the market. And we think that will be the de facto way of operating long-term. And so we started with that wedge in the customer support market off of omni-channel support. So that was our first product. And then over time, we evolved it. We started to see, oh, like we got all this feedback from our customers. And at the time we were just integrating into Zendesk and Slack or into Salesforce Service Cloud and Microsoft Teams. We were this bridge. And over time, we started to hear all the feedback from our customers on what they didn't like, what they wanted. And we realized actually, if you built a customer support platform that was built for B2B companies in particular, then you could compete really well against these incumbents. And then it takes us to where we are now, which is we're just reinventing how customer. support works completely. One Speaker 1 (5:22) of the big disconnects in the market is that nearly every CEO that you talk to will say hiring the best people is the number one most important thing that CEO focuses on. But there's not a lot of focus on the pitch to potential employees. When you go into the market with a fun and adventure pitch, how does that resonate? Speaker 2 (5:43) It resonates really well. So it's funny, when we were raising our Series B and our Series A, I remember talking to Merit, our partner at BCV, and the first thing I told her when she came in was, hey, our motivation for doing this thing is fun and adventure. And it kind of stops people because they're like, what? I've never heard that before. No one's said that. And so the motivation, it can't be there alone. It has to be paired with, obviously, an objective that is in alignment with what an investor would want if you want to take their money. And it turns out for us, it is. Right. Because we're playing this giant board game of capitalism and we're trying to get to a billion dollars in revenue. And so the way we think about our market and pairing the fun and adventure, which is the motivation, that's the driving force with the objective of get to a billion dollars in revenue, those pair super well, because then it's a long-term sustainable fuel for us to keep going. And then the objective is very aligned with building a big business. And so we work backwards from the big business. And every time we make a product decision, every time we think about what market to enter, we're thinking about how do we get to that order of magnitude? Because a lot of founders, what they'll do, they'll just say, hey, like. We're excited about this direction. We think it could be good. And they don't really think in terms of what is the end goal, like how big should this company be long term? So when we started, we set a very clear objective of revenue scale, and then that informs what type of products and markets you're willing to build and enter. Speaker 1 (7:15) I wonder if a lot of this is just being differentiated in the market. Again, CEOs talk about hiring, but they don't think about themselves as basically this hiring machine in the market competing against other companies. And when I think about some of the greatest recruiters in the world right now, Elon Musk, Alex Karp, Palmer, Lucky, they're all so singularly different, so weird, but they have a very specific view that they have. And they polarize every person into either this is like the greatest company or I would work at this company over my dead body. This episode is presented by Juniper Square, the operations partner for private markets. Speaker 2 (7:52) It's interesting to, yes, recruiting is one of the most important things. I also think another thing that I'll mention is I think the greatest gift a founder and CEO can give to their company is product market fit. Speaker 2 (8:05) And it kind of has to start there because in order to attract an incredible team to come work at Pylon, I need to give them the potential that they get, the potential energy that they can make kinetic, right? I need to give them, here's the boat we need to assemble. And if we assemble this boat and row in it, like we will go to the promised land. And so product market fit has to come first or has to be like. The potential for it has to be there and very clear. And I think because we work backwards from a very ambitious goal, it's very easy to explain to someone, okay, here's how we're going to get there. Because we've already done the math. We are thinking about this in terms of this order of magnitude, revenue, et cetera. And so you have to start with product market fit for yourself, and you need to be honest and clear with yourself. And then it's actually very easy to communicate that to your team and people you try to hire. Here's what we're trying to do. Here's why we're doing it. Here's our plan. And there's no ambiguity of like, oh, does this really make sense? No, we're like, hey, we want the most certain path to get to this revenue scale. Here's our goal. Here's the market. Here's the incumbent. Here's our plan. Here's how you can break down the math to see that the market's big enough. And then also, the thing that's hard about recruiting is today, there's so many great companies to join. And so how do you differentiate yourself? For us, it's almost always the market. Speaker 2 (9:30) And so when someone's interviewing at Pylon and they're interviewing at Anthropic, it's really hard actually. Anthropic and OVNI are the two that are like, it's kind of hard to argue with their market. But most companies are very easy to compete against because just from the start, their markets are small is the problem. And so they didn't think long-term about how potentially big the company can become. Whereas we have thought about it. So for example, let's say we have an offer out and someone else is offering. double the equity. I will go, I will do the math in front of them. Here's company a that they're interviewing at, and then here's pylon company a like we can look at the existing market. It exists, right? There's already some market out there that we could try to size up and understand how big this company or product could be. And it is one to two orders magnitude smaller than what pylon can be. You look at the previous generation of SAS era companies. they have given a really good relative sizing of market. So customer support is a $40 billion SaaS category. And then the AI version, the agentic version of that is going to be even bigger. And then you like a much smaller company, let's say an SEO company. And hey, the agentic version of that, like the biggest company there is a billion dollar market cap, right? Not even a billion dollars in revenue. So you take an SEO company, for example, let's say we're competing with someone who's doing like agentic. engine optimization, like the AEO stuff, they're competing against incumbents that are worth a billion dollars in just total market cap. And then we compare it to Salesforce Service Cloud, which is $10 billion in revenue. It's literally two to three orders of magnitude higher, which is crazy. Speaker 1 (11:14) And from the very beginning, you wanted to build something really big because you believe it takes about the same effort as a small company. Talk to me about that. Speaker 2 (11:22) When we were starting the company, we were considering whether we should bootstrap or, hey, maybe some ideas are in markets that are easier because they're less competitive. But the problem is we realized for ourselves, no matter what, we were going to put in 100 % of our time and energy into whatever we do. And so you may as well go for the most ambitious. company you can. And actually, in some ways, it makes a lot of other things easier. For example, hiring great people is much easier when you're trying to be really ambitious. It goes Speaker 1 (11:52) back to this equity pitch. You could... Entice people to join your company because they're equitably worth it. Speaker 2 (11:58) It's equity and it's also just fun, right? That's one of the things that's underappreciated by people is you don't have that many companies that you get to work at, right? Where you can put in 100 % of your effort. You aren't going to be able to do that for infinite companies. You have a small set that you can select and be like, I'm going to go all in on this company. And you don't really want to do that if the ambition's not there, right? If it's like, oh yeah, we're going to kind of like get to $10 million in revenue and just like cash out, sell, et cetera. It's just not exciting. The people who joined Pylon are there because they're like, this could be a generational company. This company could be far bigger. Speaker 2 (12:41) than like any company I've been at before. And so this goes to the original starting point. Like when we had started the company, we started literally six days before Chachi PT came out. So it was like, it was crazy timing. And at the time, a billion dollars in revenue felt very ambitious. Then this year, we're rethinking that and we're like, look, what is actually generational now? And generational now means actually you get to trillion dollar market cap, right? The goalpost has moved. And so we went back to the drawing board. We're asking ourselves, how do you get to a trillion dollar market cap? Because that's what generational means now. And so we went back to the drawing board. Okay. Let's like look at our market. Let's look at our positioning. Let's look at how pricing is going to evolve. And we basically figured out that the like fully agentic version of what we had built for customer support teams is actually something that could 10x. the revenue potential of the company. So if we're charging $100 per unit right now, actually in the future, we're going to be charging $1,000. A way to explain this is in the SaaS era, let's say you are Salesforce. You might charge $100 for a seat in the platform. But what's interesting and is happening in the market now is people are adopting Cloud at their companies or they're adopting Chat and they're paying $1,000 per person for that. And so suddenly, and it might be for the same type of work that Salesforce might have been providing. So let's say you're using Salesforce's ticketing system for customer support. We're seeing support team members now spend like $1,000 per person per month on cloud. Now, the question is, can Salesforce go and build whatever features that someone is trying to build out or use with cloud? And the answer is they can't because they're not very good at building and they're too slow at this point. And it leaves opportunity for people like us to say, hey, we actually can build that out. And so we can be in the customer support category. And the new AI products we build actually are competing for the spend that is now going to clod that otherwise Anthropic just takes or OpenAI takes. And essentially, everyone's using the general purpose models to try to. Speaker 2 (14:59) build or use them in a way that are not optimized at all and are super expensive and slow and just not worth it. So our competition moves from, Hey, we used to compete with these support platforms to now we're actually competing with Anthropic for a very specific job function. And so we're trying to move spend away now from Anthropic, from Claude, from chat to Pylon. And that's in order of magnitude more than what they'd be spending on the previous products. Speaker 1 (15:28) I've been investor in Anthropic since its series C. Congrats. Thank you. And I never believed, I think it's somewhat an absurd thesis to think that it's going to replace very specialized, very verticalized solution to B2B. I don't think that was ever part of the pitch. When people think about that, it's maybe an interesting thought experiment. But when you look at how companies operate, how they choose software, how they deal with compliance, how they deal with execution, it's just not possible to replace. with one solution from one company. Speaker 2 (15:59) Almost every repeated Claude prompt or skill is a company waiting to be built. Speaker 1 (16:06) It's kind of like the old Craigslist. Exactly. That's right. Where you have all the Craigslist categories in each one. Airbnb was one of them. Speaker 2 (16:15) Exactly. Airbnb, like maybe Thumbtack or like dating like Tinder. Right. So, yeah, I agree. It's exactly the same thing because what Claude is doing is it's filling the gaps. And I can give you a very specific example of this. We have a bunch of cold callers at Pylon, SDRs, and their job is every day to go pick up the phone. They use some cold calling software. And after each call, I want them to get coaching. And so when we first started adopting Claude, I said, hey, take your transcript, go paste it into Claude, and then let Claude tell you what you should have done differently. And, you know, I was like, wow, this is so cool. It makes so much sense. And then I realized over time, I'm like, why are they going from their cold calling software, copying that transcript, then going and pasting it into Cloud, and then going back into the cold calling software? They're totally disconnected. If that product had the coaching just built in, that would be way better. It would actually be a better product. And so I think Claude is filling the gaps because people haven't caught up on the product side and built the right thing yet. And so this is where, for example, I look at, once again, you go back to Salesforce. They have all the advantage right now in terms of distribution, in terms of brand, in terms of relationships, but they just can't execute on the building. So they have agent force, yeah, blah, blah, blah, but it's not like I don't think it's real is what I'm hearing from people. And so it's an execution challenge for them. Like they have all the advantage as like they own all the data, they own all the context, they own all the relationships. They should be able to