Zeynep Yavuz-Willson, Partner at General Catalyst: GC is an early stage investor and AI transformation firm

0:00Would you say GC is still a venture firm? No.

0:05Zeynep Yavuz-Willson:How do you define or describe it then? That's really what we all think about. Are we partnering with the most exceptional people and are we the first people that they call? Are you playing the role of a transformational partner for portfolio companies? Absolutely. We are very focused on transformations. AI can really transform the way we do work. So we are an investment company, but we're also an AI transformation firm. Very interesting investments. We focus on early stage. We're very active at the seed stage. Our mission is to partner with exceptional people. There are a lot of great investors in Europe and a lot of great firms.

0:35Zeynep Yavuz-Willson:We care a lot about what we call GSI, like familia. So the people in our network. So what does it come down to then? It's about the customer relationship. It's about the trust. What are the biggest mistakes that a venture investor today could make?

0:51Hello, welcome back to the Scaling Europe show. Today I'm joined by Zeynab. Welcome.

0:54Zeynep Yavuz-Willson:Thank you for having me. Thank you for having us and these amazing offices. How are you doing today? Very good. How are you? very good now i think you're one of the most exciting investors in europe in this age of ai investors investments like lagor dwelly very uh interesting investments right both focus on the services industry yeah so just like get straight into it why is that the bet that you've made a couple times now in this age of ai what does that say about where you think the value is going to be captured yeah maybe i go a little bit to the like the story of how this like ai transformation thesis was born.

1:27Zeynep Yavuz-Willson:So you might know, you know, at GC, we're an investment company. And at our core, we're very much of a people business. So we focus on early stage, we're very active at the seed stage, our mission is to partner with exceptional people, as early as possible in their founding journey. We are intentional about that. And it's also a little bit kind of you can't plan it, right? You meet exceptional people anywhere, anytime. And that's a little bit kind of like the fun part of the surprise part of the job. And then we have the intentionality side, and that's quite important as well. So when AI happened, we kind of got around all together as partners, and we were trying to map out where we think the biggest impacts are going to be, what are going to be the second, third order consequences of having AI diffuse into different economies.

2:15Zeynep Yavuz-Willson:And one of the topics we had was services. So a lot of service categories that are very text-based and voice-based where AI can really transform the way we do work and also the way people experience those services is going to change. And actually that is quite different in the US context versus the European context because a lot of these services are quite local. And also the industries will be transformed with AI in different ways. And I think Legora and Dewey Lee are actually perfect examples of this. So step one was like we knew AI was going to transform services and these categories, like these services categories are just so much bigger from a TAM perspective compared to software, right?

2:56Zeynep Yavuz-Willson:So that's super exciting. And beyond that, it touches everyday lives. Like if you think about accounting, you think about property management, real estate, law, you know, these are touching everyday lives of people. And I think that's what really inspired me. The second thing that was important was that I knew that for some of these industries, by just selling AI tooling to them, we could transform them. But for some other industries, selling them technology was not going to work. And I can talk about that in the context of Legora, where I believe selling the technology is making real difference in the way lawyers and legal professionals are doing their work today.

3:39Zeynep Yavuz-Willson:Whereas in the property management space or in the real estate agency space or in accounting and insurance, we just didn't have that. These industries were not buying the technology. We were not seeing the same market pool. but the opportunity was still there to transform these huge industries a lot of investors would look at property management or real estate and will say i'm not investing in this again yeah yeah yeah because so many people have been burned including myself like i remember when i got the way least first like when i saw the first pre-seed deck i was like property management or like real estate no because it's so hard right yeah so you need to think about this in creative ways you can't be like i'm just going to do a seed investment and see how it plays out yeah so there are not that many firms i think who can really get behind this thesis and say i'm going to think creatively i'm not going to have like a rigid way of investing and it's not just going to be capital to do this anyways but i am going to go try to transform or help people transform property management and real estate and accounting and insurance even though these industries are not buying any technology looking at very specifically then given your i don't know your experience being burnt in this industry, what made you kind of get over that?

4:50What made you think actually what they're doing or the approach that they're taking, I'm still going to get involved?

4:55Zeynep Yavuz-Willson:First, I mean, I met them and I think like when I first met them, I was kind of like you see their hustle and they had such deep industry insights as well because they had built a business in real estate. And that's important because they had built a business in real estate. They knew that driving organic growth in that market is closely impossible structurally. because a lot of the contracts, rental contracts, as you know, in the UK are three to five years and people will just renew them. So if you want to churn that landlord or tenant away from an agency, good luck. You know, you have that window that's going to happen in three to five years.

5:28Zeynep Yavuz-Willson:Same with property management. You know, a vote happens in a building once a year and you need to convince everyone to vote to change to you. So like driving organic growth in this market is like really hard. Then if you were doing venture investing, right? Like all you care about is growth. Like all you care about is organic growth. So obviously that makes sense, right? You do classic venture. It doesn't just fit in your mandate. So you say, I'm not going to invest in this property management, leave it. But then with our mindset, what you can do, and this is how we do our investments out of our creation fund.

6:02Zeynep Yavuz-Willson:We still have the ambition to say, okay, yes, venture, classic venture investing, maybe it's not going to work, but this is still a huge industry that's arrived for transformation. So let's think of a way that will work that can transform this industry. And there we knew that inorganic growth to get started, to get to some scale, was key to effectively seeing how we can start that transformation. And when we get to a certain scale and we drive better customer service, then ultimately organic growth will be there. And maybe the organic growth is not going to be at the same rate of a Legora. It might not be as fast because the market structure is different, but that's okay because our goal is not necessary to invest in the fastest growing company.

6:46Zeynep Yavuz-Willson:We want to invest in the company, be part of the journey for the most enduring company. And that's what I'm focused on. And so when you look at Lagoora and you look at Dwelly, so your thesis was Lagoora is in an industry that can buy the technology and implement it at C-value there. Dwelly is a very different type. It's like back to front. If you want to get the growth, you have to buy the growth. Right. You have to buy the company. You have to buy your customs. Yeah. Because you cannot get your customers. You know, in law, there was just such market pull. Yeah. You know, everyone. And it's just the way also the legal market is structured.

7:18Zeynep Yavuz-Willson:If you look at the law firms, you know, there's AM200 and there's like the notable companies of Europe. And when one of the prominent law firms buys a technology, all the other law firms are watching that. So they're like, OK, I did also buy AI. So the adoption of AI in legal was just so fast because there was a market pull. Well, then you look at property management. Yes, if they did buy tooling, they would also be transformed with AI. But they're just not buying because the structure of property management in real estate is that it's incredibly fragmented. So you don't have the same domino effect and the market pull.

7:57Zeynep Yavuz-Willson:Even if you were to build the most amazing product, they might not buy. Yeah. So and we see that particularly, I would say, in Europe compared to the US. And this is why I'm I'm very excited about like investing with GC and partnering with founders with this strategy in Europe, because I know the plain vanilla venture way of doing is not going to work. And we need to think creatively. And at GC, we can think creatively. Right. We have so many different investing strategies. We think about capital solutions creatively. And if we don't have the capital solution that's required for that company to succeed or to disrupt that market, we'll innovate it.

8:35Zeynep Yavuz-Willson:We'll come up with it. That's interesting. So you've got the capital and the flexibility to adapt to different models of the businesses. When you look at the services industries across Europe, right, let's say you take that go to market that you have in the law firms where people are buying technology versus the roll-up play. where are you what other service industries are you looking at where you think okay this is a prime opportunity for ai you mentioned accounting maybe it's accounting maybe there's other services industries that you're thinking are yeah what are those services and and how are you thinking about equipping those services with capital so um there are a number of them that i mentioned and the other thing is that i would say you know doing this in the u.s and europe is different yeah but we've done a lot of the thesis work holistically as a team so as a global team and we've already mapped out those markets so property management is one real estate is one legal is one yeah and insurance and within insurance you have mgas you have the tpas like the the claims management you've got bpos so like it goes really deep within insurance and the brokerage space then you have wealth is also another category and and particularly by the way, if you just think about accounting, real estate, property management, wealth, insurance, these are not even in-house functions necessarily.

9:56Zeynep Yavuz-Willson:What I mean by that is that they're selling to SMEs and they're selling to people. So these industries will just continue to exist. They will never be fully disrupted with AI. People will be living in buildings. There will be property managers. People will be looking to rent and people will need to file their taxes and people will need to get insurance. And all of these are regulated industries. And that relationship that you have with the broker, that customer relationship actually is the key mode. So what we try to do with the AI transformation work that we do is that how can we free up the time of the agents, the real people who work in these industries, so they can really focus on the relationship side, because that is the mode.

10:37Zeynep Yavuz-Willson:It's not the admin. You know, the admin needs to get done. That needs to be automated so that you can really focus on the relationship. And I think, by the way, that's really interesting is that we always are talking about or trying to understand like the moat in AI, right? And I think more and more people are realizing like even the moat in AI is that customer relationship, is the trust that you have with your customer. It ends up being kind of the only moat now that we see in AI. And like the challenge of AI, in my perspective is not going to be a technological challenge in Europe. Obviously, AI is a technological challenge and, you know, robotics and there are different layers of it that needs to be still figured out.

11:21Zeynep Yavuz-Willson:But the pace of acceleration on the tech side is happening incredibly fast. Yeah. What's not what we're not catching up with is more the change management piece. So the diffusion of AI and how are people going to be using AI? And that's I don't think that challenge is really solved. and you can't like move fast break things because we're working with real people. You know, people need to adopt it. Yeah. And that requires trust, word of mouth. And it might just not move at the same rate as, you know, some companies haven't grown historically. So to go back to what you said maybe like a minute ago, you talked about how relationships is the moat.

11:57Zeynep Yavuz-Willson:Yeah. What does that mean in reality for companies that are building technological products? So having, you know, a front row seat to Lugora's journey. Yes. How is that, how are they using that moat to sort of grow quickly and maintain their base? So on Nagora, I mentioned right in the law firm market, it's a very tight knit community of people. And the customer references you have is everything. I've done obviously a ton of reference calls on my customer calls on Nagora, you know, and I'll meet someone like I'll speak to a legal buyer at a law firm, I will not name it. in the reference call he will say you know i i bought this because you know i believe like i believe in max and i know i know you know kind of people like max will change the world this is like an ai manager in a mid-sized law firm and he's thinking like the person he's buying the technology from and the team he's buying it from is going to change the world yeah so that that relationship that you establish people that level of trust you know you make them feel that way and then of Of course, after you sell the technology, the DAU, MAU numbers, so the engagement is incredibly high.

13:08Zeynep Yavuz-Willson:I think that's the mode. The level of engagement you have, the way you make your customers feel, you trust them, that's really the mode. Another example is Langdock, for example, another one of our portfolio companies where they have such strong word of mouth. and you can hire a massive sales team to, you know, nowadays, you know, and it is important, you know, the deployed engineering resources, these are important to like implement change, et cetera. But actually word of mouth is if you can have that, that's so much more powerful because that scales. Yeah. And that's really interesting. And then you're making this bet on the services and I guess on the end uses and the applications in these companies.

13:52Yeah. That's where you're thinking the value is going to be. Where do you think about the rest of the AI stack then? What's your view on, say, the model layer, open or closed source, the chips, the underlying technology, the hardware? You've made these bets sort of at the very diffusion and you talk about people actually using it and adopting it. That seems to be where you're taking your bet. That's where you think there's most of the value to be captured. what does that say what are your views then on the rest of the stack and the value that will ultimately be captured by the hardware players at Frontier Labs?

14:24Zeynep Yavuz-Willson:I have some points of views and but also watching very closely on how things change so I focus on the applied AI layer right so I'm thinking more about how we can apply AI how we can have the trust of a customer I look at the engagement and how people are using the product yeah and I believe that relationship is the defensibility. But the rest of the stack is also important. And I think open source is a big tailwind for a lot of these application layer companies. We're already seeing model companies move to the application layer, right? I mean, we all work cloud, we work in cloud. So that is an application.

15:00Zeynep Yavuz-Willson:That is how I engage with Anthropic. And I think that will continue to happen. But now the application layer players also have the opportunity to move down the open source layer. And there are a number of companies, I will not name them in our portfolio, outside of our portfolio, that since the December moment where I would say the costs of AI have really significantly gone up, capabilities have gone up, but costs also have gone up. And open source, I think, is now there is more consensus around open source catching up with a bit of time lag with a lot of the frontier models. a lot of them have a very active open source and moving to our own model strategy.

15:42Zeynep Yavuz-Willson:So I don't think we can say yet open source, like open source is going to make applied AI have the moat and like things open source, it's all going to work. All the companies that are building up the application layer have to be actively thinking about this today. And if they're not thinking about it today, I'd say that's a concern. That has definitely been the most interesting switch in the last nine months. Nine months ago, it was like the frontier companies, the labs, they're coming for the application companies. Anthropic, they're going to come for all the application layer companies. They're going to do the legal use cases, the vibe coding, whatever.

16:19Now it's like actually these application companies are going to replace the providers of the models themselves with open source.

16:25Zeynep Yavuz-Willson:Exactly. So what does it come down to then? It's about the customer relationship. It's about the trust. Actually, whoever has that engagement layer has, in a way, a leg up. As long as you have the same performance or better performance, or same performance but better trust, the focus company that's in the application layer and the customer relationship should ultimately win. And is that true, do you think, for consumer-facing applications as well, where there's less of a relationship, but you've still got, I guess, like brand? and you've got the relationship is a bit more ethereal than a one-to-one customer success relationship.

16:57Zeynep Yavuz-Willson:Yes, and that's a good point. It's trickier. For people to switch, consumers can switch a lot easier than businesses, right? So I think it's something to think about. But ultimately at the consumer level too, the brand affinity and changing people's behaviors and when people are adapting, they adapt but they don't like to change that much you know developers are different for example you know they will actually be constantly testing new tools and switching they're kind of like the worst people to sell to actually you know they are constantly trying new things and they will change and they want to find the next best thing and you know what consumers do consumers will say if you're they'll say they'll be okay with whatever is good enough yeah but they care about cost so they will maybe switch because of cost because for them if something is good enough like it's good enough.

17:48I guess it goes back to your point on trust. It's still about trust, but I guess consumer-facing applications build trust and loyalty in a different way, I guess, to the

17:57Zeynep Yavuz-Willson:sort of typical B2B. And also, if you look at the business side, the other thing that we always talk about is the data modes, right? And that's another question. Is there really a data mode? I don't know if anyone can today come and say, I know for sure there's a data mode, but what I can tell you is that if you look at our AI services companies, whether that's a Legora or that's a Dueli or that's an integral in accounting, they're effectively layered where the work gets done. Legora is a platform where the law firm and the legal client is collaborating. Dueli is a platform where the work gets done, where the tenant is having requests and the real estate agent is responding to it.

18:41Zeynep Yavuz-Willson:Same thing with accounting. It's where the work is done on integral where the customer is having questions for the accountant and all of that, that is data that's being generated on the platform. So we're talking about effectively this reinforcement learning environment that you don't pay for because it's part of your stack. And I think a lot of the applied AI companies, if you look at it in the last two, three years, they were so focused on go-to-market because it was all about like, I need to build the brand. I need to be the first mover. I need to be the trusted brand and legal, trusted brand and real estate.

19:13Zeynep Yavuz-Willson:I need to build that. And now that that has kind of happened and customer support and a few other categories, the next step is, OK, now I need to go deeper in my technology stack. Because why? I have this threat that's coming from potentially the model companies. And two, I have this tailwind of resource. And three, I have all this data that's being generated on my platform. so now that you have some of that scale and you have that data at scale it's the right time to actually start thinking about how am i going to feed that data into my own models and potentially even deliver a better experience for my customers than if i would if i were to be staying on a closed source or um in a frontier model yes it's using the head start that you've got to capture the data use the data that you've got to improve the models and maybe you're building on top of the the frontier yeah interesting on open source you have the data you're generating the data i think up until now they they were just not focused on it and i think it was a mix of like the frontier models obviously are the frontier model so they are very good at what they do but i think cost is becoming now a topic as well and and i think now the taking capabilities in these applied AI companies they're also saying like i actually think i can build the right model for my customer because I know my customer best.

20:30Zeynep Yavuz-Willson:I have the right data to fine tune. So I have to go down that route. I think it's really interesting, the point that you made that it's the application layer company, especially that you've invested in, but where we're seeing most of the value is one that sits between two people, either people staying in properties and their property managers or lawyers and the people they're interacting with. I think there's something really interesting in that, that it serves as the layer of communication between two people, which I'd never really thought about, but it's interesting to think like what are the other where are the other products where are the other communication layers in the ecosystem where people should be building or are building that we haven't seen yet you know what other because it's all about the network effects right yeah this is where you end up having network effects like uh legora was a good example where it's the law firm and the legal customer right because then uh the in-house effectively let's say then if you are on Legora as a law firm and you say you tell your customers I want all of you to be communicating through Legora or why haven't we seen the same thing in accountancy or have we have other companies like that you know we've got Legora we've got Harvey but we especially in the UK we have this huge history with the big four accountancy firms we've got these amazing accountancies firm we haven't there doesn't seem to be a player in that space that has grown to the level of a Legora or Harvey.

21:51Why do you think that hasn't worked so well?

21:53Zeynep Yavuz-Willson:So I would say some industries definitely had faster escape velocity compared to others in AI adoption. And we've definitely had that in Eagle, but a lot of the workflows are tech-based. We've had that in customer support where a lot of the workflows are voice-based. Then you look at something like accounting and there are, and I've studied this in each European country pretty deeply. And then obviously we invested and integral that is doing this transformation in Germany. And you've got a few things that make your AI transformation job quite difficult. One, you need to work with the local accounting platforms.

22:28Zeynep Yavuz-Willson:So you could attempt to rebuild the ledger. So in Germany, for example, you have DATEV that has pretty much 100 % market share. But that's like building another business to try to disrupt DATEV. And also DATEV has a lot of trust. it's actually owned by the accounting firms themselves. So the accounting firms themselves own this accounting technology. Imagine that you don't want to replace that. Yeah, yeah, it's interesting. So you want to partner with that, which means that you need to build an automation layer that's on top of effectively a system or record that already exists and used by everyone.

23:04Zeynep Yavuz-Willson:By the way, those integrations that you have and the depth of those integrations, I believe will be another mode in this AI diffusion. Then the second thing is that in accounting, you've got so many edge cases. And this is not just text-based workflows, right? So there is quite a lot that you need to do from a quantitative side as well. So the automating that is actually a lot trickier. And there is literally no room for error. Because you are doing something legal, you're doing a legal filing, and there cannot be any room for error. And the other thing is that, again, I'm coming back to this, the level of fragmentation in a market slows down adoption of AI.

23:43Zeynep Yavuz-Willson:The more concentrated the market, the AI adoption happens faster. Why? Because the decisions are actually driven top down. If you think about how a lot of these customer support solutions or legal solutions that are bought, it's boardroom decisions. When people use it, obviously they're happy with the tooling, but so much of that is happening. And I think this is why diffusion of AI has been super fast in some of the verticals is because the decisions are happening at the board level and it's being pushed down by the CEO and the chief. There's always an AI strategy that's being discussed in the boardroom.

24:16Zeynep Yavuz-Willson:And that makes sense. Therefore, particularly in-house functions like customer support, Regal has had such high adoption. If you think about SME accounting, like who's driving that change? We need to do that change bottom up. And that's much harder to do. Yeah, that's really interesting. Okay, so if we look at the European market more broadly and we look at this idea that, or there's varying degrees of diffusion or adoption across Europe, what do you think is unique about the European market where maybe we have an opportunity to adopt or diffuse AI faster than Americans? Maybe it's because we have...