Founders in Arms
00:11:05 · Rajat Suri, Jean-Denis Greze
[00:11:05] S1: we had couple 100 users. And and how is it today? How how are we scaling from a couple 100? It's February this year, right, we're talking about? Yeah. This is this year. [00:11:15] S2: That's right. It's less than twelve months. Yeah. Well, we launched Vermeil in June. And... Yeah. I mean, we don't talk about the numbers too much, but we're... We've added, over a million each of the last two months in terms of ARR. It's tens of thousands of users. It's paid, so you have to pay $1,549.99 or $1.99. So, you know, it's like... It could be a pretty expensive product, but I think it delivers enough value to be worth that. It's still mostly word-of-mouth, although we started marketing, which I'm pretty excited about, paid marketing and influencer marketing and all the things that you do. But mostly, it's still people try it. 85% of people don't like it. 15% of people like it enough to pay. And then those people tell someone else about it, and and we make it easy for them to tell someone else about it, and then they they bring more friends, and it grows. Cool. [00:12:05] S1: And and so I'm new to it. Myself, looking at it before we had this talk, and, like, what... You know, as someone who is, like, already using CloudCheck, TBD, why would they use Town?
00:12:15 · Jean-Denis Greze
[00:12:15] S2: I mean, to... Today, I would say the the main trade offs are, first, it can do much more and it's much more proactive than those two products in the market. Number two, it's pretty simplified for the mainstream user. So I would say, like, Claude Code or Claude Code can do most of what we do. But you you kinda gotta know what skills are, and you gotta know what connectors are, you gotta set up your routines around email and around meetings. And most people don't wanna do that. They just want something that works out of the box. So that's... The main reason is our users are not using any of these tools, and they're busy. They're smart, but they're very busy, and so they don't have the time to do all the setup that's necessary. And then what's interesting and funny is that we have users at companies where they built their own version of Town, and they wrote me an email like, hey. My version is better. And then two weeks later, they write an email just like, hey. I'm tired of, like, upkeeping my thing because it takes real effort, right, to keep a product at edge, then they start using our product. I think it's easy to look at these things and be like, oh, I could build that. Because you could build that. You could build it in a week, and it would be more custom to you, and you might like it better. But then because the technology's moving so fast, you don't you don't wanna spend every weekend making your little, like, personal AI better for you. So that's a big part of it. Then there... Sorry. There's, like, meta talk. Yeah. So we sell to businesses. So it's PLG, but then we sell to the business. And I think that some points that resonate with businesses because they're true is, look, two years ago, if you were a business, you would have bet on OpenAI, and that would have been a bad bet. If you're coding, you should have bet on Claude. So for us, it's like, look, no one knows who the winner is going to be at the labs, and so you probably actually want the interface or product provider for you to be lab neutral because I don't care who wins. I will use the models of anybody that are right for your workload, that are the right price, the right level of intelligence. And so that argument, when you go to mid market firms, like 500 to 2,000 employee firms, like, it really resonates.
00:14:19 · Jean-Denis Greze, Immad Akhund, Rajat Suri
And we have a lot around how we how we treat data, how we let the business own the data, like, how we are model neutral. That that is, like, meaningful. Are all buyers gonna buy into that? No. Probably not. But it's such a huge market that there is, like, a big chunk of it where you can you can win with that kind of differentiation. That's structural. Right? OpenAI, I mean, it could, I guess. But they're... I don't think they're waking up anytime soon and, like, you know, peddling cloud models to their users. [00:14:47] S0: Well, Grump did this thing where they are peddling Anthropic and OpenAI, which is [00:14:52] S2: very Well, they need money. Yeah. But that's because they make money on the Anthropic models. Right? Because the Anthropic models... I I... My interest... I don't know. My guess is that by [00:15:01] S0: the way. Yeah. [00:15:02] S2: They're selling anthropic models that are served on their data centers because they have the deal. And so I think they make money they make money on that immediately. [00:15:11] S0: You seem kinda relaxed, but, like, I feel like... Like, how does it feel to be at the center of this kind of... You know? Yeah. Like, I feel like everyone is, you know, instinct, muse, rock. Like, it's... It seems like the center... I know you're saying, like, oh, this is, like, different, but, you know, a lot of these people will also try to go off next [00:15:31] S1: to me and then. Right?
00:15:31 · Rajat Suri, Jean-Denis Greze, Immad Akhund
[00:15:31] S1: to me and then. Right? [00:15:32] S2: Yeah. Totally. I mean, you're, like, it's the eye of the storm. Look. In life, you're either irrelevant or irrelevant. Obviously, it's great to be relevant and alone. I don't have that luxury. Yeah. [00:15:41] S1: That would [00:15:42] S2: say I have the luxury. [00:15:42] S0: I feel like Plaid had that for a really long time. Plaid was [00:15:45] S2: almost alone. There was a lot of competition for Plaid, [00:15:47] S0: but I think They all sucked. [00:15:51] S2: You. Yeah. They would... But there's also... You know, Plaid has, like, a consortium of banks as a competing product with some place place dark. So [00:15:59] S0: That's true.
00:16:00 · Jean-Denis Greze
[00:16:00] S2: It was tough. Like, Plaid was tough, but, eventually, we got to network effects, and that's why... Honestly, like, Plaid that the network effects was winning because we were better, but it felt like we had to keep winning. I think as soon as we got network effects around people connecting the the app and the friction we could remove out of the flow, like, then the business really starting to roll because the conversion improvement was high and couldn't be replicated either by the bank's network or by third parties because we had most of the demand. But to answer your question, the way I look at it is it is extremely competitive, for sure. Everyone knows that. Why is it competitive? Because it's a huge market. It might be the largest market ever. Right? There will be other larger markets, but there's billions of people that will use assistants for work. And as the assistants can do more and more, they're not really assistants. Not even really chief of staff. They're like coworkers that can automate more and more work so people will spend more and more tokens for work to be done by them. So it's a huge opportunity. I never woke up. Like, when we decided to go all into this, we knew that Microsoft and Google and OpenAI and Anthropic and whoever else you want to put on the list is gonna be a competitor. That's the truth. You have to ask the question, how do you build something that is different from a marketing, a feel, a product perspective so that in this giant world of users, there's set of people that are gonna prefer your ice cream. And my ice cream flavor may not be vanilla. Everyone loves vanilla. You know, it might be like, you know, mint with, like, cherry on it or something.
00:17:36 · Rajat Suri, Jean-Denis Greze
[00:17:36] S1: Red leaves and cream. [00:17:38] S2: Well, that would be good. People people like that. Yeah. Exactly. But it... And so I think the the way we think is, like, we have to we have to zigging while these are. It has to... We have to end somewhere different. And then the final part, which... So what I would say about you and Mercury is, look, we are... There's some things in our product, like email and calendaring, that I guarantee you we're spending more time than everyone else. And we're very focused on email and countering, and it comes out of the box, and it is very good, and it will only get better. And I think there's people who really want an assistant that day one is really good at email and countering, and I will pay... Like, our experience on that will always be better. We will think about the details, and I think people notice. Right? And over time, you can't do all the things, but you can pick some things that you're great at. And if there are people in the market that really want what you're great at, you you can win. Right. Right? You can win at that. And if you just