SaaStr 2025 Partner/Workshop Stage
Five investors debate what makes vertical AI startups fundable. They want founders with insider knowledge competitors can't Google, revenue from actual customers not experimental budgets, and a long-term vision even if early TAM is small. Pick a vertical you can stay in for a decade and prove AI actually works.
Chapters
Please take your seats and silence your devices. Please welcome to the stage CEO and founder Runway, Siki Chen.
All right, thanks for coming. Who's excited to talk about financial modeling in SaaS? All right, all 10 of you. Love it. Okay, so I'm Siki. I run a company called Runway. We're an FP&A platform for data-driven finance teams. We currently help a little over 200 companies simulate how their business actually works. These include companies like AngelList, MyFitnessPal, Superhuman, etc. Um, and over the course of building this company over the last couple of years, obviously we've seen a whole lot of finance models, all stripes. And, uh, we've never met a customer who doesn't come in and feel embarrassed about their finance model. It feels like, oh, it's super complicated. My data is a mess. And that's pretty common. But a finance model usually is technically correct, right? So you have like accurate KPIs, it's GAAP compliant, it's accrual-based. We also found, and this is pretty common with every finance model, is that there generally, there's a lot of frustration because these finance models are so accurate. And so clear, but they're still ignored by the people who need them most, the sales counterparts, the marketing leaders, the CEOs. And we spend a lot of time building software thinking about how can we fix this problem. How do we make it so that other people who have a stake in the company and in finance actually understand and care about the financial model that ultimately drives the business. So, It's not that there's anything wrong technically with the model. It's going to be accurate most of the time. It's going to reflect the right forecast and the right actuals. But the main problem is actually disconnected with the key decisions that a company makes in a business. And when you think about a model, one of the ways that we think about what a financial model represents is it's easy to think about a financial model as just a spreadsheet. Right. And it's so much more than that, as we know, if we built a model before. The way we like to think about a financial model is as a simulation. My background is in games. I used to run product at Zynga, and one of the more useful analogies for what a great financial model does is it's like a game engine. So if you ever played a video game like RollerCoaster Tycoon or Civilization, there is an underlying simulation that simulates a part of reality. It's an abstraction of reality. And what great simulations allow you to do, especially in finance, is effectively to allow you to time travel into the future to understand the decisions that you might make today and what is the impact of it in the future. So how do you build a great simulation like this? Um, when I think about a game engine, in the form of a simulation, I think they have 3 parts. So if you ever played a video game, you have controls, right? If it's Mario, you have the joypad. You can go up, left, right, down, and you can make, make Mario jump. Then there's logic. What's going to happen when Mario moves? What's going to happen when a Koopa hits Mario? If you increase ticket prices in your RollerCoaster Tycoon park, what's going to happen? So those are what controls are. Then there is a logic. And then there's feedback. And feedback is, okay, you're— you pressed a button, you've increased prices, you put a new attraction in your RollerCoaster Tycoon park, and there is some simulation that it affects. Then you can see visually what happens, right? There's more people at your park, maybe. Or Mario dies. You get some kind of feedback. Great simulations have all three of these things. And so when I think about building a great model that actually gets used and people can understand, it starts with controls. What are the actual decisions that you get to make in a business? Is it hiring? Is it a product roadmap? Is it product decisions? Is it hiring decisions? Is it a marketing campaign?
These—
this is the most useful starting point because simulations and great models aren't just to be looked at, they're to be used, right? You want to use them to refine your thinking. So one of the mantras we have at Runway is finance isn't just accounting but in the future, right? It is also accounting. You want to understand your budgets and your forecasts and your revenue goals, but it's also about something more than that. It's about thinking. So developing intuition for the business, understanding causal drivers of it, Starting with the decisions is what helps you develop that intuition and get the rest of the company bought in and understand the impact of the decisions that you make. Oops, I'll go back. Logic. So obviously once you have your controls, you want to wire them up and see how your controls will affect the KPIs you care about. This part is relatively straightforward, but for SaaS companies, and technology companies in particular, there's a really powerful tool, which is we have the ability to build technology to increase efficiency. So one of the more useful things is to identify what are the costs of scale to different things. And generally, there's two types of things that scale. One is cost of scale with people, right? If you have more people, you're gonna need more G&A, you might need more SaaS. So that's one type of scaling. And the other type of scaling is the type of scaling comes with as you have more customers and revenue. You might need more engineers or more customer support people or more account executives. And the wonderful thing about technology is that it is leverage. It is leverage in a real sense, right? So if you don't have a shovel, right, and you dig a hole, it's going to take a while. If you do have a shovel, you get to dig more holes or dig the same number of holes with fewer people. That is a key leverage point. And so identifying the ratios of how your different costs scale is critical because then you can model the technology impact on these ratios. So for example, as your product gets better, um, you're a typical customer support person, maybe able to support 30 customers per month or 40 customers. But as the product gets better, as it becomes more PLG, as your documentation gets better, Maybe that 30 or 40 becomes 50 and 60, and it has a direct impact on efficiency and profitability. For account executives, as your product gets better, maybe you start with, typically you can close about 10 deals, but as enablement gets better, as you develop new AI features, that ratio increases. So understanding and modeling these leverage points, these ratios, is super important because then you can connect these ratios to the things that you do in your product and the decisions that you make in your business. And of course, then once you have these controls, you have these leverage points, then you can simulate how it affects the downstream KPIs that you care about, whether that's revenue, margin, profitability. This is how you make it directly visible. And you kind of repeat the cycle until your assumptions get tighter. So just like any simulation or any software simulation, you're not going to get it right the first time. It takes some iteration. And what you want to do is as you execute, you're going to see where you're correct and where you're off, which assumptions you thought was going to be X and it ended up being Y, which relationships you thought drove retention or conversion or new AR and actually didn't. And then you can adjust the logic and change your simulation over time. One of the examples of controls that we have, so I'm kind of a model nerd. When I ran product at Zynga back in the day in FarmVille, we created dozens and dozens and dozens of these models. But we did a really good job of exposing controls that tied directly to what we do day to day. So we might have a decision between, hey, we should do a big Christmas event. And if we do a big Christmas event on Christmas, it's going to lift engagement by 20%. And some percentage of that 20% is going to stick around. And then we had another decision, which is instead we could have improved a tutorial at the same time, and that seemed like it had a smaller effect. It's about a 2% lift in retention. But if you have a right simulation, you can get a better answer what's going to happen in the future. And as it turns out, over a 6-month time range, that lifts revenue by 15% because a 2% retention compounds. And so when you have the right simulation and right controls, you get to make better decisions. And over time, you can make better decisions in your day-to-day because you have developed an intuition as you use your model and play with your model sufficiently. And so when you have the simulation and you have this model, it becomes the interface into your business. You can prioritize features based on leverage. And then over time, you monitor your forecast versus actuals weekly. You adjust your model, adjust your simulation, assumptions, and then you can even adjust your hiring plans based on how effective you were at delivering a roadmap and delivering an impact. So how do you actually go about start doing this? We have a very easy, quick 3-step process. Uh, first step is identify the controls and the different KPIs and leverage points of your business, right? What are the key decisions that you actually make week to week, month to month? Define the leverage points of your business. What are the most important ratios that impact efficiency and your ability to grow in your business? And 3, outline the outputs. And for SaaS, usually this is going to be margin, CAC, new AR, and you connect all 3 dots. Once you have that, you connect it to real data. So everything that happens in a company eventually falls to the bottom line somewhere, whether that's a new hire, whether that's a new customer, whether that's a new product feature, eventually it does something to the business. And usually the things that you do are mapped to sources of data. And by connecting the sources of data, you have a starting point, you have actuals, and you have a basis to compare as you execute over time. You define your controls and your logic, and then you build the actual leverage ratios and you manage towards them. And step 3 is you make your feedback a habit. So on a weekly basis, now you review your actuals against your forecast. When you miss or you overperform, identify why. You can drill in to understand the causal controls and see if it's a logic that was wrong or the assumptions that were wrong. So what this looks like on the other end of it is the following, which is before, and we've seen this a lot of companies, Every department has their own model. The sales department has their own model. The marketing department has their own model. None of them really talk to each other. And if you can build a useful simulation with the right controls, what you'll find is that you can have everyone use the same simulation, same source of truth. And now everyone's looking at and understanding the impact of decisions you make using the same framework. So tradeoffs become easier and it's easier to align. The second thing that happens is, imagine you're in a meeting and someone asks, "Okay, what if we push back the roadmap by 3 months, or we pull ahead this marketing plan by a month?" And usually that takes days, but if you have the right controls modeled, you can answer them live in a meeting and just say, "Here is what's gonna happen." And that is a great way to get promoted, usually. And final thing is, the people that don't care about your model now cares, particularly the CEO. The model becomes, instead of the spreadsheet, it becomes a single source of truth for your company actually thinks. And now the whole company can play the game. That's it. Thanks for coming. And I'm ready for questions if you have any. Any questions? Yeah, you can come down to the middle if you want one.
Yep.
How do you, um, how do you handle, like, um, especially like in technology investor environment?
Yeah, yeah, very aggressive models and very aggressive goals. And then that's—
then of course, you know, you still have reality that you're living in.
So you want to think about, okay, what are the, like, the realistic expectations?
So just kind of like, how do you, how do you handle that balance? Because I found myself at times having to keep up with like 2 or 3 different models that are like, okay, CEO for investor, actually running the business.
Yeah, uh, version control is a big issue, and so typically you will end up with a couple of models. But just like when you play a video game, there's like different possible futures, right? You can put in different assumptions and you can maintain a, you know, base case, a high case, a low case, and you get a range of possibilities. And, you know, not shilling Runway particularly, but there's different platforms that you can use to make that version control and scenario management a lot easier. The problem with— and I've actually— this is the reason why I started a company. In 2020, I was running a company called Sandbox VR, and then COVID hit. And so we were literally being triaged by Andreessen Horowitz on who's going to like get the emergency funding because no one else was going to fund any companies. And we had to create these scenarios, whether COVID was going to be 3 months because Elon said it was going to be over by June or a year or 2 years. Andreessen was the only fund or people that told us it was going to be 2 years, and they were spot on. We ended up with these like dozen spreadsheets. And when we update one, it won't update the other. And so having good version control is actually really important to keep it clean. And there's tools for this. I saw another hand raised. Hello.
Hi.
Hi.
Typically, how long does it take to create a model and see the outcome?
Yeah. Do you already have a model?
No, not really.
So what you described with everybody doing their own kind of thing, siloed, that's what we have.
Yeah.
Getting it all together, and typically how long does that take?
Yeah, I mean, we've seen people get started even before the first kickoff meeting. So we have white glove support, but you know, it's a version— what we build is a version of Notion basically that does math and connects to your data. And so having a consumer-grade experience is really important. But I would say on average, it's about a 2 to 6 week implementation depending on the complexity of the business.
Thank you.
Yep.
Yes, sir.
How do you handle transparency of your financial data to the rest of the company?
That's a great question. So, some context. This is my first B2B SaaS company of any kind. My entire career has been in consumer software. And there's so much to learn about B2B, and I really enjoyed it. But one of the more useful things is being able to take learnings from the consumer world to the enterprise world. And so one of the stories I think about is, I think it was around 2005, 2006 or so, when Facebook was just getting started and they did a newsfeed. And their main competitor was MySpace, which was about 10 times bigger at the time. And they famously observed that the more control they gave users over who got to see what, the more users would share. Right? So MySpace is like everything's public, people will just post GIFs or whatever. It— Facebook at the time when they had a news feed, they, uh, let it— gave you very fine-grained control so you can share just with specific groups of people or just your friends or the entire world.
Right?
And the more fine-grained that control is, what they found in their data is that more people shared. And so if you think about, um, a spreadsheet, right? It's kind of all or nothing. Whether it's Google Sheets or Excel file, you have to share it all because all calculations on both those products are done on the front end. And so what's really important is to be able to have very fine-grained controls, right? So for example, even if you wanted to share an entire spreadsheet, you can't because you can't just hide or encrypt a single column called salaries, right? So what we do is we give you very fine-grained control, whether it's column-based or row-based. So you can say, I want people to see everything, except for salaries. And what we found is that if we give people the control over, uh, how much of every— anyone in the company gets to see, people end up sharing a lot more. So one interesting, uh, thing that we've seen is, for example, kit.com, used to be called ConvertKit, is one of our customers. And what they did after, uh, they start— got started is they did an all-hands where they give gave everyone access to Runway, the entire instance, because they were able to high salaries and they did a scavenger hunt because they implanted some kind of bug in a model and broke the, the company into teams to find the bug and everyone was successful. And so having extremely fine-grained controls is how you do it. Some companies might not want to share anything. Some companies might be completely transparent and having the ability to control where you want to be is really important.
All right.
Thanks for coming. Really appreciate it, guys.
Please welcome to the stage Vice President of Product Advocacy and Market Strategy, Make, Darren Patterson.
Good morning, Saastr. It is excellent to be with you again. You know, for the last 15 years, I've been working with companies both large and small to bring about the future of work. That means having employees that are engaged in meaningful processes that know that their work matters. And it means companies are efficient and effective in being able to stay competitive. Most recently, I get to do that at a company called Make. Make enables you to automate and integrate some of your core processes in your business in a visual and powerful way to accomplish the best you can be as an organization. And in my role, I'm incredibly lucky to be able to learn about, engage with, and collaborate with some of the world's most amazing companies. I've recently been obsessed from learning from founders about how they do things that are differently in order to get extraordinary results. Recently, I've learned about companies like Guideline, a company that is democratizing retirement, especially small businesses and their employees. I've also gained a deep appreciation for companies like Snyk, a developer-focused platform that enables you to secure development processes by ensuring each of your engineers is a security guardian. And I've been absolutely fascinated by up-and-coming AI-first companies like Lovable, who are radically challenging the need for companies to hire developers in the first place. Now, in some ways, these three companies, they're incredibly different. Each one of them is operating in a different industry with different lifespan and different business models. But what do they have in common? Well, I can tell you they're all incredibly successful. It's not too often we actually come to Saastr and we talk about mediocre companies, right? But actually the way that they achieve success is incredibly important. You see, each one of these phenomenal companies has achieved success through a very specific approach. An approach that puts them within the 95th or even the 99th percentile of all similar companies on a super employee, important metric. That's ARR per employee. Now, one small caveat here, uh, Lovable, they're not quite at $100 million ARR just yet, but it's gonna happen in no time. They actually achieved $30 million ARR in 120 days with a team of 18 people. It's incredible. Now, while this ARR per employee metric, which actually is all about efficiency, often gets overlooked in favor of other SaaS financial metrics like your ARR total or your ARR growth rate or customer acquisition costs and NPS, I'm here to tell you that efficient growth has never been sexier. It's no longer the case that you can build a product, test for product-market fit, and then get some investment and just hire like crazy to see how fast you can get the rocket ship to grow. Yes, investors are actually looking for something different now. Not just companies that can reach that elusive $100 million ARR mark, but it's looking for companies that know how to grow. In fact, it's kind of counterintuitive, but in fact some would even say that it's easier to grow with less people. Think about all the time, effort, and energy that actually goes into hiring more and more people. Not only the hiring processes, but how many calories are spent on all adding those new management layers, decisioning processes, propagating messages, meetings, and lots more meetings, I guarantee that. All of this energy could actually be spent on understanding your customers and innovating on your product more effectively. So what does it mean to grow efficiently in practice? And as a result, of course, how do you become the world's smallest $100 million ARR company? Engaging with and researching these companies, I can tell you they have some very distinctive types of cultures, key technologies, and business practices that enable them to grow like this. Of course, it all starts with the basics, starts with the fundamental piece of their culture. Culture is how you think, and how you act in order to get things done. And it starts by the type of people you surround with this, surround yourself with. Obviously, every single company has a distinctive culture, but it's incredibly important, especially early stages, to be very, very selective about who you hire. The co-founder of Lovable actually emphasizes that they believe in hiring super high-agency people that are very focused technical talent. This also means planning on fewer management layers where even more senior hires are player coaches that can actually do their share of the IC work. Now, second, as a core cultural element, these companies embrace experimentation. They're never satisfied with the status quo, and their employees are constantly trying out new ideas. They're constantly tweaking and refining to question everything and continually improve. And if you're serious about about absolutely having a culture of experimentation and innovation, it means you're also serious about embracing failure. Failure actually increases learning, and learning faster than anyone else actually enables you to have a significant opportunity for differentiation. Guy Pagiornia, the founder of Snyk, he actually reminds us all that we actually probably overestimate the extent to which people in the past have actually succeeded and how much those people actually knew what they were doing. It's comforting advice for all of us. Now, we also talk about the technology practices that these companies have. And in 2025, and especially at this conference, everybody here knows what I'm going to say, right? Yep, you guessed it. Of course, this company radically embraces AI in all their internal and their external processes. They use it for personal productivity. They use it in their core processes, and they constantly evaluate what's new and bringing it into their corporation. But you can't actually deploy AI effectively without automation, preferably no-code democratized automation. In other words, unless you expect your employees to sit in front of Claude or ChatGPT or DeepSeek and conduct their work all day long in front of that, you'll definitely need to empower them with key methods of applying AI into their core business processes. Using no-code automation tools has the added benefit of supporting that experimentation culture. Let's face it, we don't always know where AI is going to provide the biggest bang for its buck, how to tweak it, and how to continually improve it within your core process. Now, if we think about what are the biggest opportunities for AI and automation, we think about all the things that AI is really good at and core processes in your company that need to be automated. That includes understanding what your customers are saying about your products online and be able to respond respond to that analysis, that sentiment. Being able to summarize market research and content, being able to quickly generate content at a scale we never thought was possible, being able to classify support tickets and make sure that they get to the right person, verifying all that content we create actually meets our brand or compliance standards. Transforming data is no longer nearly as difficult as it used to be. And of course, inreaching data to support core processes and customers. Lastly, not— lastly, but not least, AI is incredibly good at extracting the right type of information to be able to support your core processes. That means that you have incredible opportunities in very straightforward ways to be able to create highly personalized sales assets and marketing engagement, to be able to support customers like you've never been able to do before, or even support backend processes to validate contract or invoice data to run effectively. Now, we've talked about what are the core kind of pieces of culture, as well as what is the core pieces of technology that they use. But they also have a very specific approach to business practices. And first, that's the employees in these organizations routinely demonstrate growth mindsets. They have a core belief that their smarts aren't just fixed, that they don't know everything, and that you have to work out your brain in the same way that we all go to the gym, right? Every week and week, day. Yes. To work out your body. They know that they don't know everything and they're constantly embracing new things. And it's a very important part of their practices. The second major practice adopted by these super efficient companies is actually very difficult for, I suspect, many of us in the room, particularly if you're a perfectionist or if you think and dream about what you think the future is going to be. And what I'm here to say to you is to stop building future-proof solutions instead. And future-proof solutions often think about every edge case and meet every single expectation. And instead, it's super important that you apply this to both your internal and your external processes. And in everything you do, you ask, what's necessary to prove that this concept works? And in the smallest value and iterative way that I can test out that it's the case. So test your ideas, learn more, iterate, and then and only then should you think about scale. This allows for flex— fast flexibility, faster iterations, and ultimately more sustainable growth. Wow. I know that was a lot in a very short period of time, but let's do a very quick recap. First, it's important to recognize that high-performing companies don't just grow fast. They grow efficiently. The way you grow matters both in the short term and the long term. Efficient growth is the new cool kid on the block. Second, the best measurement for understanding efficient growth is all about ARR per employee. Some of the world's best performing companies boast incredibly high metrics, meaning they're lean, mean, SaaS machines. Lastly, companies that have a chance of actually becoming the world's smallest $100 million ARR business share key facets of their culture, their use of technology, and their internal practices. They focus on and being incredibly selective about the people they surround themselves with. They embrace experimentation and they definitely embrace failure. They also embrace technology rapidly. Naturally, they use AI. We all know that, but they embed it into their virtually every one of their business processes through a process of democratized no-code automation so they can quickly experiment and learn. At last, companies like this are very particular about their practices. They embrace growth mindsets. They don't assume they know everything. And they're constantly building small and worrying about scale later as they experiment and learn. I hope that you were able to pull a few pieces of interesting insight from this presentation. I'd love to carry on the conversation, hear your thoughts and perspectives. You can visit myself and the Make team at booth SG205, and I hope you have a fantastic conference. Thanks, Dastur.
Please, please welcome to the stage from People Data Labs, Ben Eisenberg and Lindsay Warren.
Hi everybody, I'm Lindsay Warren, VP of Growth at People Data Labs, and this is Ben, CEO at People Data Labs, and we're super excited to chat with you today about activation to expansion. Expansion, how the best PLG teams are staying ahead in today's climate. Ben, you want to kick us off?
Yeah, sure, Lens. Um, so when we're talking about PLG, let's see, our slides are— there we go. When we're talking about PLG, the thing that I think about is the user base that these companies have access to. The goal of PLG is to get as many free users as possible using your product and then to convert those users into a paid plan. and the world's most successful PLG companies have millions and millions of users that they're drawing from. So the natural question is how do they cut through the noise? Uh, and in the era of doing more with less, a lot of the innovation amongst go-to-market teams has really been leveraging their own data as well as external data to improve and sift through the noise of all of their free signups. This is something that's a full funnel exercise. A good PLG motion goes from signup and activation to first signature to expansion and retention, using data across that funnel to maximize the ROI of what they're getting from a given account. And so here's a classic PLG pricing page. Uh, you can see the free plan here on the left. You've got a team plan where users can go and attach their credit card, and then the real moneymaker, the enterprise plan. And the goal of any PLG company is to get as many users on that enterprise plan as possible. And if I'm thinking from a business perspective, It's really, really important to make sure that we get as many users as possible to that enterprise plan. And the company that we wanted to showcase today and talk a little bit about is GitHub. And the reason we wanted to talk about GitHub is they've got a really, really easy-to-understand PLG motion, from the sense that they've got hundreds of millions of active users using their platform, and they all really fall into one category, which is engineers. If you could understand the engineering headcount at any of GitHub's companies, customers, if you could understand what technologies they're investing in, what they're doing, as GitHub, you'd really be able to predict the value and the opportunity with any given account. And so the question that we are hoping to answer today is, how do you identify those users and those accounts that matter?
So let's dive in on that. Who are the users that matter? Uh, the best PLG teams are looking at both first-party and third-party data. This is gonna sound really simple, and I promise you that it is. You can use AI to do all of what we're going to talk about today, but we're actually just getting back to the basics. And this is what we've learned the best PLG teams are doing. So let's look at first-party data. We all have access to this. We have a PLG Motion. We're just looking at, you know, user velocity stats, right? And then we're layering in how many people have signed up at any given customer account. Who are my repeats? Who signed up? You know, how do we— 10 users at any given customer account, 50, etc. And then these teams are layering in third-party data from any source. Doesn't matter where you get it. There's lots of third-party data out there, and it's specifically how many potential users could I have at any given customer account? So let's dive into that a little bit more. The best PLG teams are focused on what is any given customer account's growth potential. And when they ask about growth potential and they're trying to figure that out, They're asking how well penetrated are we? Again, they're looking at that first-party data and then they're laying in that third-party data and getting even more granular and more nuanced. And this is all dependent on your buyer persona. It's different for every company. Um, at GitHub, they might be looking at what teams and departments in their, you know, if their buyer persona is engineers, um, and comparing that to overall total headcount. So we look at the, the screenshot at the top. We pulled this from GitHub's homepage. It's just their logo bar. And we took a look at Duolingo. So on the left, we have total headcount. And you can see that if we're just looking at total headcount, you know, Duolingo might not look like a high potential growth account for GitHub, right? Like, you know, hiring is kind of steadied off a little bit. But within their buyer persona, if we look at the graph over here on the right, their engineering headcount growth has continued to rise. So this is a really interesting signal and why it's important not only to look at overall headcount growth, but also buyer persona headcount growth. And so Ben is now gonna talk about the flip side of this. It's not only growth potential, but risk potential.
Yeah, and this is the thing as we talk about product-led growth as a full funnel exercise, the same stats and metrics that Lindsey was talking about are also being used to understand what the potential of an account is as they're continuing to work with a company. So you can take those user signups that you had before, and then as you're going further down the funnel and trying to understand the customer potential, you're actually looking at the activity of those users. Because if users are not actually using your product, obviously they're more likely to churn. You're also pulling in that third-party data. So as Lindsay was showing on the headcount graph, you're looking at the health of the business. So you're not looking backwards at the headcount, but you're looking forwards to understand if a recent layoff, recent hiring, or a steady state is an indicator of the likelihood of that account to renew or expand. And what we see the really elite teams doing is actually looking at that really deep into their buyer persona. So understanding that at the department level, looking at individuals within that department, including executive sponsors who might have impact or influence on a renewal, as well as champion tracking. So understanding if the champions who help drive drive the deal to close in the first place, if the champions who are the power users of the product are still at the company. Because if those people leave, that can be a really, really valuable signal for the health of the account, as well as for the health of other accounts where those champions end up going to work. And so again, we went back to our favorite homepage, the GitHub homepage. We found a really exciting company, Ford, and we found two really interesting executive movements that we wanted to highlight and talk about really quick. So the first is Mike Crabtree. He was sort of a more junior product owner at Ford, but then he moved on to what I would presume is a much smaller company in a much more executive role. If Mike was involved in using GitHub or his development teams were using GitHub at Ford and he had a good experience, it might make sense as an account rep to go after this new company that he's at and try to understand what their technology strategy is, if they're hiring engineers, and if they have everything that they need, you know, from a technology perspective. Inversely, Rahul here was the CTO of the commercial segment of Ford, and he left to go work at a very large company as well. But when a CTO turns over at a company, you really, as GitHub, probably want to think about if you still have the right executive sponsor in place. And also, even if every single executive is using GitHub, the plans that they make about their headcount has an impact on the value of this account to GitHub. Because if this account is going to change their engineering headcount, that is likely to have a significant impact on, you know, the value of the account to GitHub.
Okay, so I'm the VP of Growth at PDL, right? It's my job to understand our PLG motion as well. And after talking to all of these really, really great, uh, PLG companies, we decided to apply this to our own strategy a few years ago. So this is our pricing page. It looks very similar to GitHub's pricing page, right? We have that free a plan, we have a little small paid plan where you might attach a credit card, and then a larger enterprise motion. So let's dive in. These are the signals, very similar to GitHub's. We work with technical teams, uh, that are working with data at scale. So I'm looking at that same buyer persona score and trying to make aggregations on what customer accounts in my PLG motion have engineering headcount, and is that headcount At a steady state, is it growing, is it declining, and what signal is that telling me about this company? Our buyer persona is a little more nuanced than GitHub, which is why we don't have 100 million active, monthly active users. We're at a much smaller portion than that. And so I have to layer even more insights on top of this, which I won't go on about. It's our buyer persona, but it's, it's really interesting and allows you to keep layering and layering. The next score that we utilize is a customer utilize. Asian score, right? So just layering on top of that buyer persona score. And then I'm using this as the growth leader and marketing leader and in charge of the sales development teams to then kind of power and structure our funnel. So we use these interesting signals with first-party and third-party data to see, okay, what PLG email sequences do I wanna send? What's my strategy with that? And then what do I want my BDRs to be doing? I want them to go after the accounts that are not well penetrated using these two scores. And then that's also how we direct our sales team. So it really starts to get into most of the funnel, and we found it to be very successful.
Okay, so takeaways from this presentation. So number one, the best PLG companies are combining data. So they're taking their internal user data, their first-party data, their usage data, and combining it with third-party data to really holistically understand their customers and build a motion that goes beyond just what they understand about their customers. They're doing that by scoring accounts and users, and we highlighted a couple of different factors today that could influence a score. We highlighted how buyer persona, headcount growth, and departments could influence how you target accounts. We also talked about utilization and the first-party data that you can weave in to understand what your account is doing today. All of this gets used together to really build better signal, and signal helps you with churn, it helps you with growth, it helps you with identifying the accounts to go after. And all of that just comes from making sure that you're scoring every single account and making that score as accurate as possible. So yeah, if you want to come talk to us more about this or talk about how some of our customers are leveraging data, feel free to come by and visit us in booth SG216. And thank you so much for listening to us today.
Thank you, guys.
Please welcome to the stage CEO and co-founder, Propensity, Sumner Vanderhoof.
Hey everybody. Oh, thank you. Appreciate that. Uh, thanks for coming over and watching the presentation. On, uh, essentially like how to optimize your outbound motion, especially in the age of like anti-spam crackdown. So if you guys are part of a growth team or have responsibilities for revenue, you've probably seen click rates go down, call rates, like connection rates go way down, uh, but in a strange sense, like, there's been more calls than ever, there's been more emails sent than ever, so like every rate has gone down, even like the total numbers have gone down. It's in part because Verizon, Google, even AT&T are adding AI tools to combat the AI tools, right? They're putting anti-spam blockers in place. Google just yesterday announced that they're adding a bunch of features to block spam text messages. So if you're using text as a channel, that's going to get a lot harder pretty soon, especially— what were they blocking? It was like anybody that sends like the you know, when they're trying to like get you to click because you have an E-ZPass toll due and they're trying to get you to click on that. Or like, if you guys have seen the ones about Coinbase, if you have a Coinbase account and they're trying to get you to call the number, those types of spam messages are messing with everybody's flow and they're clamping down on that stuff. So it's making it harder for everybody. It's like a few bad apples are ruining the party for everybody else. So there's a pathway to still doing outbound right and getting the engagement that you're looking for. Without getting blown up by the anti-spam kind of rules and the kind of the spam filters that get layered onto everything. So this flow, the next couple of slides are just going to walk through what those steps are that we do at Propensity and how it works with our sales team and what the outcomes are for that. So the first thing is, whenever you're doing like any sort of marketing sales alignment motion, make sure that you're tracking people, not just companies. So There's been a big push over the last, specifically over the last year or so, to do contact-based marketing. So de-anonymizing contacts that visit your website, understanding the contacts that are clicking on your ads. So everything goes down to the individual instead of just the company. So what you're really trying to do is find the interested parties, like, so that you can focus your sales team around the people that are actually interacting instead of just working a cold list. Essentially give sales the warmest list you can. The phrase we like to use is like, anybody visiting your website is gonna be the warmest cold lead that sales reaches out to, so you gotta give them that information so that they can target the people that are actually interested in the product. Those types of people will actually raise their hand more frequently than the people that you're just hopefully, just working a list and hoping they pick up the phone. Multiple page visits, so scoring those types of actions as well, like a single website visit from an individual might not be enough for your type of business to denote warmth. So it might be like multiple page visits over the course of a month or something like that before sales reaches out. But the idea is scoring contact-level engagement at the website visit, the click visit, and then route those engagement moments to sales right inside their CRM. So before you send sales any leads, before you start as a salesperson working a lead, make sure that you have some sort of employment verification tool in place to know that the people you're reaching out to are the right people. It does no good if you deanonymize a website visit and you think they're at Company A and you think their email is whatever, you know, is, and it's an incorrect email, it's an incorrect phone number. You're just burning cycles. So you want as much accuracy as possible. So this is really a quality over quantity approach to outreach. Instead of just spraying and praying and just missing the mark 99 times out of 100 or 999 out of 1,000, you're looking to have more quality at-bats because you're reaching out to the right people. So anything that goes into the CRM system, anything that goes into the hands of sales should be pre-vetted and pre-qualified as a an accurate lead for them to go after. If you're in the sales team, basically what we wanna hear you say is like, give me more of this data because it's actionable and I can actually work this information. And then of course, like the easy answer here is like, provide an easy-to-use like working list out of the CRM system. Don't make them go to multiple systems. Keep everything where you can create an operational workflow out of it. So route the data directly into— if you're using Salesforce or HubSpot or Zoho or Pipedrive, whatever it is, feed the data into the system that the sales team is working on. Don't make them learn 2 or 3 systems or hop around. One of the things I like to do with our sales team is pre-verify not just the email addresses, but also the LinkedIn URLs. So like, and we put that information directly into Salesforce so they can click on the LinkedIn URL and instantly connect with somebody straight from Salesforce, and they never have to go to like Sales Navigator or any other tools to do it. So they're really efficient about their touchpoints. And that gets really important because we want them to personalize everything. We don't want them to reach out like in bulk. We, even a year ago, we were doing a lot of like automated sales, like using sales automation tools like Apollo.io and Outreach and different things like that. To do bulk touches across like multiple sequences. We've completely gone away from that. We don't use any like automated sales touch tools anymore. Everything's a personalized touch that comes through every rep on our team. So it's even more important for us that like every lead is qualified, every email address is vetted because they're spending time in their inbox sending those messages out. This is how you stay out of the spam box. Right? You're not sending 1,000 emails a day per rep. You're sending only the high-quality leads, the emails, you're only connecting with the high-quality leads. And we're not using an automation tool. It's not like a numbers game like it used to be even a couple years ago. Now it's about a quality game. So you visited our website, we vetted your email address and your employment, we passed that off to sales, and sales reached out with a very personalized touch within 24 to 72 hours. And now that engagement turns into a conversation. The other key piece is we're not presuming interest at the point of like, okay, just so you visit, because you visited our website doesn't mean that you're going to jump on a demo with us. So we're not asking you to take an action like that. What we're doing with our messaging is asking you to opt in. You didn't fill out a form on our website, but what we want you to do is like invite us to the party. So our messaging is almost always something like, hey, saw that you were on our website, do you mind if I send you more information about Propensity versus one of our competitors? Or do you mind if I send you this piece of content? Our team just recently released this tool or this, this widget to help you do like campaign planning. Do you mind if we send you this? So what we're looking for them is to say yes, please send me that, because that starts the conversation. So we never include attachments or links or like, hey, book a demo immediately. Nobody wants that in their inbox. But they will say, you know what, yeah, I am interested in that. I'm going to raise my hand to a sales rep that's reaching out. And that personalization really, really helps open the door, build the relationship, and start the conversion process. It's, it's funny being at an AI conference because a lot of this is like anti-AI, like the AI tool is used to help the sales team craft the personalized messaging, but they're clicking the buttons, they're doing the process, they're reaching out across the different channels to each rep. So this is how it helped us. So you can see this is over the last, like, almost exactly a year. So we really ramped up, like, our AI-driven automation touches middle of last year, and They were just like flooding the market with messaging, like probably a lot of your teams are doing right now. And you can see our average touch per sales rep has gone down substantially over the last year, but our booked demos per rep has gone up on average month over month. So we're doing less touches, like 85% less touches, and we've doubled our per rep booked demo because of this process, because we're only giving them the high-quality leads. We're only giving them the ones that visit the website or take some action indicating interest, and then we're asking them to like opt in before sales reaches out. So they're getting— the sales reps like it because they're getting paid more commission because they're getting more deals and they're doing less work to do it. And we love it because we're not like spending a ton of cycles trying to optimize 8-week sales sequences that don't work anyway. And then our customers love it because they're not like— we're really transparent from the onset. And we're not like just flooding their inbox and hoping that, man, we got the quantity game going. We hit the, you know, the email at the exact right time when they were just thinking about the thing. So it's basically helped us transform the way we can predict rep outreach and rep engagement and then focus on just like the market that's interesting. So now marketing's job on the other side of this is to produce more of those hot leads. So marketing comes to the table and says, okay, if we're gonna scale out our outbound side of things, we need to scale out the number of hot leads that we're giving them based on this metric. So everything they do now is tied into generating more website visitors, generating more clicks on their ads, everything that's trackable, and then we hand that off to the sales team. And then if that spigot stops for some reason, the sales team starts to, upset because this is where they make their money now. So they're making their money out of the hot leads that marketing's generated based on kind of the new framework of kind of the spam market that's out there and how everything's being routed. So hope that helps you guys. I think that's the last slide in the presentation. Yeah, that was my takeaway. So thanks for the time, everybody. I appreciate it.
Hello everyone.
Thank you for having me. I wanted to talk about how to build a GPU NeoCloud. I'm not sure how many of you aspire to build a GPU NeoCloud, so I'll try to reflect on Nebbius's journey over the course of the last 1.5 years or so and try to bring out the lessons that we learned, which we hope, and I hope, would be applicable to most of you whatever the stage you are at in your business. So we started with a relatively simple premise, right? How do you build a GPU neocloud? You hoard some GPUs and you hope for the profit at the end of the, at the end of the tunnel. The only thing you need to figure is phase 2. That was pretty much the plan with which us, Nebius, GPU stealing gnomes started the business. Um, when we started figuring out the phase 2, turns out it turned out to be way less simple. We had to build a great product, we had to bring it to the US, we had to create the brand awareness, we had to hire a great sales team, we had to automate— we still do— and we had to make the AI accessible. So step 1: build a great product. Turns out for us it was easier than for most because from our dramatic birth story we inherited one of the brightest engineering teams in the industry. So we started with making a complex product for sophisticated customers, which turned out to be simpler than building a product which is accessible for all. Our product was relatively recently ranked as, well, in the second tier of the industry ranking and were used by leading AI players worldwide. What have we learned? Well, lesson number one, play to your strengths. If you are great, your team has a history and that history determines what you're best at. If you're best at engineering, start with engineering product. If you're best at sales, start with sales. Lesson number 2: leverage your network for customers. Having a large network among the venture capitalist world and the Russian-speaking tech community in the Valley and beyond helped us a lot. We found some early adopters who helped us refine the product later on. And then lean into their needs. The way you arrive at a great product is that you listen carefully to your customers, you embrace especially the criticism, which is hard to hear but is the most useful type of feedback you can possibly get, and work hard to productize those, those things that they ask you to do to make them accessible to all of your customers. The next challenge was bringing the product to the US. Because of our history, we started off in Europe And then we quickly found out that most of our customers are in the US because US collectively accounts for 80 to 90% of overall AI investments. So we had to swim across the ocean to actually bring the product to our customers. The things we learned while doing so are the following. First of all, if you're going to enter a new market, start by spending ample time on the ground. We started spending time in the Valley, long before we decided to invest into our first data center. It took about a year. Challenge the constraints. We were told that you cannot sell overseas capacity to the US. Turns out it's not true. You cannot sell it for all the purposes, but you can totally sell training, training capacity located in Europe to the US customers. You need to just talk to them and challenge the constraints that are given to to you. And then when you commit, you have to commit with conviction. Once we made a decision to go to the US, it took us just a few months to open the first data center, which we did a couple of weeks ago, and we're already in the process of building a 300-megawatt site in New Jersey, which would be our second site in the US. Create brand awareness. We are in a weird situation that the team behind Nebius has worked together sometimes for decades. But Nebius as a brand is 1.5 years old. So we had the awareness challenge. People didn't know who we were. We needed to create this brand awareness, which we're still doing today, but the lessons we learned from that are the following. So first of all, when in Rome, do as the Romans do. You will probably not find a city outside of San Francisco where people try to sell software and, you know, tech products via billboards. But surprisingly, don't underestimate the power of billboards on 101. We actually had sales made by— made to people who saw our billboard and just reached out. Second, make it personal and own your story. We stem from a big Russian famous tech firm which in a dramatic set of events starting in 2022 launched us into the journey which we are today. Turns out that telling personal stories and the story of our birth, our own personal dramas, that worked way better to connect to people than just showing a brand book. And then most publicity is good publicity. Since English is not our native language for most of us, we made silly mistakes. Including, for instance, typos on our swag, which, by the way, made for great viral content because people took pictures of that and made fun of us. And that publicity is also good publicity because it builds awareness. Step 4, build a team. Since we're selling mostly to the US-based customers, we had to build a US-based marketing and sales team from scratch. Which meant going from 1 employee in the US in July 2024 to more than 40 people in less than 9 months. And we had to, we had the privilege of getting the people who had the experience working for the leaders of the industry, both on the hardware side for other neo clouds, for major hyperscalers and whatnot. The lessons we learned from there are these. First of all, if you can afford it, try to work with an in-house recruiters team. You can then ask them to focus on the specific profiles and on specific teams that you need. They will work in a way more targeted way than most external recruiters. Second, go the extra mile for the A-players. We wanted the best people in the industry. Guess what? The best people in the industry are expensive and demanding. Lean into their demands. They're worth it. The third lesson is be ready to change yourself. Research has shown that most organizations cannot grow by more than 30% year over year in headcount while maintaining their own culture. If you're preparing for explosive growth, your culture will change. Embrace it. It's inevitable and it's not a bad thing. You have to be open to, you know, merge the culture of people incoming with your own and arrive at something new. Automate. When you're building something, you don't know how it's gonna work. So the first sales, the first product outreaches, the first conversations are always gonna be done by hand. With time, you figure out the right way to do it, but then automate, automate, automate, right? The process should be you figure out the stuff you wanna do and then you automate it to make it as smooth for your customers as possible. And then also try to think where the puck will be because if you reflect on the state of the AI industry, it's in the state of overblown investment and underblown revenue. It will not stay that way forever. So try to think how the industry may change, how, what are the scenarios, and try to spend at least share of your resources to prepare for the future, not only to run your business today. Make AI accessible. There's no shame in borrowing ideas from others. So we do sometimes develop our products in a way which we've seen in the market. For instance, we recently moved also into inference pay-per-token type of space, which allows us to target the earliest-stage startups who just need an endpoint for a model and some, you know, market-leading open-source AI model to work with. What we learned from that is, first of all, it's okay to borrow ideas. Your product development does not have to be unique. You just have to do it better than everybody else. Tech in general is based on a time-based advantage. Meaning your product is as good as the amount of feedback and iterations which went into it. So if you start iterating faster than the rest, you'll get there. The important learning is don't cut too many corners because we did. The thing which we would not do today if we had to do it all over again, we would not make the inference product and the GPU cloud product entirely separate because that creates confusion among customers. And that basically is the— are probably the most generic things that we learned in the last 1.5 years, and I hope they would help some of you with building your own early-stage products. Thank you.
Please welcome to the stage Director of Product Management, PingCap, Xin Shi.
Hello everyone. My name is Xing. I'm the head of the product growth at PinCap. Before we dive in, a quick show of hands. How many of you had heard about TiDB before? Awesome, awesome. I saw several familiar faces. For those who are new to TiDB, a quick introduction. This is the one-stop scalable open-source database platform specifically designed to power AI and SaaS businesses. Today, I will share some of the real-world insights that we gather directly from our customers by working with them on scaling AI and efficiently minimizing infrastructure headaches and accelerating innovation. So first of all, AI isn't just an enhancement is fundamentally reshaping the future of SaaS. On average, the AI SaaS market is experiencing a 20% year-over-year growth, and the top 1% of the customers we've been seeing are growing even faster, often tripling their use cases and exceeding that kind of rate. Whether it's generative AI SaaS directly integrated into your business or is augmenting infrastructure of the AI, we are seeing explosive growth and adoption. So next, this remarkable growth means your total use base has been doubled, tripled, and your data volume could grow even more significantly. But this impressive explosion comes at a cost. While many amazing discussions we have heard about today is about GTM, POG, everything, You know, no one talks about like how to build an infrastructure for your SaaS business. Your infrastructure actually the critical foundation to support it all. Poor infrastructure choices can amplify security risks and inflate operational cost and even create severe, you know, bottlenecks for data handling. Ultimately becomes the bottleneck for your business while you are growing. We have see, uh, we have seen, uh, 4, um, you know, major challenges, uh, and these challenges are not, uh, not just, uh, theoretically, they are driven with different behaviors and the patterns from the tenants in your SaaS. Some tenants spikes on compute aggressively, others store massive cold data without touching it. Some of need fine-grained security controls and the other abuse shared resource unintentionally. These patterns can push your system architecture to its limit. So in the next few slides, we will talk about the each challenges and how we tackle them one by one with our customers. First of all, we talk about security. Security is definitely the priority. While we talk a lot of tenant security in SaaS, we are not just talking about logging and encryption. We talk about isolation between noisy and quiet neighbors to prevent abuse and overactive neighbors, tenants, and to support fine-grained and role-based access control tailored for each of the customer's needs. On top of that, auditability and compliance are crucial, especially for enterprise buyers. We have learned that security isn't just a trade-off with performance. It's designed for our requirement. You might hear from your engineering team, this and that might not be achieved because of the security reasons. But with the right database architecture, we can deliver both. So next, we talk about the traditional architecture with dedicated resources per tenant and is how the efficiency and overhead for the maintenance. But a bigger issue is the margin. While you're merging with your POC motion, like 90% of the users come and go and you are provisioning peak capacity for all of them. How you handle them? We often see, you know, teams and provisioning resources to avoid outages and perform and to handle performance issues. But it is a silent— is a silent merging killer for your business. Shifting to a pooled serverless model dramatically increases resource efficiency and allows granular control. A database decouples the compute and the storage, uh, and with usage-based models gives you the tenants exactly what they need— nothing more, nothing less. It's only the way to scale sustainably. Many SaaS companies' platforms today are powered by Retrieval Augmented Generation, or RAG, If you look at, you know, so-called industry best practice, building a good RAG system typically requires multiple database systems— vector stores, SQL database, analytical, and you name it. What does that mean? In this practice, to serve a single retrieval request, we often see 6 to 8 different queries across different systems. Sometimes it's even more if you factor in, like, you know, re-ranking, enrichment, and all the logging. Not only is the architecture complexity, but it's super fragile. It creates serious challenges for performance and the maintenance inconsistency and gives you a challenge for your product team's creativity. That is why we invested in consolidating these modalities. Vector embeddings, SQL, JSON, full-text search, all of them into one unified database. And AI data types can be stored in, in one table. It's a game changer for building a fast, reliable, and scalable and cost-efficient AI-powered SaaS. So next we talk about the data. Data is actually the core value of your SaaS business. And also, it powers all the AI framework. But not all data has the same value, even though it might cost you the same to store and maintain them. Keeping everything in hot, expensive storage is not sustainable. So when that comes into play, that's why it's essential to, you know, implement layered storage within a single database. Hot data should drive real-time user experience. Warm data enables iterations and learning. And cold data retains the long-term strategic play of your business. All of this must live within the one roof to ensure your AI pipelines don't break along the complexity. Underlying, your storage strategy based on data temperature is not just a cost to play is actually the fundamental of the scalable and intelligent system design. So as a database vendor, TiDB, we have seen a dramatic shift in the past 2 years. Now over 30% of, you know, over 70% of our new databases on the cloud are created by either AI applications or AI agents. And what does it mean? That really means AI and SaaS is not a side trend. It's the mainstream. It's the new core of the cloud growth. So with that, our goal is super simple. We want to empower your infrastructure that scales as fast, as smart as your SaaS and AI. By working closely with our customers, we help them not just keeping up with the demand, but turn infrastructure into, you know, really competitive advantage. So here are some numbers from that work consolidating over half a million databases. We see a remarkable result— over 4 times faster feature development, 60% of the cost reduction, 80% of less operational overhead, and nearly 1.4 performance increase. Increase. These, you know, technical wins immediately turn into your sustainable business advantages. So a quick summary of the key takeaways to succeed in AI SaaS: treat data centralization as the competitive advantage. I think you should think about it starting from day one. Prioritizing multi-tenant security early, otherwise you will be blasted in the future. Embrace multi-modal database storage, otherwise you will see a growing engineering team, but seems like they are still maintaining the same user base. Automate operations before scale overwhelming. And you have to design your data tier around the serverless and usage-based economy. Reminder, at the last— in the last sentence is like scalability is not just, you know, system elegance, technical elegance, but is a sustainable business model for you. Cool. With that, you know, I'd like to thank you everyone for your attention. If you want to know more, I'd like to continue the conversation. Come to meet our team, uh, at the booth. Uh, we can deep dive the strategies. Enjoy the rest of the SaaStr. Thank you so much.
Itai Sudan, CPO Thrive, Reece Johnson, and Head of Business Development and Ecosystem, HoneyBook, Emma Radin.
[FOREIGN LANGUAGE] Hi everybody. How are you all doing? You got some lunch?
Cool.
Okay. So my name is Itay Sadan. I'm the CEO and co-founder of Duda. We are a white label website builder. We have over 1 million sites on our platform, and we service many hundreds of SaaS platforms that use our product to offer websites to their end customers. And one of the things that is— when I talk to product leaders, one of the things that is very common, and I ask them, what's your biggest challenge? And 9 out of 10, the answer is How do I grow sustainably? And today on this panel, we're going to dig into—
we're going to look at it from the lens of product expansion.
And who better could I have on this panel than the two following people who are very experienced SaaS operators? I have Emma Radin, Head of Business Development and Ecosystem at HoneyBook. And Reese Johnson, CPO of Thrive. So our goal in this panel is going to be to unpack strategies for product growth expansion. I want you all to come out of this session with real ideas, real learnings that you can implement in your companies to provide more value to your customers and through that generate more revenue in the long haul. So let's get into it. Reese, Emma, feel free to, uh, let's go ahead and introduce yourself shortly and your companies.
Excellent.
Go ahead.
Great.
Thank you. Hi everyone, I'm Emma Radin. I lead business development ecosystem at HoneyBook. For those of you who are new to HoneyBook, HoneyBook is a leading AI-powered business management platform for small businesses and primarily focused on service-based businesses. So think photographers, interior designers, marketing consultants. And these business owners, they don't sell products, they sell their time and their craft. And so HoneyBook is a platform to help them run their business. We started in 2013. We're based here in San Francisco, and our mission is to help small businesses be successful. Since then, we now have over 100,000 paying customers and have processed $12 billion on the platform. And one thing that's really exciting with AI now is we're seeing more and more 6- and 7-figure businesses of one. Over to you.
Thank you. My name is Reece Johnson, Chief Product Officer of Thrive. I've been a product guy my whole life, so Enjoyed working with, as running the product for a couple S&P 500 software companies. Had some 4 fun chapters in my career with all good strong exits, $2 billion, $6 billion, $4 billion, and $10 billion. So it's been a really fun career where we focus on Itay's question quite a bit, right? How do you grow, right? How do we grow? If you gotta grow, growing, is what is the premium of any entity. And at Thrive, we focus on small businesses. Primarily, we focus on trying to help them grow. And we'll get into that.
Awesome. So at a very basic level, if a SaaS company wants to grow, it has pretty much two options. One, you win, acquire more customers, either different customer types, could be customers in different geographies. The second option, which is what we're gonna dive into today, is offer more products to your existing customers, right? Create this more deeper integration, deeper connection with your customers. It's super powerful to be able to sell multiple products into an existing customer. And it's so much more efficient than chasing new logos. So kind of kicking this in, Reese, maybe you can tell us where did your company start in terms of the core product and how has the product played a role in your company's growth journey?
Excellent.
Let's hit the first slide then. Okay. So this is interesting, right? When I joined, The company had a product value prop, do it all small business. I mean, do it all. That's pretty broad, right? It's like, that's a little scary to me, right? The last gig I did, I was with Francisco Partners, right? And a very, very solid private equity company, right? Return on equity may be number one in private equity entities. And I remember meeting with the CEO, his name's Deepak, And he says, they really try to coach the operators. You know, I'm an operator. And he's, here's how to win, Reece. There's 3 ways to win. Focus, focus, focus. I'm like, oh crap. It's like, so we've adjusted. This is an adjustment that we've done just within the last year, right? Industry-leading marketing and sales platform. So that's what we focus on, growth, marketing and sales platform. Hit the next slide. And so we just take a traditional funnel. At the top end of this funnel is we help customers get found on the internet, right? Listings, websites, social, reviews, and we really try to get them clicks, calls, and impressions, right? Next slide. And then we just acquired a company, and I'll talk through the logic a bit about why and when, in a moment, but it was Keap. Hmm, $80 million company, and they focus on selling— so it's a CRM construct, but for us, most importantly, marketing automations. 42% of their customers did marketing automations. That's kind of what their main value prop was. So now we get a bunch of customers to come into a small business, and we start running these automations, and then What's grow? Nurturing those customers, getting them back, make sure they don't fall off, right? So this concept of how do you know when to grow? How do you know what— like expand your product expansion? That's going to be a fun one. We're going to get into that one in a minute, right?
Absolutely. So kind of, Emma, shifting to you, tell us a little bit about the product evolution at the HoneyBook, what really triggered the different shifts along the way?
Yeah, a lot of small businesses get into the business thinking they're going to be heads down on their craft and with clients. And the reality is they're spending all their time on the back end, administrative, and they're losing sight of why they started a business. And so our product is there to solve that pain point for them. We started in 2013 with a really simple problem. For the events industry, which back then was still using fax machines, which feels crazy, for invoicing. And so when we launched in 2013, it was to build a really simple, beautiful invoicing tool for the events business, photographers, planners, et cetera. And from quickly there, we just started to expand beyond and beyond to support them. So a payment solution to solve what was there was still a really big pain point back then around payments, which today is very commoditized. And the way that we think about growing our product is really understanding our customers and their needs and being focused on our customers, back to the focus, focus, focus, and building around their jobs to be done. So since then, we've really built this full stack, which we call their operating system, which includes an invoicing tool, contracting tools, payments along with templates and automation so that all of the stuff that they need to do to spend time with their clients can be handled more seamlessly.
Excellent. Reese, when, when we talked about it, you mentioned that the LTV impact of offering more than one product, kind of moving to the second product, is pretty material. Is there anything you can share about that?
Yeah, just real quick, the financial metrics around the second product for us The first product is less than 10,000 LTV. Second product was almost 4 times that, actually a little over 4 times. So huge difference when we get that second product. It's not linear, it is exponential LTV increase.
Super interesting. So the revenue expansion is, is really interesting, but there's one more advantage to offering multi— to going multi-product, and that's increased retention. And one of the things is as you start to sell more products into your customer, you become so much more important. You become the connection. It becomes a deeper connection. And one of the— I kind of brought a slide up here from our friends at Vendasta that I really like to use. And it shows how every additional product that gets added increases your retention. So the first product that you sell into a small business, after 2 years you have about 30% retention, which is not great. But with second product and third product and then fourth product you sell, you get to 78% retention, which is, you know, pretty dramatic, a pretty dramatic lift. So I know that that's also something that is very important to many of the, many of the companies that we talk to. It's not just getting— one thing is, is the LTV increase, but that higher retention is also super important. So the big question that becomes is, how do these companies figure out what is the right way to introduce, to expand? What is the priority? When do you go past that first core product to offering more products. Reese, how do you in Thrive or any of your previous companies, how do you determine when is the right time to add that additional product and what's your kind of mechanism to decide about priorities because there's so many opportunities to grow in different directions?
Hey, that's a fabulous question. I would start with saying when not to. So just so we know when not to get into When do you want to stay with what you're doing? If you've got a business, and let's describe an excellent business as rule of 40, right? So you've got growth and EBITDA added together of 40 and above, and you've got that asset and that's happening, especially if it's weighted more towards growth, you're a one-trick pony and that pony is awesome. And careful, don't distract yourself. You've got something really good going. Ride that pony. Right? Just keep going. Now, here's what, here's how I look at it's time to make a change. It's time to expand, as Itay said. When that growth starts slowing down. So that business on the right that I started with, that all-in-one business, we got it to over $100 million. Single-digit growth. It started slowing down. So that second product, the one on the right, where we added Marketing Center, which Denise is in charge of. It is some really awesome AI use cases. It's growing quarter over quarter, over 60%, Q1 last year to Q1 this year. So really fast. It's now our fastest product, over $200 million in ARR now. And the reason is when we try to adjust and move to, as Itay says, your second product, your next product, if your first one isn't is not Rule of 40 and it's slowing down, you know it's time to change. You've got to do something different to keep your excellent company a Rule of 40.
I can do priorities later. Yeah.
Yeah. And is there anything else you want to share about the decision-making? Is it when you choose that next product, is it something— are you doing multiple bets? Is it like a single bet? How do you kind of decide on where to—
Good question. So a lot of my career has been with the consolidators in the industry. So it's very inorganic is the expansion strategy is most of my career. And when you have cash, most tech companies do one of three things, right? They buy companies, or they buy back shares because they keep on giving it out to employees, Or they pay dividends. Most of the companies I've been with haven't had debt. So we've just been buying companies. So this question has really, really been on my mind a lot. So I came up with an answer, and I go, there's got to be something on this. As I prepared for this, and I started searching it, there's something called an Ansoff model that I found. I didn't know this, but that model validated my intuition like crazy. So every time— If you look at the Ansoff model, it starts your current market, and if you move horizontally, it's— you basically take that market, that buyer, and you add a product. It's called the product development strategy. And when you move horizontally in that model, that's the easiest and best way. Every time we moved horizontally, where we identified a buyer, and the buyer we naturally normally sell to and have that distribution strength, Most of the time it was a successful acquisition. Every time we went south in that model, new buyer, similar product, or diagonal, we failed.
Makes a lot of sense. Emma, I know you're at HoneyBook, your approach is more research, more community-centric in terms of how you decide on kind of the next products to build or partner. Can you elaborate on that?
Yeah, I think I'll cover first the partnership piece and then some of our research piece. So, um, I lead business development, which is our partnership team. Drew here, anyone who wants to talk partnerships can reach out to Drew. But, you know, where we don't build, we partner. And the reason for not building for us is expertise, time, um, and capacity, essentially. And so for partnerships, It's part of the holistic company strategy. You know, there's not a product strategy and a partnership strategy. There's a company strategy and everything ladders up together. And our partnership strategy is to work with other B2B adjacent tools and services that support and add value to our customer base. Because what we're seeing for small businesses, it's ultimately a Frankenstein solution for them, and that is painful. And so from our partnership approach, we're able to do some of the horizontals, integrate in, and make it a more robust center of gravity, as we say. Our goal is that our members can really use HoneyBook to run their day-to-day. And some examples of that, we heard from our customers, because we do hear a lot through our community and our research team, is Canva was a key tool for them to create assets for their brochures, their proposals, proposals, how they communicate with clients. And right now, until we did an integration, it was really, um, a crappy experience. And so what we did is we partnered to enable an integration, and that way our customers could easily upload from Canva and build out their templates and proposals through HoneyBook. And that delightful experience creates a lot of love with customers. Another recent partnership that we've done is with Uprise Financial. They might They might be here today. In this uncertain economic climate for small businesses, which is really top of mind for us and for small businesses, we did a really quick integration with them to make available financial advisors so that today that they can connect through Uprise and connect with financial advisors. Those are the types of approaches we're taking.
Cool.
Thanks for touching on that build partner buy decision. Obviously there's a lot of advantages from building. You have complete control, but you know, maybe it takes a lot of time and money. You can acquire, that makes it maybe a little bit faster, but then you have to deal with integrations, you know, a little bit of the culture chaos of integrating another company. And obviously you can go ahead and partner and and then you're leveraging expertise of a third-party company. But you really need to take care of, you know, this— how do you get like this tight alignment? How do you drive shared goals? So there's no like clear answer on what's the best between all these. Reese, anything kind of from your side on, you know, between the build, partner, buy? Like how do you approach those decisions?
Yeah, I, of course, I'm biased towards build and I've got to be careful with that because I like building. And so I have to know, you have to know your own bias, right? And 'cause you don't want a mentality of, you know, NIH, non-invented-here syndrome within the company. But from a build standpoint, it's you have unique IP, build. If you know what you have is unique, then you build. If you have confidence in some, just ideally some, you know, some really empirical data, but confidence that you can do it better or faster than others, you want to build, right? So that's the primary build construct. As far as from a buy perspective or a partner perspective, you kind of have to buy if you know the market's moving and you're not there. If you know there's a trend, because I remember when I was at McAfee, I was running product there for years, And spyware came up, right? And luckily we could build. And within a year after releasing, we had $50 million more, right, on that one category. It was a great build. But if you can't build and the market's coming at you and you know you're going to miss the window of that curve, right? We've all read Crossing the Chasm, right? That's the Silicon Valley Bible, right? If that inflection point, right, if you're going to miss that and it's in your world, your category, And you can't build fast enough, you dang well better buy if it's your core, right? And then from a partner standpoint, I like to partner when you know the distribution strength is there with your sales force, your customers are pulling for it, and you can add it to your platform, but it's not core. Partner's beautiful.
Well, thanks. Okay, we can't be in a panel in 2025 without, You know, we've gone kind of almost 25 minutes without mentioning the word AI. Emma, actually, you said it in your intro. We did a study of about 750 different small businesses, and what we found on the left-hand side is about 30% of the businesses said that they make purchasing decisions today influenced by if the company that they're considering has AI features or not. On the right-hand side, when we asked them, but how many, how many of you are actually using AI features today? And the answer was much larger, 69%. So I think we're way past where AI is a buzzword. It's important in your marketing material that AI is mentioned. And obviously customers are making real use of AI features to streamline workflows, empower the users, and basically generate value. Emma, how does HoneyBook leverage AI in its product?
Yeah, for the past 2 years, we've been building features into the workflow for our members, whether they have a sophistication to adopt AI or they just don't want to adopt a new feature. Our strategy has been that AI is part of the embedded workflow. And so we just built trust with our members to make sure that AI drove the results for them. And we started with like the mundane tactical things, again, trying to free up time for them, increase productivity, and remove the friction. So examples of that were meeting note-taker or email note-taker, uh, helping them identify leads which were high value to respond faster to AI automations. Throughout our existing core product, we have built in AI. We also have an AI chat, which helps them not only understand how to be a more effective business owner, but about their business. They can ask questions about how many bookings do I have this month, how should I change my pricing, what does it look like year over year, to help me write email copy for Instagram. All of that is part of the platform, and it's paying off. We are seeing that our customers customers who use our AI are 2x-ing their bookings and 94% higher profitable.
That's really cool.
Reece, same question to you. How's Thrive leveraging AI in its offering?
You know, when we try to help a small business get found, there's a lot of things that they need to do, right? So content generation is the slam dunk for us. It's write the social post for me. Write the email campaign for me. Write the service description for me. Write the website for me. Can you? All of those 4 are huge relative to AI and a small business who, you know, maybe a home services, HVAC or something, plumber, they don't want to write a bunch of stuff, right? They need that help. They want that help. So AI has been magnificent. That's chapter 2025, right? Chapter 2026. Leads are coming into the business. Leads have different characteristics and attributes. How do we take those leads with their different— with the different metadata associated with that lead, geography, zip code, anything you know about that prospect, and pop it into an AI automation for lead nurturing? Is our research area for the next year where it becomes more strategic.
Super interesting. We see AI is front and center for you guys and and it's not kind of this future thing. It's right now and it's happening. So I see we're kind of nearing the end of time. I want to kind of wrap up with maybe a last question to both of you. What's been one of the more surprising things that you've kind of discovered along your product expansion journey? Emma, let's start with you.
Great. Mine is really simple, but I found it true at every company I've been at. Which is staying focused on your core. You know, it's really easy to chase growth, chase innovation, you know, wanting to make all sorts of pivots. But at the end of the day, staying focused on your core customer, staying focused on your core business, staying core to your mission is always the right choice with everything going on.
Reese?
Yeah, I would have to say the thing that surprised me most happened about 5 years ago. When I, I've been very fortunate in my career to where I figured I'd go ahead and take a break and retire. And holy smokes, I cannot believe how much I love you all. I love this industry. I love software. I cannot believe how fun work is. Don't take this for granted. What you're doing, this journey you're on, this mountain you're climbing, enjoy it. Have a blast. It is fun. And I always felt like I had to work, you know, because I did, right? And it was brutal. And I had an attitude of I had to. Holy smokes, when that disappears, I have to tell you, everything changes. What you're doing is fun and it is awesome and it is rewarding. Just eat it up and enjoy every day. That's the biggest surprise in my career.
Yeah.
Amazing. That's a great note.
I was—
we were at Duda just celebrating 15 years last month, and I was talking to the team, and it's exactly— it's about that. It's about the journey. And there's a lot of fun things in the journey, but there's also some— a lot of challenging things that each one of us has experienced. And it's all about the learnings, overcoming those, leaning in and learning along the way. So thank you, Emma. Thank you, Rhys. Uh, fantastic to talk to you.
Thank you. What a great partner, Duda.
Thank you very much. Thank you all.
Partner, Accel, Sarah Ittelson.
All right. Hi, everyone. Thanks for being here. As was mentioned, I'm Sarah Ittelson. I'm a partner at Accel, the venture capital firm. And I'm joined by Jesse from Decagon. And we're here to talk about the real playbook for scaling an AI startup. We're going to try and unpack some of the how of what has been sort of a tremendous journey to watch the Decagon team go on recently. So Jesse, why don't we start by having you do a bit of an introduction about your journey?
Sure. Hey everyone, I'm Jesse. I'm one of the founders of Dekagon, and for those that aren't familiar with us, what we do is we're building an AI customer service agent. So we primarily work with large companies that have a lot of users, and these users will have, you know, questions and issues and things they need to do. And in the past, you'd have to go to a human customer support agent and wait on hold and get them to help you and so on. And so the AI agents that we're building are essentially there to have a conversation with them, give them a great experience. It can be over a phone call or live chat or, or, or whatever really. And, uh, yeah, as Sarah mentioned, we, we partnered with Accel at our Series A, um, and yeah, we've been scaling very quickly. And, um, yeah, today's discussion will be mostly about, you know, our learnings from that and so on.
And maybe I'll just offer a mention to partner Accel on our early venture team. As Jesse said, we invested at the Series A. I've been at Excel for 3 years. Prior to that, I spent a decade operating in different startups and had the privilege of being a part of the hypergrowth phases of a few different companies— Uber, Uber Eats, and then most recently, Fare, the B2B wholesale marketplace. But I'd say this current moment and the scaling that's possible within these AI companies is, is unparalleled to even those hypergrowth moments of before. And so excited to talk about it today. And I think, you know, our headline that we submitted for the event was 0 to 7 figures. And I think we, we already are at 0 to 8. And so maybe I'll just open it up, Jesse, if there's any sort of general commentary that you want to offer about that trajectory, or we can just dive into sort of the how.
Sure.
Yeah. I mean, so we started the company late 2023. This was after GPT-4 came out. I think like many people, we were very interested in what sort of things the models could do and, you know, you could build the agents and it was very early days at that point. Um, I think we, uh, did a whole process of like finding the right market to build in. I think that was very helpful because the rate of your growth, I think, in my opinion, is like mostly determined by which market you're in. And so we, we figured out, hey, customer service is going to be probably the big first use case for AI agents. I think in hindsight that's roughly turned out to be true, um, and there's just so many large companies out there that are willing to, to kind of invest in this because they can see the ROI and they can, you know, make it, make a very credible argument of like why they should invest in, in AI. Um, and so yeah, growth was really fast in the first year. Um, we essentially went from zero to eight figures in ARR and it's mostly on the back of like I said before, right? People can see the ROI and so they're, it's easier to justify them partnering with us and kind of spending because they know that, hey, if we spend, uh, $250K, like, we can point to $800K in savings because the AI agent's able to do all these things that we had to pay humans for before.
Cool.
So, well, you touched on sort of, uh, studying the market, choosing a good market. I think one of the key things we've observed is just the degree to which you know your customer. And maybe, uh, you know, this is sort of a platitude in company building, is that you should really deeply understand your customer. But I think, you know, what does that mean in practice? What did you do at the founding of the company? What are you continuing to do as you build to stay close and build that deep customer understanding?
Yeah, I mean, at the beginning you don't have any customers, right? So you're, you're basically just talking to potential customers. And the goal with those conversations is to figure out what they actually get value from. And there's not really a great way to do that consistently. You just have to have built a lot of intuition in these conversations. So for example, you could be talking with the wrong person and they're feeding you stuff that's not really that useful and you waste time building it. Or you talk to someone and they're super excited about this sort of product you're pitching, but you get to the end of it and it's like, oh, they're not actually willing to pay that much for it. So they were excited about it, but there's not really much there. So a lot of the knowing your customer piece is to have these conversations and be really aggressive about, you know, what's, like, what they actually care about and what they're willing to pay for and what they will actually get value from. And so I think we did that process pretty aggressively for this company, and I mean, that's how I mentioned before how we decided on this market. But our view is that you have to go through that process, otherwise You're just guessing at what people actually want.
And maybe give people a sense of sort of order of magnitude. How many customers were you talking to before you started building? What's the rate at which you're, you know, obviously now you do have customers you can go and get feedback from, but yeah, order of magnitude, what were those conversations before you started actually like writing code?
Yeah, order of magnitude, maybe like 100. So it was course of a month. We I would try to pack every day as much as possible with customer conversations. We would have a good conversation and then we would kind of get some sort of commitment from them that like, "Hey, if we built this, like how much is this worth to you?" type situation. And then like, you know, we'd have all these conversations during the day and then at night we'd kind of crank out the product and show it to them the next day. And the first thing you do is, it's unlikely that it's going to be like the final thing. And so you have to have a very open mind about iterating and trying new things. And eventually, once we had talked to enough people, I think we just felt enough signal of people pushing us towards this direction. And that's how you build conviction in, in what to work on.
And then a lot of what you're describing there is that early founder-led sales motion, which I know is a big part of the journey. But sort of how has that changed now as you're building a much bigger business? But how are you staying close to that customer feedback and customer your understanding?
Yeah, we spend a lot of time with our customers. Like pretty much every customer we have, we tend to generally work with only larger customers, but everyone we have, we spend a lot of time with them, like weekly touchpoints at least. And that allows us to be super close to what they need and what sort of things are missing. And if someone feels like something is missing, they might not just proactively tell you. And so you do have to be proactive about getting them to share what they're thinking and what sort of things could could be the next on the horizon for the product. So that's what we do now, and then obviously we're still a pretty young company. We're still doing a lot of go-to-market and sales and so on, and so I'm very involved in that, and a lot of our team is involved in that, not even just the salespeople, and that helps us stay pretty in tune with the customers and the potential customers and so on.
And this wasn't the first company you'd founded, so were there any lessons learned from prior experiences, or maybe, you know, in your community of founders, are there, are there traps that you see folks falling into in this, in this bucket?
Yeah, so, uh, before this I started a different company. Uh, it was eventually acquired by Niantic, and then that company was a consumer company, so very different. But probably the biggest learning was along the lines of, you know, the discovery process, where for that company we just did too much of the, like, sitting by ourselves and thinking about what would be a good idea and then building that and like doing that over and over again. And it was quite tough because we were grinding pretty hard and like eventually we just launched a bunch of things that had no traction like whatsoever. And that's when you, that's when people tend to burn out more. And so I think the biggest learning for us was to not do that and instead just spend a lot of time with our customers beforehand.
Okay. The next big kind of pillar that you've been super thoughtful around is team building. Uh, when you and I first met, actually, I think I was first sitting with your co-founder because you were closing employee number 5, and it was amazing to see sort of your focus on talent and hiring, you know, especially at those early days, but it's continued and been, uh, I think a really important source of sort of differentiation. So how do you How do you think about getting the right people on the team?
Yeah, I think the team is more or less a reflection of what the founders care about. And so whatever my co-founder and I thought were important, I think that's reflected in our team currently. Um, we're very proud of the team we've built. It's obviously a mix of, you know, sales, go-to-market, engineers, PMs, deployment folks, and so on. Um, and Everyone has different preferences. I would say for us, we knew we wanted an in-person environment. So even today, we're essentially 100% in person. We really cared about having people that were very just intelligent and could adapt easily to things, right? So I think that, that kind of selects for a certain profile at the early stage where you're working with both engineers and salespeople and people to be very comfortable in this environment where there's not that much structure and there's not that many rules rules to follow, and you can't just like sit down and someone gives you a to-do list and stuff like that. So, um, yeah, I don't think there's like one right way to build a team. It's just the founders have to really be clear about what, what things they, they care about and what things they're not willing to move on. And then once you're able to find enough people, that, that kind of becomes the culture of the company, and then it just continues from there.
And then that's sort of, uh, you know, the team as a reflection of the, of of the founders. Maybe do you want to talk about sort of finding your co-founder and what was important in that process and, and, and how you guys came together?
Yeah, finding a co-founder I think is just very luck-based. Um, it's the same thing as finding any other, like a good friend or a partner or whoever. Um, my co-founder and I, we happen to be in the same stages of life, so we both started a company before that was acquired. We, I think we officially met through A16Z. Z who led our seed and they backed my last company as well. And so it was kind of happenstance. We had a lot of mutual friends. And I think the most important thing for co-founders is you have to be in like a similar mind space. Like your skill sets can be different, but like the, the level of commitment, the working style, the, the type of company you want to build. I think if those are pretty aligned, then it's just a lot easier. And Yeah, again, I think like co-founder team, it's just, it's so variable because everyone is different, right? And like every candidate you meet is also going to be different. So you have to just be open-minded and like, I mean, be open to new things and eventually like you'll find someone that I think is a great fit.
Would you have advice for anyone who's in that sort of co-founder dating stage? I feel like I talk to a lot of founders or sort of pre-founders who are in that exploration, like what are the dimensions that are maybe okay to compromise on and which are the ones that you think, you know, you should absolutely hold out even if it means delaying starting your company as much as you may want to right now?
What would you advise?
Yeah, so if you're looking for a co-founder, the easiest thing to do is just to experiment and like build something together. I think that'll give both people or the whole group like a really good sense of what, what it would be like. I think this is kind of unique to co-founders because if people are considering co-founding, they probably do have the time to invest in this. If you're just hiring someone, maybe they have a full-time job and there's not really much time for you to get like a, a trial in, so to speak. But for co-founding, you can definitely get a trial in. That'll be the best signal by far. I think the, the biggest thing you can't move on or negotiate on is is just the level of commitment, like, people have to being a founder. And I think you'd be surprised, but I think there's a pretty wide range of that. And if one person's more committed than the other, then it's just not gonna work no matter what. So that's the first one. And then probably the second one is just, like, working style. Like, are people working at roughly the same pace? Are people kind of, like, putting in the same amount of time? Are people kind of, like, kind of gearing for the same types of goals, right? Like some people want to build companies because, um, you know, their approach is, is what I mentioned before, which is like it's very customer first and you're just kind of doing discovery. You're not tied to a certain specific idea. Other people might be more tied to specific technologies or specific types of companies that they want to build. And so I think if those things are aligned, you're generally good. And after that, it's just more of a personality thing and, and that sort of thing you can, you can really figure out in a trial.
And then when it comes to the broader team, we'll talk in a second about execution, but I think you have a very high execution, high intensity culture that you've built to achieve these results. How do you screen for that in the hiring process? You know, you speaking about commitment and sort of the co-founder, but in all your employees, I think that cascades through. How are you screening people to bring the right level of, you know, sort of rigor to the job that you guys expect?
Yeah, I mean, it's something you can pay attention to when you're interviewing someone and you get some signals, but it's more or less you just ask them about their past experiences, right? Like, tell me a project where you just invested the most time in, in your life in, and, uh, you see what they say. Uh, you can kind of get a sense for, you know, the favorite projects that they've done in the past, like, You know, how hard were they working? What sort of pace did they have? And then you can also get back-channel references, right? And so if that's something you're optimizing for, it doesn't really matter really what you're optimizing for. You're only going to get so much from the interview, but you can kind of see examples of the past work. You can see examples of, you know, like past relationships they've had, and then, you know, ask them. So yeah, I don't think it's like anything too different than interviewing for other things.
Do you do any active like counter-sell?
For us, not really. I think we, by the time I think the final interview happens and like I'm talking to them and so on, I feel like we've already screened for all these things. Like people have already opted in to being in office every day. People have opted in to being part of the culture we have. And that's great. And so my job is mostly to just to talk about the past experience, like see if it's a good fit for our team, see if their personality is a good fit and so on.
Well, let's get into that sort of execution, culture and how you've built that. How have you set up the team to sort of execute at the pace that you have?
Well, in the early days it's easier because everyone's in one room and you know everyone. You're just like talking all the time, right? If you're, if you're in the same office. I think nowadays we're, we're really just trying to make sure that everyone has north stars that are aligned in the same direction. And so if you're talking about sales or engineering, like people have to know what they're working towards and like what the important goals are and And if they're not working towards one of the important goals, then it's probably not that important. Um, and then it's, it's very measurable, right? Like for us, we are a SaaS company, like I'm sure many folks here, and, uh, the main thing you're measured on is ARR and how many customers you have and like what's the quality of those customers and so on. So these are very measurable metrics. And so those, that kind of becomes North Star that everyone's marching towards. And we have ambitious goals. And so that kind of naturally drives some sort of incentive intensity and ambition with the team. And then along the way, there's a lot of checkpoints, right? It's like, if a customer gives us really good feedback, we treat that, it's like, we celebrate the wins, we share that with the team. If a customer goes live, if we close a deal, if a customer renews for 2 years or whatever, right, those are all wins and we celebrate all of them. And so people across the team know that what they're working towards are these high-level goals. And, Yeah, I think that so far has driven the intensity and the execution, but again, we're still early. I think there's still a lot of things for us to figure out. And one of the things we are gonna hopefully figure out is like how do we keep this execution culture as we grow? 'Cause right now we're about 100 people and we're growing quite quickly. And at some point it becomes harder to maintain the culture just by like saying stuff, and so you need to have systems in place and so on.
And what are you observing as those behavior, that sort of accountability culture, that cadence, How, how is that playing out in practice in what is sort of an unexpectedly like rapid turnover in software, right? Like your market, as you mentioned at the top, like you, you researched it, you were sort of correct in identifying that customer support might be one of the first largest to be upended by AI. That has meant that people are sort of going into purchase decisions, maybe out of schedule that they were originally planning to. Maybe they're having to do POCs or, or proof points differently. Like, what are you seeing in terms of the SaaS go-to-market that has maybe evolved in this sort of AI moment?
I think the biggest difference with AI agents is that you can really— you should be able to really test and see the ROI. And so in our case, right, it should be very clear if you were deploying AI agents for someone, They should be saving money, first of all, because, you know, the AI agent should be able to resolve a good portion of their conversations, and that's a very quantifiable metric. Uh, the second one is customer satisfaction. Are customers actually happier interacting with this AI agent instead of waiting for a human? And if both of those are going up, then it's like a very, very easy business case for them, because it's like, hey, we're saving a boatload of money, and people are happier, which means we're going to generate more revenue in the future, and people are going to be retained longer, and so on, and then, you know, We kind of break down that math for them and that's how you're able to kind of drive sales faster. I would say that that is a pretty necessary component of any AI solution nowadays is because people have come to expect that and because AI is just so easily testable. People are expecting very hard ROI. I think the two qualities of our space that have been very helpful for accelerating has been number one, uh, there's ROI factor, right? It's really easy to measure what you're gaining from it. And then two, it's easy to deploy. Like, you don't have to do it all at once. If you kind of pick one surface area and deploy it to, you know, 10% of the customer base, you can already start getting feedback and, like, you can be live in production. And if that goes well, then you can— you get ramped up and it's very easy to justify, right? So it's easy to get started and it's easy to measure. Uh, and that's how you get these buying cycles to be shorter and these kind of, you know, as Sarah mentioned, to happen more frequently because you're not waiting for like these big projects to take place and make sure everything's perfect and like test it 1,000 times before they'll even entertain some sort of commercial discussion.
And then I think a source of conversation is just how quickly everything is changing and evolving and advancing. And so how have you structured yourselves to say, sort of at the tip of the spear of what is possible in the technology and getting that as quickly into production and sort of expanding the surface area of what Decagon can do for folks?
It's relatively easy, I would say, right now because we're still a pretty agile team and customer service is one of those industries where it's so large, like there's so much surface area for us to climb. Um, but in terms of on the engineering side, I think we, we've always taken the approach that like, hey, we should be building while planning for stuff to be changing underneath, right? So the models are changing, uh, they're getting better. The new models are good at different things, and so you have to be able to evaluate them and figure out which models are good at what. Uh, there's new techniques coming out all the time as well that people are figuring out, and so you have to be building in a way that is conducive to that. And if a new model comes out, you should be able to like instantly instantly evaluate it and then swap it in and out and have minimal sort of leverage or lift there to get that done. So that's one big piece. Um, and the other piece is when, when you're kind of thinking about building, we, we own the application layer. And so the application layer is kind of like the final product that the end customer is buying. And that helps a lot with this because what we're, what we're tasked to do is we're just solving the business problem, right? And that, that's not going to change no matter how the technologies are changing. So as the technology evolves, the business problems still stay the same. And that's what's powerful about building the application is because you're, you're kind of being priced and you're kind of being evaluated by solving the business problem. And the technology improves, then that should help you. Um, it's not necessarily the case if you're, if you're more in the middle, right? If you're kind of an infrastructure thing, then maybe, maybe you have to think more about, okay, what happens if the models change? What happens if the applications change? So, um, But in general, we're trying to build in a very flexible way such that as the technology improves, that should improve our product too.
And then maybe switching gears to mindshare and breaking through the noise in this category. We've talked a little bit about sales process, engineering culture around staying at the edge on AI. How, from a product marketing positioning, getting the word out about Decagon. There's so much talk about different AI solutions and a lot of pressure tops down for folks to sort of have their AI narrative. How are you breaking through all that noise to be in consideration, to gain preference?
Yeah, I mean, I think at least right now we're a lot more well-known than when we first started, of course. And so the sort of messages that we're generally pushing are number one. Decagon is kind of unique in that we're GenAI native and we're kind of seen as the sort of more modern solution to a lot of this automation that's happened in the space. Because our space has had a very long history of automation, so the natural question for a lot of our buyers is, okay, well, how are you different, right? Like, those people are using GenAI now too, right? So, I think we've really pushed this concept of like, hey, we built from the ground up and because of that you're able to do so much more with with GenAI models, the whole paradigm that you build these agents is gonna be different. And so that's been helpful. And then I think the other piece on the product marketing side is that like we have a sort of solution that actually works, right? That we have so many case studies now with big customers like Hertz or Chime or these customers that are seeing a lot of impact. And they've done case studies with us where, hey, we saved this much money or we were able to repurpose as much of our support team or we were able to drive higher NPS and higher retention and so on. And so when we do our product marketing, a big message is like, hey, AI is really hyped up right now. There's a lot of use cases, but this is one where there's actually real ROI and you have businesses out there that are actually seeing fundamentally powerful impact. And so that's what we emphasize when we do product marketing.
And then, you know, with the trajectory you've had, how do you think about goal setting, raising the bar, keeping that ambition high? I think there's a healthy level of sort of dissatisfaction, you know, despite— I think that, you know, some folks might go be patting themselves on the back, and that is not at all the attitude on your team. And so how have you built that into the culture? How do you think about driving ambitious goals from here?
I think everyone on the team knows that we're still early. Like, the market is so large. We've only touched like a fraction of a fraction of it. And so there's still so much to do. And I don't think anyone feels like we've achieved the goal yet. So I don't think that's hard at our stage. I think the other part of it is just keeping people focused on what could be, right? We've gotten a lot of customers now. We have a lot of things figured out that we didn't have a year ago, but we still have so many things to figure out because the goal is— the final goal is so far in the future. We want to be working with the biggest companies out there. We want to be deployed across the entire surface areas. We want to be handling the entirety of their customer interactions. And so there's still always so much more to do.
Awesome. So we've touched on knowing your customer, building the team, creating a culture of execution. Any sort of other, you know, main messages you'd share for folks that sort of want to go on a, you know, are on a similar journey in the audience?
No, I just think picking the market is very important. I assume most folks here are involved in SaaS, and I think the nice part about SaaS is that it is a very structured way to build a company, and that's helpful because you can, you can measure revenue, you have, like, concrete customers that you can talk to, and so on, right? And I mentioned before, my last company was a consumer company, and the benefit of consumer is you can grow a lot faster, but it's just, like, way less structure, way higher variance, and you don't have, like, single customers who are paying you a lot of money that you can talk to. So, I think within SaaS, being able to find a market that's like a good market and has the properties that I've been talking about, I think it will just make everyone's life a lot easier. So yeah, that's the last thing I would say.
I think we have a minute for questions and there is a microphone right here and they will keep me honest. But if there are questions from the audience, if anyone wants to pop up to this microphone, we can take them. Oh yeah, go ahead.
Hi.
You said about talk to customers.
But how do you do it in the daily basis? I mean, you have some venture capital or someone else that can connect you straight to Rippling or Notion or those kind of people.
How do you do this network to talk to someone who you're gonna use your solution? Yeah, you definitely want to find the companies that are most known for working with startups. Setups, right? Um, like Rippling is definitely one of them. And, uh, then you just, just figure out some way in, right? So like maybe your friend knows someone or your VC knows someone or maybe it's like 2 hops, but like if you just do it enough, I think you'll be able to find someone there and then maybe that person's not the right person or maybe they don't have a use case but like their coworker does or so on, right? So you, it's just kind of like, uh, uh, you have to find the shortest path in and then you just try to do that again and again. So So when we were first starting, we were just aggressively lining up conversations with literally anyone that would talk to us. And yeah, that's just what you have to do. [SPEAKING CHINESE] Great.
If there's one more, otherwise we appreciate your time. And hopefully this was helpful. Oh yeah, there's the music.
Thank you.
Thanks, everyone.
Hello everyone, thank you for coming here. Very happy to be with you. Um, I would love to start with just introducing myself. Uh, I'm founder and general partner at Day One Ventures. It's an early-stage firm that invests in companies and leads their communications. We back pre-seed and seed-stage companies, and we do PR for them. So we make sure that companies that we backed get the best publicity and the best exposure. We have grown from $11 million under management that we raised 6 years ago into middle 9 figures, um, AUM, which was quite impressive. And 11 companies of, uh, that we backed on early stages became billion plus in the last few years. And some of these names you might know. You probably know company World or Worldcoin. You probably know company Remote You.com, Owner, Artisan, that's represented here quite explicitly. You would probably recognize Superhuman and maybe even use their products. Just some of the companies that we backed. I would love to speak about how you can— I would love to speak about how can you generate really significant sales funnel using communications technique, and there are a few steps to this process, and the very first step in this process is messaging. I think, and I would give you like a quick framework, and then I would just give you a couple examples from Owner and from Orkid. So I think first question that you really need to answer is why now? And I think it's really important to connect what's happening with the company with the current reality and explain why audience should be thinking about this company right now. I think second is it's really important to outline important market trend. And third, you have to find some brutal problem metric. Find competition and explain what's the differentiator. Explain why this exact founder will be the one who will win the market. Then the last one, find some notable traction metrics to show that the company is having momentum. So I want to explain it on the example of Owner. I think first thing that Shows why now is because right now, um, 40%— 40% of restaurants in the United States, mom-and-pop restaurants. So out of this 40%, uh, the future is jeopardized by much bigger restaurant chains like Domino's because they have much bigger R&D budget and something that small restaurants cannot afford. And big platforms like Grubhub take around one-third of small mom-and-pop restaurants' profits, which is really a huge number. So it's very little left for these restaurants to develop and to grow. And it's exactly related to the problem that Owner is solving. It's helping small businesses to fight big players and help them to survive, thrive, and run their business smoothly and focus on what they're best at, which is not technology, but which is running their company. And it's exactly what we showed here. We take this message, mom-and-pop restaurants struggle from predatory big platforms. We take exact market metrics and shares that 40% of all restaurants in the country, mom-and-pop restaurants. We showcase this market trend and then we explain that Adam, who is high school dropout, started building this company not for wide audience but his He started building it for his mom who had a grooming business and been struggling to achieve any significant traction. And then there is a company, Orchid, which is whole genome sequencing for human embryo, and this company has gotten really impressive PR. They moved from zero to over $4 million in revenue in less than 12 months just with the help of PR without any marketing spend, and this company has been doing something very innovative. Right now it takes much, it's much cheaper to sequence genome. This company is doing, looking at 100 times more metrics. They do 99% genome decoding on the stage of human embryo and help to help you to choose healthiest baby possible. And again, we connected back to Nur's personal story when her mother was going blind because of genetic disease that cannot be treated, and she chose as a mission to fight potential, to fight this problem and make sure that all future generations wouldn't have to struggle from disease that you, that comes with genetics and comes with mistaken genome. It's like a good examples, and both of these companies has, I will speak a bit later about it, but both of these companies got covered in media from ranging from Bloomberg to Forbes. New York Times, Wall Street Journal, and has been leveraging our PR work in order to grow. Strategy. How do you approach? You created messaging, you know what you want to speak about. Now it's about how you do this, right? What you do is you try to understand what are the business goals. Is your business goal growing customers? Is your business goal to to increase amount of people working for the company and applying for the company? Is your business goal to fundraise? And based on what are your business goals, you decide on exact message, you pick particular news, and then you go forward and you pitch particular angle to certain media outlets that would be reflective of the interest of the audience. I think one of the most common tips that I suggest founders to do when they choose which media you want to go after in order to represent your company is ask your customers what media do they read. And there are like a few examples. There is a company Remote, which is global payroll company, and we were thinking about when we were doing their new round announcement, company raised $300 million at $3 billion value, and many PR agents would say that you have to give exclusive 50 media outlets, we were thinking about how can we get as much coverage as we can. So besides using this angle about fundraise, we started using other angles which had to do with new products and new global talent marketplace that this company is launching. We started speaking about particular spots that's the most attractive for remote workers, and we started speaking in the context of how, like, is remote work here to stay in the context of lots of people returning to the office. So we were able to leverage what's happening to the company right now by coming up with multiple media angles. Each of them would be tied for particular media outlets and would result into multiple media stories. And there is a second example, company DoNotPay. Maybe you know it's robot lawyers that help consumers to fight for their rights. For example, if you got parking ticket that wasn't fair, And you decide to fight it back and get your money back, this robot will do it for you. And you only get charged a monthly subscription if the robot succeeded. So what happened, it was a story that happened in COVID. And they basically helped people to— it was COVID-related story. And we basically used this media hook to get exclusive in Today Show, company revenue skyrocketed. It was already millions of dollars in 2 days. It was 70 times— 70% higher than revenue before Today Show episode, and that led to preempted Series A lead by Andreessen, which was amazing. So this is like couple examples how exactly you can use and leverage different angles that as present for the company life and as present for what's happening with industry right now and with the world right now in order to get excellent results and use it as a momentum to boost company brand awareness and as a result achieve very significant results such as new round of funding. And you can give— have a look on this. And the next thing is execution. You have to Care a few things in the speech, and I give here a couple of examples. Speech pre-embargo and post-embargo. I think what's really important is to take care about subject, and the subject should look as good as potential headline in media story. Think about what kind of headline would be really viral so this media article will go viral on X and other platforms, and try to use it in your subject line. Make sure that it's very relevant, and from the very beginning, start answering the question, why now? Like, it has to be answering fully the question why reporters should take care of it now, not tomorrow, not in a week from now, but exactly now. Why tomorrow this news would not be new, right? Don't try to add as much information as possible. I would suggest try to add only things that convincing reporter to take meeting with you and to speak about the company. And be very explicit with ask. Immediately log this opportunity and suggest like, hey, would you like— here's a pitch— and then would you like to have a call with founder at this time, at this day? And there is really good example of story which we secured for company Able, which is doing AI for police software, and it was few things in founder bio. One of that was that founder used to work for Elon, and we used some of this, some of the elements of his biography in order to get media to care about the company. Second thing is that we used in the pitch, we were speaking about how this founder would go on 20 ride-alongs in order to study more about their customers, and basically he came up with exact products solution when he was going on these ride-alongs. And it again helped us to create something really compelling and stories that both media would want to write about and audience would love to care. And as you preparing for the stories to come out, it's really important to do media training. It's really important to— it's really important to rehearse and You could— you don't have to have media agents in order to do it. Try to explain products that you're building to 10-year-old or your grandmother. Record yourself with a 60-minute— 60-second speech. Watch the video. Try to understand what you can do better. Write it down. Try to rehearse 2, 3 more times. It doesn't have to be anything extensive. Another tip that I really love to do, try to speak about your product and your subject and the news you're speaking about with people who has nothing to do with industry. Try to explain it to barista, try to explain it to Uber driver, and it would be like the best way to create something that's simple and compelling. And the next thing is launch search. I think right now we live in the time when getting media stories are not enough, so I do think it's really important to make sure that your distribution online is robust, and as a part of it, it's really important to prepare compelling posts for X or for LinkedIn. And I think one of the things that's really easy to do— every one of us has at least like 50 friends, right, that has quite good followship. So you must do the very compelling post. You must start with clear— just trying to answer the question why people should care about it. Try to be as sharp and as short as possible. And once you posted something online, ask 50, 100, 200 people in your network who has biggest amount of followers to repost, comment, and retweet. This is a very easy hustle, and mostly, in most of the cases, people agree to do this. It does really create magic. Both of these posts went viral and got more than 1 million views, right? So it wasn't just New York Times story about Orchid, and it wasn't just Forbes story about company Superpowers. It's also now portfolio. It's literally mechanics reaching out to your closest network, your family, your friends, your colleagues, who would post, repost, retweet, like. That overall boosted the brand awareness of the company and overall boosted their recognition and boosted this media stories as company already secured. And here's like some of the results that I'm really proud of. As I already mentioned, Do Not Pay got term sheet, got revenue 70% up. It was already millions of dollars and they got to 70% revenue up just 2 days after the launch and got very successful Series A led by Andreessen. Artisan, as a result of our launch, it Forbes and VentureBeat, got more than 1,000 sales meetings, which is amazing. Orkid has gotten from zero to over $4 million in revenue in less than 12 months, and PR was the only tools they've been using for marketing. They had no other marketing spend whatsoever, and you could see some of the media outlets that it got featured in. And Owner, after we announced Series B, and the goal for Series B announcement was to recruit as many people people as possible, as possible. What happened that not only they got, not only they got more than 700 people applied for jobs, but this company has also got preempted with new round term sheet. So it was really impressive because company hasn't been planning to raise, but then it got preempted with new rounds. So on top of 700 job applicants, this company, which is B2B company Again, it's not a consumer company, which is usually easier to get publicity for. This company also got term sheet for the new round, which was incredible. That's just some of the examples, and I hope that will inspire you to think creatively. This is absolutely free media channel. Lots of this learning's just based on first principles and experience. I don't think it's very difficult. I don't think you need PR people to do this job. It's as simple as I just outlined, and I would highly advise every one of us to try this approach and to always combine getting media stories with going direct to your audience.
Thank you.
Thank you, Masha. That was a very great presentation.
And just a few words.
My name is Dasha. I lead GDM efforts at Puzzle, which is also one of the portfolio companies of Day One Ventures, a modern accounting in software, and I also write for Forbes about venture capital. And I meet a lot of founders, and whenever I meet founders who are backed by Day One Ventures, I always know that it's going to be ambitious, bold company with a great mission. So great job, and I'm really impressed by the work that you are doing. And a quick question to the audience: is anyone working on their storytelling right now or PR? Raise your hands. Okay, maybe more people should do that. And that's actually my next question. So storytelling is something that sets you apart from all other VCs and something that you're really great at. Can you explain, Masha, why storytelling? Why is it so important?
Well, I think right now we live in the time when people's attention span is very short. And since many centuries, our brain and our consciousness is organized in the way that we remember information and stories. And the more compelling the story is, the more dramatic the story is in many ways, the easier for us to remember something. And I feel like right now every single day the competition for someone's attention is only increasing. I think that importance of turning what you build on, I think everyone has a story. I think what it takes is reflecting back and trying to understand what happened to you and trying to practice and rehearse and just find a way how to speak about what happened to you in an interesting manner so it catches someone's attention and then figuring out what kind of moments were the most important and the most turning moments for whomever you're speaking with and turning it clear storyline. I think ability to tell stories is a huge competitive advantage. I don't think you have to have particular education or experience in order to get good at this. I think what's really important to always pay attention on what was unusual about the way how you got here, and I think it's always like practice. If you're good in storytelling, you would be You would be successful with fundraising. You would be successful with recruiting. You would be successful with media. It's something that every founder would need to be good at at any stage of company life and something that would be boosting the rest of advantage. So I think in the current time, to sum up, with attention span we have, it's really important to gain the skill, and you can turn the skill into a weapon.
Yeah, that's great. And you know, the other question is you definitely have your own playbook that you leverage at Day One Ventures, right? You have your own like tips and tricks. So what are the founders that are not like following your playbook usually get wrong? You know, very often founders reach out to top media reporters, you know, and they say, oh, I've raised $500K, can you tell my story to everyone? And while it's a really big milestone for them, I understand, it's not always very interesting for a reporter to cover another just round. So what are they doing wrong?
I think there are a few things. I think one really important thing is to start correctly. We always start with a question, what business goals should communication solve, right? I think it's really important that you do it not just for the article, article and not to brag about the article to your friends, but you do it with a clear picture in mind. Why do you need it? And build your communications campaigns based on that. I think number 2, it's very often that companies started years before and founders having hard time to connect reality that's happening right now with what, what is that that company is building. I think it's really important to work out the strong answer on the question, right? Why now? And if you try to practice the Sunset, try to see, literally pitch it to your friend, to your family, to friend of friend. Use any opportunity to tell the story and try to see reaction in someone's eyes. Try to see if people get energized and people really resonate, want to ask more questions, or they're turning away. I think this is very important. And the last one, I think it's a very typical thing. I'm sure some of you and lots of you will raise venture capital, and it's very often that founders use like raise money as a big validation point. I'm really trying to teach our founders and our team to never speak about fundraising round as its own, because many companies raising money right now— raising money, it's like day one. What's important is like what have you built with that? So I would highly suggest to never speak about funding round. The best would be, hey, we're launching product that's doing XYZ, and by the way, we raised $50 million Series A. Never start it like, hey, there's a company owner and it raised $50 million Series A. Always start with like, here's like what we built, here's why it's so pressing and important right now, Here is why there is a reason why small mom-and-pop restaurant will die without this product. And only then like, and we have this backing. So this backing is just validation that you have a good why now answer. So never start promoting funding round, always promote what you're going to do with it or what you've already done with it.
That's definitely just the best tip you can get really. And I know we have a lot of early stage founders as well in the room. At what stage, like when you're a startup founder, should you even start thinking about PR, about, I don't know, working with agencies or just reaching out yourself to different media and reporters? When is the time?
I honestly think that as soon as your vision is solidified and you know what you're working on and you outlined plan for— you at least can clearly answer questions like, hey, here's like where I want to be 12 months from now. PR is always about picturing the future. It's not about picture like, here's like what it is today. It's always like, here's like a problem and here's the future that we desire to be in and here's how we're gonna get there, right? Here's our game plan and our game plan is superior. That's why we're better than competition. So I would say it's almost like never too late if you still deciding on mission you You might pivot, you might build different solutions too early, but as soon as you have solid vision for your company and you know where you want to be 12 months from now, because you're gonna be telling to reporters and publicity about this plan, meaning that you would have some commitment to follow this plan, right? It's a good time. I think early is better than later. I don't think it's better if your product or MVP is ready, but if your product is If it's not ready, you can always get users on the waitlist. The faster you get your idea exposed to broader set of audience, the better, more polished this idea will be because it could literally bring in your life people that would be life-changing for your company journey. So I actually encourage to start doing it earlier than later.
Yeah. Another question, like let's say you've made a big announcement like a product launch or a big like funding announcement You got a lot of leads, let's say like Artisan, right? 10,000, 1,000 sales leads. But how do you make sure that you keep working on your PR and your communication strategy? That is not just like one-time thing, which is success, right? But not a long-term success. So how do you keep building these relationships to make sure all this PR, all these efforts lead you actually to business growth, to new leads and customers?
Well, I think first of all, it's really important that there is some frequency and I think when you're early-stage company, pre-seed and seed, I think speaking about yourself 2, 3 times a year is enough. Once you get a bit bigger, Series A, Series B, 4, 5 times a year is enough. I think it's really important to take, in my opinion, it's better if you take big steps and you showcase something that's really impressive. So try to see what is, like if you think about the next 2 quarters, What is the most important business milestone? That's one. It's really impressive for people in your industry. It's really giving something really important for your customers. And three, that's like achievable and you're gonna shoot to get there, right? Try to think about and try to time it alongside with big fundraising push, along time with big recruiting push. Push. So try to plan it right before some important moment where and when it's really important that your company is known, right? Try to be very practical about it because your time as founders and executives is priceless. It's the biggest resource, not cash, but this time that your company has. So try to be very, very practical and try to be very clear about communications goals, but I genuinely think that maybe like 2, 3 times for pre-seed company, 4, 5 times a year for Series A, Series B company is a good timing. I also don't think you should be announcing something like every week, but I think if you come up with this news and then prepare, it should be really big splash and just try to get as much media as possible. Try to get all top tier media outlets. As soon as story is good, it's reachable for literally every single company. And two, don't forget about growing your community online, talking to your community in between, not only about your company but about whatever you care about. And use this online friendships and connections to boost distribution of this news. Don't rely on media only. Great.
And last short question, wrapping up our discussion: what founder stories excite you?
I think everyone love underdog story, right? I think they all make us believe more in ourselves. And I think if you look at like most successful people, even if you look even now at people like president or like Elon, I feel like lots of these people, when you see them speaking, they still do feel themselves like underdogs. They still feel like they're not they still feel not fully accepted and they still feeling like they need to prove themselves and they still feel almost like a little bit a stranger in the room. I think it's very typical for many leaders and for many strong people to feel like a stranger in the room and to not necessarily like facing the world which is very welcoming to the ideas but accept just being really committed to their vision having a lot of faith in themselves in the time when other people don't believe in them. I think these stories are very powerful and for everyone. And I think if you like look back and at what's been happening to you and think about the time when you was the only person who kept believing your idea, that's help you to craft this like underdog story that pushed you. And, uh, helped you create what you have created. Thank you, Masha.
Let's give a round of applause. Thank you.
Welcome, welcome. Thank you everyone for being here. I think we have a amazing group of investors here to talk through, I think, what's on the top of a lot of founders' minds. How do you become a differentiated vertical AI company? And so just to start us off, to make sure everyone's super familiar with who we have on the panel, my name's Glen Borak. Uh, I'm an investor at Jump Capital, an early-stage venture firm based in Chicago. Uh, we invest in all things software. But we're here to talk to our panelists today. So maybe just to get things started, would love a really brief introduction. And then maybe just to start things off, what's getting you guys excited in vertical AI these days? Obviously, there's a lot of pitches coming through. What resonates the most with you right now? Liz, maybe we'll start with you.
Sure.
I'm Liz Wessel. I'm a partner at First Round Capital. We're a seed and pre-seed venture firm. And before working at First Round, I was a founder and CEO for 7.5 years. Um, I would say what gets me really excited these days is when the founder has either a really, really specific domain expertise in their vertical, or— and I'm happy to give like examples— but, or some like really unique insight that either is contrarian or, or something I've not heard from any of the other founders in that space. I would say those are the two things that typically perk up my ears in the early stage.
Awesome.
Hi everyone, I'm Yaz. I'm a partner at Emergence Capital. We're a B2B software fund based in San Francisco. We've been around for 20 years only investing in software, and more recently I've been talking about the applications of AI within some traditional industries that we've seen. And one of the things that I find exciting is that AI has unlocked tremendous opportunity within verticals that otherwise weren't venture-backable. So when you think about AI today, you can deliver it to your customers in the way of a product itself, but also in the way you operate internally to be able to scalably reach customers that otherwise were harder to do so in a unit economic positive way. And so now when I look for really exciting opportunities in this space, it's what verticals have been overlooked for the last 20 years of cloud software and, and where can new really great iconic companies be built.
Hi everyone, I'm Amy Cheatham. I'm a partner at Kostanova Ventures. We're an early-stage firm primarily investing in B2B software at seed and Series A and been around for about 15 years now. What gets me excited right now is a couple of things. One, founders who deeply understand their customers and their true north star is understanding them, talking with their customers, and deeply, deeply, deeply ensuring that they know what good looks like from the customer standpoint, even if that's different from what their initial insight was when they started the company. And then I'll piggyback on what Yaz just said of there's so much services spend in so many verticals that is now addressable as software revenue, and we're just in the early innings of that shift. And so very excited about what that means for vertical AI in particular as it expands the revenue opportunity for those companies.
Awesome. A lot of great things to double-click onto. Liz, I think one thing you mentioned about kind of really, you know, having that insight into a specific problem. I feel like nowadays with AI startups, you know, you're kind of getting pitched nonstop variations of maybe the same sort of problem set or maybe a slightly nuanced take. How do you kind of break through if it's like a novel insight and/or if it's just kind of hype and something that maybe is more surface level? What's your kind of evaluation criteria?
I think it's often pretty easy to tell. Maybe it's just because I am taking so many pitches each day, it's easy to differentiate. But like, from a founder perspective, or from my perspective, when I am learning something and I'm like, oh shit, is that true? Like, and I have to ChatGPT or Google it or something like that. Or, um, when you show me that you know something that honestly only someone who worked in that industry would know, then I'm usually really leaning in. And so I think it becomes pretty obvious when someone is sharing something that's something somewhat surface level. Um, like if I think about one of my portfolio companies, um, Beacon, they're in like the physical intelligence space selling to physical security teams. Uh, it's an AI vertical SaaS company, and the 3 founders come from like the 3 biggest incumbents in that space. And so when they came to pitch me, I learned so much, and all of it was not Googleable. Um, it was things that— it's not like they were sharing proprietary secrets from their prior employers, to be clear, but it was things that very clearly only people who had spent the prior 15 years of of their careers in that industry would know. And, and I've also seen that, by the way, with founders who are younger in their early 20s. Maybe they didn't spend 15 years, but they spent all of the last 30 days talking to 10 customers a day and clearly came up with insights.
So awesome.
And yes, at Emergence, whether it's Viva or otherwise, you know, within vertical SaaS, you guys have always had a really, really strong position and thesis. What kind of lessons do you think can be taken from vertical SaaS into kind of some of these verticalized AI plays that we're seeing? And do you evaluate those differently or are they kind of similar in nature?
I think they're so similar to what we saw in vertical software or industry cloud or, or some of these preceding large companies that have been built. I'd say one of the big differences now is How thoughtful are you around data collection? What are the things that you're incorporating into your customer contracts that allow you to continue to innovate and be at the forefront of, of the model outputs that you're delivering for your end customers? And are you able to deliver that through your product UX really seamlessly? Those are things that vertical software companies often talked about, is we have unique insight because we are really close to a customer that has very specific needs. Now that's more important than ever because that data capture is actually fueling so much of the engine of that business. I think the, the other thing I'd share, uh, with respect to the Vivas and the Doximities of the world within vertical software is we've often talked about layering the cake. It's you land with one initial wedge product that people desperately need and your unique expertise delivers a really compelling new take on how to solve that problem, but then it's how do you continue to deliver additional value through incremental product SKUs. I think what we're seeing now, related to the first question you asked, Liz, is so many of these pitches sound the same. So many people are, we're going to be voice AI for healthcare, or voice AI for this, and they have one use case. I think if you are doing just that one thing in a market that's moving as quickly as we are now, in 12 months your company is likely dead. How can you really and quickly capture feedback from your customers, increase your surface area and the products you deliver towards them, and really continue to capture more and more value from that, from that end customer?
Maybe I'll add one thing to what you just said because I completely agree, and I think a lot of what we see is people biting off maybe too big of an industry category, and it leads them to taking something that sort of ends up looking like horizontal software and they're sort of branding it as vertical, but they're only building this, this one specific thing. And I always encourage founders like maybe start with a smaller chunk of the industry. And if that means, you know, within legal tech, for example, if you only have to sell to personal injury firms at first, well then fine, but build everything that they need in order to run their firm so that you're eating all that spend. Instead of being this sort of, you know, not to pick on voice AI again, but instead of being this single-use tool that then can't absorb the rest of the workflow, I think it's much more exciting to pick a smaller category and then expand from there.
And Amy, you've been spending a lot of time in legal, like I know Hona and a few others, and you mentioned AI unlocking bigger opportunities. How do you kind of think about whether the space is actually big enough? 'Cause we always talk about TAM, we always talk about, hey, is this a venture-backable business? How do you kind of weigh those trade-offs between being narrow enough to have a focus, but also be big enough to be venture-backable?
I think this is a constant debate, and I'm sure even amongst the 4 of us, we all probably would disagree a little bit on this one, but I am of the mind that if you back the right team that deeply understands their customers, they're gonna figure out a way to build a big enough company. And that's not always true, right? Of course, there are, you know, things that are exceptions to that rule. But I think if you're building in a category where your customers really want a product, you're going to figure out how to build more surface area. And I think we've seen that again and again with— I'm sure people probably said that about Viva early on, that there was no way it was going to be big enough. And I think the prime example that I'll point out of one of my portfolio companies is a company called Ivo that's building in the contracting space and initially started out as a contracting tool, but Yaz's earlier point, quickly have realized that there's a much bigger, longer-term data play that eventually means they can turn into a next-generation AI-native contract lifecycle management tool. But if initially what they started building was really, really good inside your Word document contract editing, right? And that seems pretty small. And so I think if you're a seed investor and you're meeting that company, you're like, ah, I don't know how big this can be. But I think once you have that data flywheel, there's a lot more you can do with that.
Just to go off of that, like a piece of advice I give to a lot of founders when they're pitching pre-seed, seed, and even A is going off of her point, like the TAM should often in vertical SaaS feel pretty small at the seed pitch for what you're doing today. But it's on you guys, the founders, to be able to portray your vision for what this could become. And that doesn't mean that you should be building that vision. So I like to tell a lot of my founders it's like an Act 1, Act 2, Act 3. Act 1 is what you're pitching at your pre-seed, your seed, maybe your A. Act 3 is the product that we're celebrating at your IPO party. And Act 2 is everything in between. And often founders don't know what Act 2 looks like, and that's totally cool. But Act 1 and Act 3 are pretty important to be clear about in your pitch. And if you're starting to build Act 3, at your pre-seed, seed, or even A round, you're probably gonna look like a mess. Like, you're like way too all over the place. So I, I couldn't agree more that like a lot of investors— I can speak for first round. Our worst pass ever, uh, biggest exit that we should have invested in but we didn't, Twilio was. I think it was our biggest. Uh, we passed because of TAM. And I think after that happened a few times, and I knew Uber, everyone— we were very lucky to lead the seed, but like everyone else famously saw it and passed because of TAM. And I think a lot of investors have learned their lesson but still look to the founder to be able to explain what the bigger TAM looks like in 10 years.
Yeah, I, I couldn't agree more. And what I'd say is I think most VCs suck at running TAM analyses, myself included.
Me too.
And so your job, especially within verticals where you are the domain expert or someone on your founding team is the domain expert, being able to articulate why it's a compelling opportunity is so much of the job.
I—
with— you mentioned Veeva. When Veeva pitched, their initial product was a $400 million TAM. They're north of $3 billion run rate business today, and they thought that they were going after a $400 million TAM. However, what they were able to articulate was how much spend on IT happened within the life sciences industry and the other adjacent industries like med devices and and so forth, and biotech. So when you think about how do I actually articulate Act 1 or Act 3 or however you described it, it's am I building something that matters for a group of people that matter? And the path to doing that, you don't have to have 10 products when you're pitching a seed. The path to doing that doesn't necessarily have to be clear, but the end goal does. I love that framing, Act 1, Act 3, and Act 2 being unclear.
And Liz, it sounds like— so TAM is not as much maybe of a consideration early stages, but it's in the name, first round capital, oftentimes a check even before there's a product, even before there's anything. What red flags do you see when people come and pitch you? You see pitches day in and day out. What are some things that stand out to you that make you stand back and maybe, you know, question some of the things.
Yeah, I mean, I'll kind of be a little all over the place and just name a few. Um, one, like, if they're for— if their pitch is for Act 3 and that's what they're building today, I'm usually like, this is going to be way too hard to bring to market if, if you're trying to build something that other corporations have taken 30 years to build and all of that. So like, you know, not something very specific, but that's— I'm just reiterating what we just talked about. The biggest one is just when when the founders don't know their stuff, right? Like, it keeps coming up. And so hopefully you guys are all taking the hint that it's probably the number one reason I certainly pass on an opportunity is when the founders just clearly don't know the space or their customers as well as they think or they are claiming to. And it shows up by either the founders not knowing their numbers, not having unique insights, not being able to answer basic or even more not basic questions about their customer usage. Just the other day I had a pitch and they had like a few design partners and I asked some really basic questions about how the design partners were using it, how many times a week they were logging in, how many users were logging, and so on and so forth. And they couldn't answer any of that. And that kind of told me they're just not spending enough time with their customers. And usually founders don't make it when when that's the case.
Yes, one thing, you know, we've already talked about a little bit today is kind of building a, a broader platform business, building something beyond kind of just the voice AI use case. I'm curious, in this day and age when, you know, models are becoming commoditized and there's a lot of different ways to kind of think about differentiation, what at Emergence matters to you guys? Is it, you know, it's— you mentioned data. Is it distribution? Is it the workflows that you've built? How do you think about building a moat in the age of AI?
Yeah, it's a great question. And what I'd say is there's so many different tactical things that can enable a company to break out from the early stages, right? It could be distribution, it could be some unique proprietary data access and so forth. What I'd actually generalize this for in particular when it comes to verticals is, are you really capturing the value that you're delivering versus all the other existing horizontal solutions that exist? So one of the things that we look at when we're evaluating a vertical software company, say, I'll come up with a random example, but we're a vertical for sellers within a particular industry, and the comparable company is an Apollo or a Salesloft or and outreach, but you're purpose-built for financial services. If you are charging per seat less than what those incumbent players are providing, then you're probably not capturing upside value. You're just focusing your go-to-market on a smaller, more targeted group of people. What I care about is that the value you're delivering by being focused on a particular vertical is far higher for the end customer. Customer because you're building for them. And that was the case in software, and it's even more the case with AI. As models become more commoditized, you being able to leverage data in the way you run retrieval or prompting or tuning your models, that becomes incrementally more valuable, and you should capture that value. And it's hard. We invest at the Series A and Series B. It's hard to price as a premium solution to existing solutions that could have been around for decades. But holding the line is really important. Otherwise you end up constraining your TAM too far, too early, and then you're no longer a venture-backable outcome.
Makes a lot of sense. Amy, at The Seed, do you look for the same things? Like how do you kind of think about the long-term differentiability of businesses that are coming to you when they're so early? Maybe they're incubating in your office, maybe they're, you know, just a couple hundred K. ARR. What do you guys look for at Costa Noa in terms of defensibility?
Yeah, look, it's really hard. I mean, things are moving incredibly quickly. I was just on a, our spring update call with all of our LPs this morning, and I said to them verbatim that this is a really hard time, especially at the infrastructure layer, to figure out differentiation because things are just moving so quickly. And so I will say, like, for us, especially in data and AI infra, it feels really hard to figure out where to deploy capital right now. And all, and all transparency, I think at the application layer it feels a bit more clear, which is probably why I'm sitting on this panel. Uh, but it's really hard, and I think it comes down to founder, their kind of founder-market fit. Um, and then to me, like, we really like investing in highly regulated industries, and whether that's financial services, healthcare, we've done a lot in national security and defense. And those are places where I think Yaz was mentioning I mentioned earlier, you know, these horizontal solutions, I think they have a much harder time penetrating highly regulated industries. And so the purpose-built solutions tend to be able to charge a premium there in a way that they can't if they weren't purpose-built for that industry. And so those are the things where we've been able to carve out a little bit of deeper understanding of what that moat and differentiation might look like going forward. But for sure, I mean, I think historically we talked about moats as being technical, and I think increasingly it's becoming data and your go-to-market and your ability to give customers the ability to trust you with that data. And I think they're much more likely to do that in a highly regulated industry if you come from that industry and if you're purpose-built and deeply understand what they care about in that sales cycle versus something that's more horizontal that might not. But it's hard, and I think these things are shifting underneath our feet right now.
Totally. And just a quick follow-up to that. You mentioned kind of maybe the, the technical prowess that is needed to kind of get something out the door and get some traction is lessening. Bolt and Lovable are being used repeatedly. How do you kind of think about evaluating the team in that aspect? Is there less of a kind of waiting on the technical capabilities? You often see you know, CS PhDs on founding teams and people love to weight like kind of their capabilities. How do you think that changes with the different weights on kind of building a business?
Look, I think it's really exciting that you can vibe code something over the weekend. Um, I, you know, I think we've hopefully all done at least a little bit of that. And if you're not, you should, but we're still a long ways from having something to be production ready. Like I tried to build an app that, scrapes my Gmail to tell me every piece of clothing I've bought in the last 12 months and puts it into an app that I can make a closet. And let me tell you, it was not good. Uh, and so I think we're a long, long ways away from that equating to somebody building something in defense tech in, in Lovable. Um, but your point is not at all lost. Like, I do think it's a really important one. Look, we still deeply care about backing technical founders, and I don't think that's ever going to change, but That does mean that you can get to scale with a lot fewer people. And I think it's going to change the way that venture thinks about how much money these companies need early on. And it may change how many rounds of capital people are raising over time. But it certainly doesn't change the profile of what we're looking for.
Makes sense. Yes, just one follow-up on kind of some of the different things you've talked about in terms of Veeva and some of the other things that Emergence has looked at in the past, I feel like traction has always been something that is top of mind for, you know, Series A, Series B investors, momentum, trying to better understand kind of how that business is doing. And one thing at least I'm seeing at Jump is, you know, I think something coined experimental recurring revenue where a lot of companies are using a lot of different AI solutions. They're trialing them, maybe it's 3 months, maybe it's more, and then they're kind of churning off. How do you kind of evaluate that revenue stickiness at a emergence, and what are you looking for to see how long-term that revenue is?
Yeah, it's harder than ever as a VC to assess revenue quality now in an AI world where there's so much experimental budget being deployed and top-down pressure from executives saying, we need to have an AI strategy, go buy a bunch of stuff. Especially when you're speaking to vertical businesses, in particular highly regulated industries like healthcare and financial services, which is you now have people that have these mandates to go deploy. And for the first time ever, they're unlocking these budgets and there's dozens of AI startups going after them. And so one of the things I tell our founders and founders we're assessing is, is you have to get out of innovation budgets as fast as possible. Who is the end buyer, right? If you are building a solution, for a financial services company within their back office. Are you talking to their back office teams and helping them deploy that more effectively? Or are you talking to someone in the Center for Innovations that wants to try out 30 different solutions? It's okay to start there, but to the, to the extent that you can, you have to move as quickly to the actual budget owners over time. Otherwise, the churn rate is going to be incredibly higher. Than, than those in the innovations department. So it's really find business value, find those that care about that business value, align your interests with them, deliver clear KPIs for those people. Otherwise you're just going to be stuck at the C-suite innovation department that are just trying to like hit a check mark for the fact that they've procured 30 vendors this quarter.
Liz, a lot of the early companies you work with are probably going to those innovation budgets. How do you kind of coach them on where they should be getting revenue from, how to get revenue that's sticky that they know is gonna be there for the long term?
Most of mine are not going to the innovation teams because they know that the quality of revenue— I say to all my founders, I would rather you have $100,000 of ARR from great customers who love you and are using you every day than $1 million in ARR from bad quality, like customers who should not be your customers. And I think any investor would say that, and I think any great founder also feels that. So I think it's exactly what Yaz just talked about. It's like, if you do start with innovation, your entire job is to just get in with as many people outside of that innovation group as possible. But, um, I am often— maybe I'm old school, but I am telling all my founders, go get on a plane. You happen to be in Omaha, Nebraska, Tomorrow, where that customer, the end user is. You happen to be in Dayton, Ohio the next day, like get on, you know, tell them that you are going to be in all these places and find out who's willing to meet with you while you're there and then get on the plane right now and go. And I think that's the best way to, if you're really struggling to get in with the end user but you've gotten in through another channel, you're somewhat in. Maybe they know that you've passed the through their procurement or security or something. So you have a leg in, so, or a foot in the door. So you should absolutely at that point be using and abusing every opportunity you can to talk to the people who are really the good customers.
Makes sense. I know we only have a handful of minutes left. I have a couple just quick questions on general pieces of advice for founders here. So one thing I've noticed, even just walking around this conference and talking to founders is that there is an abundance of startups that all are kind of claiming similar value props, tackling similar types of problems. Like if you look in sales tech, for example, there's a lot of different AI SDRs or top-of-funnel tools or research tools, and it, it feels like it's hard to kind of figure out where things sit in the market. As you kind of work with early-stage businesses, what kind of recommendations can you give on how they position themselves, how they kind of break through to their end buyer, and how they kind of stand out amongst the crowd?
I mean, one thing I'll start with is if you as a founder, and if I ask you like, who is your competitive landscape? Who's in the competitive landscape? Who are your competitors? And you say no one, that is like a massive red flag. Because either, unless this is a brand new space that was invented yesterday, a brand new profession. Either you don't know your stuff and you are completely oblivious or lack of self-awareness, or this is a horrible space where every company who tries to be in it dies. And it's, it's probably maybe a mix of both. And so I would say to start, when you're thinking about differentiation, don't go to that excuse or that response because it's, it's more common than you might think. And really just figure out where you're differentiated short-term and/or long-term. And it's okay, sometimes it's like, here's the— yeah, we, we look pretty similar today, but here is the data we're capturing that no one else is that will enable us to be able to build that moat or that differentiation long-term. I think the whole moat thing, um, it's so important at the later stages of a business life cycle, but realistically, in the first year or two of company, very rarely do you have a moat. Often it takes a couple years, whether because you're collecting data that's training and therefore getting your product to be far above the rest, or you're, you know, I mean, there's a million ways to build moats and you guys all know about those. But I would say often it's just about painting that picture in the long term. But really owning your competitive landscape day one, I would have to encourage you all very strongly to do so.
Amy, yes, anything you want to add there?
I'll just add one thing, which is that I always used to not like competitive categories, and I think that's something I've changed my mind on. I think Liz is exactly right. Like, if a category is competitive, usually, not always, it means that there's a real customer pain point there and people actually want to buy the software. It doesn't necessarily mean that it's not going to be hard to figure out differentiation, but I sometimes I definitely think it's a red flag when someone says, oh, no one's ever done this. Yeah, that's probably because the pain point might not be real, right? And if you don't have an understanding of that and you can't articulate that, that's really concerning. And I think there are places like legal tech that are incredibly competitive right now, but at the same time, there's so much spend getting poured into that category. And so I, I definitely have changed my perception of what competitive might actually mean.
Yeah.
The thing I'd add specific to this panel is if you're building in vertical AI, One of the beauties of AI versus traditional software is you have very measurable outcomes. Like with an AI solution, you're often pitching on the fact that you do something better, faster, or cheaper. Ideally a mix of all three of those things. So if you've delivered a new solution and that is part of your promise, I come in at the Series A or B, so you often have more time under the curve to prove some of those things out. You can now say Company X 'Does this, we do this, and this is how much better it is.' And talk to our customers that have tried the other solution, and this is the magic behind it. If you lead in with that, now I want to understand why your architecture is well positioned to do this or why you have some unique data. Saying those things without the proof points, now in this world, like, if you don't have the proof, I don't care. I don't care. Prove it, and then you've afforded the luxury to explain to me why the choices that you've made have led to those measurable outcomes and the delta between you and everyone else that's delivering a similar solution.
And even at the— I couldn't agree more. And even at the earlier stage at seed, like if you have a design partner or a first customer who chose you over your competitor, being able to really clearly articulate why they did, and then also making sure that they're referenceable so that often I'll invest in a company that has two customers. Like, you know, sometimes it's zero customers, they still work at Google and they're about to quit. Sometimes it's they have like maybe two customers. And so I want to be able to call those customers and understand what else did they look at. And if they chose you because you were the first one to reach out to them or you're their next-door neighbor, it, it makes it a little bit harder for me to believe that you really are beating— going to beat your competitors in the market.
Awesome.
And we have one minute left here. I, I would just love going down the line really quickly. We've dropped a lot of really useful nuggets on what you look for, what are red flags. Maybe just one last thing. If you had to give a vertical AI founder today a piece of advice as they think about raising their seed, raising their A, and scaling a new business, what do you think is kind of the most important piece of advice based on what we've discussed today? Amy, I can start with you down at the end.
Pick a category you deeply care about and are willing to get as deeply embedded in as humanly possible and be willing to do that for the next 10 years. And if you can't pick something that you want to do that with every single day, then you got to pick something else.
Yeah, for me it all comes down to the team, and it's rare to find someone that has as deep domain expertise within a vertical and is a great technologist in one person. If you have that, you are very rare. In the absence of that, find your counter part that you trust, that you can build with for the next decade, and you have the recipe to build a really big company. Each of our vertical winners, Viva, Doximity, so many others had that perfect duo from the outset of the business.
One piece of advice for vertical SaaS AI founders would probably just be spend all the time you can with your customers. And even when you start growing and you want to spend time on other things, don't. Keep spending time with your customers.
Awesome. That's all the time we have. Thank you so much to our panelists, and thank you for being here.
More from Jason
Why Your Next Hire Should Report to an AI, Tragedy Apps, and Too Many AI SDRs? | The Agents #004
Our Agents Drove Our Salesforce Bill Up 80%, Marketo Hates Agents, and We Stealth-Churned Off Notion
Managing 20+ AI Agents: Lazy Agents, Our $500K AI Bill, Stealth Churn & the Death of 60% Solutions
