This Week in Startups - Merge's Unified API Bet in the AI Era | E2033

Episode Date: October 25, 2024

In this episode: Alex talks with Shensi Ding from Merge, who shares her transition from finance to tech and highlights the critical role of integrations in modern businesses. They explore the challenges of building and maintaining these integrations, emphasizing the importance of standardized data models. Shensi discusses partnerships and the strategies to overcome resistance. The conversation delves into trust-building, enhancing integration observability, and expanding Merge's offerings to include AI applications. They also touch on financial efficiency, growth strategies, and Merge's evolution beyond APIs, concluding with insights into future plans and potential acquisition interest. * Timestamps: (0:00) Alex kicks off the show with guest Shensi Ding of Merge. (5:18) Shensi's journey from finance to tech and the importance of integrations (10:20) The challenges of building and maintaining integrations (11:19) Coda. Empower your startup with Coda’s Team plan for free—get 6 months at https://www.Coda.io/twist (13:17) Identifying the market gap for Merge and the benefits of standardized data models (17:39) Partnership building, Salesforce integration, and overcoming resistance (20:07) OpenPhone - Get 20% off your first six months at https://www.openphone.com/twist⁠ (21:16) Further discussion on trust-building and integration challenges (26:04) Enhancing integration observability and customer support (30:01) Micro1 - Visit https://www.micro1.ai/twist to get 10 free AI Interviews and 2 weeks free per hire (31:21) Expanding Merge's integration list and its AI applications (35:27) Discussing financial efficiency, runway management, and team dynamics (43:16) Growth strategies, customer base expansion, and go-to-market experiments (47:10) Pricing strategies, customer value, and market trends (50:05) Future plans for Merge, acquisition interest, and long-term vision (52:26) Evolution of Merge beyond API focus. * Subscribe to the TWiST500 newsletter: https://ticker.thisweekinstartups.com Check out the TWIST500: twist500.com Subscribe to This Week in Startups on Apple: https://rb.gy/v19fcp * Check out Merge here: https://www.merge.dev/ * Follow Shansi: X: https://x.com/shensi LinkedIn: https://www.linkedin.com/in/shensiding/ * Follow Alex: X: https://x.com/alex LinkedIn: ⁠https://www.linkedin.com/in/alexwilhelm * Thank you to our partners: (11:19) Coda. Empower your startup with Coda’s Team plan for free—get 6 months at https://www.Coda.io/twist (20:07) OpenPhone - Get 20% off your first six months at https://www.openphone.com/twist⁠ (30:01) Micro1 - Visit https://www.micro1.ai/twist to get 10 free AI Interviews and 2 weeks free per hire * Great TWIST interviews: Will Guidara, Eoghan McCabe, Steve Huffman, Brian Chesky, Bob Moesta, Aaron Levie, Sophia Amoruso, Reid Hoffman, Frank Slootman, Billy McFarland * Check out Jason’s suite of newsletters: https://substack.com/@calacanis * Follow TWiST: Twitter: https://twitter.com/TWiStartups YouTube: https://www.youtube.com/thisweekin Instagram: https://www.instagram.com/thisweekinstartups TikTok: https://www.tiktok.com/@thisweekinstartups Substack: https://twistartups.substack.com * Subscribe to the Founder University Podcast: https://www.youtube.com/@founderuniversity1916

Have startups come to you and said, hey, we really wanted you to integrate with us because then we will get our name on your site. Our logo will be in your materials. Are you a distribution point for companies themselves instead of just something that people add on to let them talk better amongst each other? All the time, which I which I fucking love. Yeah. So all these companies will come to us and they'll be like, hey, like I'm, you know, like a 50 person like CRM company and I saw you these customers and I really want them to integrate with us. If you're like a new source of truth and you're trying to compete with like HubSpot or Salesforce, it's really hard to get that roadmap time from every single vendor that is integrating with HubSpot and Salesforce. Like they might not want to spend that time integrating with you. But with merge, it's free distribution.

And again, we're helping create an ecosystem around their product. Is it free? It's free right now. Yeah. Right now. Yeah, we'll just do it because it makes our product better. This week in startups is brought to you by Coda. Coda empowers your startup by bringing words, tables, and teams to,

together. Strategize plan and track goals effectively with all your valuable data in one place. Go to coda.com slash twist to get started for free and get six free months of the team plan. Open phone. Create business phone numbers for you and your team that work through an app on your smartphone or desktop. Twist listeners can get an extra 20% off any plan for your first six months at openphone.com slash twist. And micro one. Micro one is an AI recruitment engine. to hire world-class engineers fast. Visit micro1.a.ai slash twist to get 10 free AI interviews and two weeks free per hire. Hey, everybody.

Welcome back to this week in startups. My name is Alex. I am Alex over on X. Now APIs or application programming interfaces are the software intermediaries that allow two pieces of code to talk to one another. If consumers use GUIs or GUIs to interact with software, well, software uses APIs to interact with more software. And they have become incredibly big business. Companies like Twilio just a couple years ago nearly turned SaaS on its head by offering their products via an API, with pricing

often set to hinge on usage instead of a flat rate. Today, offering an API is table stakes, but as the number of APIs has exploded, so two has the complexity of working with so many different software products, data types, and points of integration. One startup, Merge, is tackling the problem with its own API that offers access to a lot of other APIs. So please welcome to this week in startups. It's Merge co-founder Shindsay Ding. Hey, Shindsay, how are you? Hi, good to be here. All right, I'm not going to lie. I wrote that before we started recording. How did that do? Oh, I thought that was awesome. Ladies and gentlemen, you were all welcome to bask in my presence. You're a natural podcaster. I mean, yeah, your voice was, yeah, it's kind of weird

actually seeing the video component because I'm so used to just hearing your voice. Yeah, well, What's funny is no one's a natural podcaster. You just end up doing it so much. You lose all of your inhibitions and, like, self-doubt. And then you just ramble for a while and then shows come out. It's easy. Anyone can do it. So, Shenzi, I wanted to have you on the show because I had heard a lot about merge,

because I was very interested in APIs, I think, as everyone was back in the 2000, 2021, 2002 era, but a bit like no code to a degree. They became this thing we all talked about. And then the talk kind of faded away and everyone used them. And so it seemed like a good idea to kind of, I don't know, just sit down with you and talk about the state of things. But I want to go back in time to start because the founding story for Merge, I think, underlines pretty well the problem that you're going out to solve. So can you take us a little bit back in the past and just quickly run through how you and Gil,

your co-founder, started the company? Yeah, absolutely. So I'm actually in New York City right now, as you can probably tell from my janky backgrounds, but it started in New York City. So Gill and I both went to school in New York. We were studying computer science. We were in all the same classes, group projects, social group, and we were essentially just really big nerds. We were actually on engineering student council together, class president, vice president. And it was the start of a really great friendship because all throughout college, you would have different class assignments and you'd really find out like the quality, like basically whether people would do what they said that they would do.

And Gil would always do what he would, what he said he would do. And that's super, super, super, super rare. But something that I remained to like think a lot about and also remember, especially when we decided to start a company much later on. But we ended up going down different paths. Gil was like a really great software engineer. I was more okay. I mostly did it for fun. I think it was intellectually stimulating, but it wasn't something that I was going to be uniquely good at. And I did feel like one gap that I had was understanding the business side. And I had no idea how to evaluate whether a company was good or not, whether an idea was good or not, how to understand

financial statements. And I really wanted to learn that. So I went into invest in banking in New York City at Credit Suisse, RIP. It was a great two years. Really started understanding how to read financial statements, how to read 10Ks. And it taught me a lot about businesses. But I really wanted to go back towards tech. So moved to San Francisco, joined Silver Lakes Growth Equity Fund, and reconnected with Gil, he helped me find my first apartment, got my first car. And I realized that investing was not close enough to actually operating. A lot of times, yeah, and I'm sure, yeah, it's just really not quite the same experience. I would read a financial statement or like a C model and just make, oh, just increase revenue and

decrease costs. Simple. Why doesn't everyone just do that, Shenzee? Come on. Exactly. And we would go to some of the board meetings and I just didn't really understand why that was so hard to do. And so I really wanted to actually work at a company, understand what it was like to do that.

And I interviewed out a few startups and I found a company called Expans back then, called KDM, and I joined as the chief of staff to the CEO. Pause there, because this is off topic, but I really think that the chief of staff role is not particularly well understood outside of folks who have either had one or been one. So just for everyone out there who's heard chief of staff, what does it mean? Yeah, I mean, in essence, you're basically the CEO's bitch. I know that's hard. Like, there's no other way to word it.

You just do whatever they want so that the company can be successful. And I think it's quite popular in situations where, like, there needs to be guaranteed execution. And my hot take actually is, I don't think that if there's a chief of staff, usually I think there's actually an operational problem at the company because it means that you can't guarantee that when you hand something to someone that's going to get done. And so a lot of my job was just like making sure things got done and like, you know, documentation and like random, like, I did like random recruiting stuff. I like helped out of hiring his executive assistant. But it was all just like random things. But the essence of it was just making sure that when he

wanted something to get done, it got done. The joke about this is I'm so glad we invented the term Chief of Staff, so that way men can now be secretaries without using that title. And I thought that was hysterical. But anyways, Chief of Staff are very useful. And if you ever work near a CEO of a company over, I don't know, 15, 20 billion value, you'll run into them. They make sure everything works. There's essentially like the COO for the CEO is the way that I think about it. I think so, but probably less strategic. It was more execution oriented. And but I will say the one benefit of it was I got a ton of exposure to the CEO. And Tim is, I don't know. Actually, I think he was in this weekend startups a few years ago because I prepped him for the

meeting, or I popped him for the interview with Jason. And yeah, he's just brilliant. Like, every single thing he knew about politics. He had a PhD in cyber warfare. And especially in like 2018, 2019, it was a really interesting time to be learning about tech and politics and also how it impacted, you know, different global dynamics. So yeah, it was just a really interesting experience for me. And every time the company would run into something that was potentially detrimental to the business, he would just figure out a way to solve it. And that was, that was really cool to see. But it was at expanse during this time of your work life that you noticed that essentially integrations were becoming an increasing sticking point for the business. Yes. So one benefit

of being chief of staff was I would join a lot of executive meetings and also sensitive meetings. And every executive meeting, we would end up talking about things that were causing us to lose deals and what competitors were better than us at. And starting after like the first year or so that I was at the company, integrations coming up is a reason why we were losing deals. And we provided ourselves on having the best products. Like, we were really proud of that and we were really confident in that fact. But our competitors that had a less high quality product were able to win just purely based on features, like specifically integrations.

And our sales team started really like just yelling at us being like, we need more integrations. Why don't we have any? This is so table stakes. Let's just start building them. And since I was on the finance and operations team, I was really seeing it from a P&L perspective. If we didn't build these integrations, we couldn't close deals. And that was going to hurt us on the revenue side. But building these integrations were going to be super expensive because we would have to hire San Francisco engineers and it would really impact operating expenses.

And then even if we tried to move to a lower cost area, which we ended up trying to do, it's still a lot of engineers and you have to permanently maintain those integrations. And it becomes a permanent line item that you can never get rid of because there is a maintenance component to integrations that a lot of people don't think about. It's not like you can just build it and let go of it. You have to continue maintaining them. And so from a P&L perspective, I was like, this really sucks. Like, no one's really solving this problem. I don't understand why this is so hard and is really building these integrations that difficult.

And it was really fortunate that at the same time, Gil, my co-founder, he was head of engineering at a recruiting tech company. And they were pretty lean. They were Series A stage. And they had like 10 engineers or so, but everyone was busy. They didn't have enough engineers. And so he ended up having to build the integrations for his company. By himself.

By himself, because there's no. Yeah, there was just no one else. And because of that, he actually had a really deep understanding of why it sucked so much. And I asked him, I was like, why it can't be that hard. Like, I was fun. I was a software engineer. Like, you just like read the stuff. And he was like, well, first off, we had to get access to the API. And we had to get a partnership and that took a while. Then we had to get access to a sandbox account. And that costs us some money. Some API providers are nearly impossible to get access to because they're closed gardens. And so you might end up waiting years to get access. And then once you actually get access, then the product manager and the design team has to figure out what this integration looks like and what that. workflow could be. And then they started building the integration and they built it into the wrong API because some API providers can have multiple APIs for different functionality. So they spent three weeks on it, ripped it out. I had to do it again. And then once they bought it by the way of engineering time is a lot of dollars. Like it's not just three weeks. It's an enormous amount of money and opportunity cost. That's a lot. Yes, exactly. And then actually to finish the full end-to-end correct integration, it took six weeks or so. They were a lean series A team and

Gill is brilliant and he, you know, he worked really hard and fast. And it still took six weeks. weeks. But what he said was the worst was that actually when they went live at the integration, the engineers actually had to keep helping answer customer success questions because customer success team members would get questions from recruiters who were their customers being like, hey, like, I gave you an API key. Like, why is my data syncing? Or like, why is Joe's name missing from this candidate? Or like, why is his resume not uploading? And a customer success team member isn't going to know all the answers to these questions. And so they're going to escalate it to an engineering team. So the engineering team ends up having to allocate a percentage of their time

continuously to help customer support, troubleshooting, and then a lot of the times, it's end-user error. Like, they gave you an API key that's incorrect, or someone got fired, and their API key no longer works.

Right. And all of this was really just not handled by any piece of software at all. Listen, are you spending too much time as a founder all tabbing between your team chat, maybe a document editor, spreadsheets, database as well? It's time for you to consolidate all of that knowledge

into one platform. And that one platform that I use, that I use every single day is Coda. If you don't know Coda, it's like a new category of software, best described as like a collaborative workspace. It pulls together all the stuff you got going on in documents, spreadsheets, maybe a database, maybe a built an app. And it's super easy to learn.

It's incredibly powerful. In fact, we use it and we run Founder University on it. Then we had a new project, twist500.com. We wanted to make a database for the listeners of this podcast that essentially, profile the top 500 private companies. It was a no-brander. We said, oh, we should make a website. So I got twist500.com. And I was like, what are we going to do this for? Everybody in the meeting was like, Coda, Coda can do that. And that means I don't have to buy new software. It means I don't have to hire a dev shop. I can just do it myself. Coda empowers your startup to strategize, plan,

and track all of your goals effectively. Take advantage of this limited time offer just for startups. Coda.io slash twist today to get six months free of their team plan. You're going to save at least a couple hours a week per team member, and you're going to feel like you're in control. That's Codda.io slash twist to get started for free and get six free months of the team plan. Coda.coma.io slash twist. Okay. So you're sitting in meetings noticing that deals are going out the window because you guys didn't have the right integrations or as minis you needed,

even though your product was better. your co-founder actually had to go out and build a bunch of these things, seeing kind of upfront the issue. Clearly, you guys both saw the pain point, but you couldn't have been alone. And to me, what merge has built, and we'll get into unified APIs in a second, is such a reasonable and smart solution to this?

Why hadn't anyone solved this already? APIs weren't new when you founded the company. The problem wasn't new. So how would the market not close this gap, shams? Well, at the time, there were some really, really popular companies like trade I. and Zapier and Ricado that were just super, super prevalent.

And I think the belief at the time was that the onus of building integrations is actually on the buyer and not the vendor. So it was still pretty common for Expans, for instance, to sell to, I don't know, some Fortune 5 like AT&T or something. And then AT&T purchases Expanse, and then they also purchased Splunk,

and then they also purchased MuleSoft or Orcaut or Tray, and they connect, Expans, and Splunk. And then you now have three pieces of software to make two connect. And it's just kind of a pain. And you have to maintain that integration forever while also spending like $30 to $100,000 on Mule Software Rikato or Tre.

So that was just a really common workflow. And we would talk to, and Gil's company at the time would actually try doing that a few times. Like they'd be like, oh, like we don't have an integration with like workday, for instance. So you should buy Tray. And that would end up still blocking the sales cycle because it's an additional amount that you have to spend. But at the time, it was still common.

And software wasn't that fragmented. And it was still pretty common for companies to just purchase like a whole suite of like Oracle software or like workday and like SAP and then you just wouldn't really need integrations. So I think integrations just weren't that common at the time still. So that's when I really thought about it. And I think even if they did think about it, they were like, well, you can still

use a workflow provider to kind of do what you need to do. I see. So essentially, at the time, software was less fragmented. People had existing solutions to a degree, but as you found out, they weren't enough. So you went out and you found it merged and the idea if I boil it down to like the absolute nub is that instead of having

6,000 different APIs to integrate with things, Why not have one API that is the conduit for all the other APIs? It brings them all together into one pipe, essentially. Yes. And one nuance is, so technically is one API because the credentials are the same and the pagination of rate limiting is all the same. But we do have different objects or data models per category. So for instance, like opportunities are unique to CRM and you can't, you can't, there are

no opportunities in HR systems. So it's per category. There are normalized data models. Yes. And that allows you to, I presume, have more total partners because if you have normalized data fields, then it's easier to probably link in new sources of information. Yes. And fundamentally within every single source of truth, the data model is very, very similar

because humans just want to simplify and think about things in a similar way. So among like Salesforce and HubSpot, like, there are stages and there are opportunities and like they might order different things, but fundamentally they are the same thing. Same with in recruiting systems. There are candidates and there are applications and there are stages and notes. Like there are some, in some API providers might be missing some of these components, but fundamentally, there are core common models that everyone just boils down to. So I want to touch a little bit more on the people need integrations to land sales, because you mentioned software fragmentation. And to me, that sounds like so many different companies out there have a different mix of software solutions that trying to sell them something new is going to be essentially impossible unless you have every single, you know, an incoming port, if you will. So how many like pieces of software do companies have today? How many are they trying to tie together?

I don't quite understand how much of this is a day-to-day issue for companies. So I think the number now has increased around like 200 or so. I saw like a recent report and like the vendor curment about like how much software like each company has. And for us at Merge, like we're 110 people. We have like a ton of software providers that we use. Oh, I'm sure. Yeah. And we expect every one of them to just sync data with each other.

Like it's a huge pain to have to download a CSV and re-upload every single time there's a new update. That's just going to be impossible. especially if millions of rows of data are getting updated every day or every second. Like, that's just not going to work. And because the expectation now of the buyer is so high, like, there's so many different vendors out there. Like, why would you choose a vendor that you then have to do that manual work for? You just want to purchase something that just you click a button and it seamlessly

syncs with all of your other vendors so that all the data is consistent across all of them and there are no data silos. That makes perfect sense to me. But I'm curious about the people you partner with. I mean, for example, you guys announced very recently that Merger is now officially a paylossity partner. Is there any tension between you guys providing a unified API and individual companies that offer their own API and perhaps them not wanting to be aggregated politely into someone else's service and instead have more of a direct connection to customers?

So when we first got started definitely and it took a lot of evangelism, I think I reached out to like someone I hadn't talked to in like six years to like get in contact with like a private equity firm that was on the board of like one HR platform that was a closed garden and then I ended up meeting. You reached out to an old friend who was. Wow, that's a very ten-year-old. I wouldn't even call them a friend. But I had to, you have to do whatever you have to do. Oh, that's true. But you have to do what you have to do to, like, because in the end, a lot of partnerships is about, like, networking and just shooting your shot. And so at the time, we were five people. There was just no way I was going to be able to get a partnership with this, like, you know, 10,000 person company unless I really did whatever I could to get in front of like the C-suite. And so, yeah, so that's what we had to do. And it took a lot of evangelism, but that became one of our best partnerships because the C-P. CTO and CEO all like felt and understood what merger's trying to do and we were trying to make it easier for them to have an ecosystem around their product. The dream for every source of truth is to do what Salesforce has done. Like it is impossible to not use Salesforce because every single

sales ops tool is built on top of it. And if you don't build on top of Salesforce, it's like if you don't use Salesforce and you're kind of screwed because there's just so many tools that are around it. Like sure you can use like some of these really long till providers, but it makes it really different. It makes it harder. That's why regardless of whether you're in recruiting or HR or help desk or ticketing or project management, like you want companies to get built around you because then you become a source of truth and you're critical and it's really hard to turn.

Yeah. And if you want an example of this, think about, I think the company was Encino, which was the first Salesforce platform company to go public. So they built an entire public company just on top of Salesforce's platform. And that's great for Salesforce because more data information and business flow through their operations. Ergo, Mark Benneyev, It's added another 5,000 people to Dreamforce to ruin the city slightly more quickly. I thought it was fine. I'm sorry, I lived in a Ness F for a long time.

And, like, there would be this weird week when you go out to the bar and then you would get there and it'd be a little sign out front that said, close for Dreamforce private event. And you would just hate those people because that's your bar. How dare they? Honestly, okay, so when I lived in San Francisco and I would, and it was Dreamforce week, I felt the same way. But then now I'm like old and I went to, I went to Dreamforce and then I went to Imagine Dragons. And I was like, wow, I've officially aged because this is like the most.

It was fun I've had all here. Okay. Look, Imagine Dragons is not great, but they did have some bangers for a League of Legends, so I'll take it. I'm not going to judge. Speaking of nerdy things, I guess I do know the soundtrack from League of Legends and which bands made the songs. They have a lot of bangers.

Yeah, I forgot too, but they have a lot of bangers. Are you still using multiple devices and apps to run your business? Well, you need Open Phone. Open Phone has rethought what the modern business phone should be, and what's so magical about Open Phone is that it works through one. single elegant app right on your existing phone and you can use it on your desktop. Now listen, open phone is so good. I have my entire team on it. The sales team loves it. Our ops team loves it. Why do they love it? Well, they don't have to mix business with pleasure. They don't have to use

their personal phones anymore to talk to their customers and clients. And it allows you to text as well. Over 50,000 businesses use open phone today. It's not like those old phone systems that you used to install at your office and it cost you $500 to $1,000 per person per year. No, just $13 a month for open phone and twist listeners get an extra 20% off any plan for the first six months. Openphone.com slash twist. And if you have an existing phone number with other services, no problem, easy peasy, lemon squeezy, open phone will port them over at no extra cost. So head over to openphone.com slash twist to start your free trial and get 20% off. So you found it merged back in 2020, which is not that long ago.

frankly. I know in technology terms, it's a couple of cycles. But companies at the time, I presume, were a little bit skeptical because you were a small company. As you said, how fast did that change? How fast did you guys go from, you know, going through friends and around back doors to being able to knock on the front door and kind of get a deal done? Because you have a lot of integrations. I hope it's gotten easier. It was really, really hard. And it kind of sucks because now, of course, we see a lot of copycats who benefit a lot from the evangelism and the education that we did back then. But it was really, really tough because a lot of people were skeptical that we would be deep enough, that we would be able to solve their use case. And also, I know a lot of people, there's a lot of composable software, but like every single time there's a new idea for some type of composable software, you still, people are still skeptical. And I felt like what we were doing was very similar to convincing people to move from on-prem to the cloud because a lot of people felt like, no, like, I want this like in-house. Like, I feel like I can uniquely be good at it. And I would really have to convince them, like, no, like, the only thing that can be different is you're worse at it. Like, you're not going to be uniquely better at this.

It's something that is just not your core competency. And other things that are really tough were just like the business model, like we had to really understand what to do. It's we're not a transactional like API. Like I think with like Stripe and like Twilio, it's a little like, okay, one transaction or like one email, one text message. So it's a little bit easier to figure what like what pricing looks like and what and customer support might be a little bit easier.

But for us like all of it was so different. And like the closest comparable we had was plaid. But even still it was different because they were consumer, mostly consumer focused, and we were B2B, exclusively B2B. Yeah. So, yeah, a lot had to get, a lot had to get figure it out. And the customer buy-in process was really tough in the beginning. But that's kind of the value that you guys offer.

One thing that APIs do is the abstract away complexity. So this is why people love Twilio. You could just use their API, send SMS messages anywhere. You didn't have to know anything about telephony. They handled all that complexity. And in the case of Merge, you guys have gone out and talked to all the accounting providers, all the HR providers, et cetera, and done all the. work to get this set up. And that forms essentially your moat in a way. But if people can follow

up based on the work you've done and copycat you, does that, to keep the analogy going, does that drain the merge moat? So the benefit of merge is actually, so you know how I was talking about earlier, how there's like a long maintenance period and it's a huge pain? So there's a maturity curve for every single integration and also every single category that we have. And so one of the benefits of us being first and us just having so many customers onboarding and every single time our customers and their customers onboard, we find out more edge cases for what can happen in these integrations. Because API documentation is always hardcoded. Like, it's just not always updated. And response bodies from APIs can vary wildly. Like sometimes you think something will come back

as a number, but it actually comes back as a string. And then boom, tomorrow, it's an array. And you're just like, what the fuck? And it breaks your code. But for us, because we have so many customers onboarding, we just have so many data points of what those edge cases could be. And so by the time, our customers onboard onto a really gnarly, like, energy. enterprise integration, they're much less likely to run into a bug than if they had built it in house because at the first time you have a customer onboard, you're going to run into a bug. There's just no way it's going to work perfectly. So I was preparing for our chat today and I was thinking about the ability for merge to

see data in motion just because of how many different pipes you guys are running between different products. And I was curious if there was a kind of a data observability element to this. And I wasn't going to bring it up because it seemed a little bit off kilter, but you just made the point for me. Can you talk about how merge is, from my view, kind of like a second order data observability firm? So we actually call it integration observability

because it's specific to, yeah, specific to integrations and also the customer support side. So actually a lot of times, once the integration is live with our customers, customer support teams mostly interact with merge and our support team. The reason why is because we want to make sure that every single piece of data that is flowing in merge and out of merge is visible to our customers.

Because we are specifically focusing on B2B, it's just the stakes are really high. Each end user can cost millions of dollars. And so it has to be extremely reliable. You have to know everything that's going on. And you don't want anything to be like what is going on underneath the surface. Like we want it to be very, very clear and transparent. And so we built a lot of tooling in our dashboard. It's mostly focused for more non-technical people.

So like we have every API request. You can modify scopes and what data you want flowing in and out is fully searchable. So you can search like for Alex's name if Alex came through or not. You can store it. Like all of this information is very visible. Same with like there's end user error. You can see it very easily in the dashboard. It's kind of like simulating what people would use data dog in century for when they

were building integrations in-house because you would need to use both components in order to figure out what was happening with your integrations. But engineers would have to look at that. And with merge, it can be a customer support person. Okay. I just realized we've been talking for like 25 minutes and I have to move us towards business stuff in a second. Okay, of course.

Two quick. Sorry, I love learning and I'm learning a lot. So thank you, Disney. I want to talk about two quick things before we move on. One is this product called Blueprint, and it's a way that you guys are allowing people to essentially, it seems like crowdsource integration recommendations. What's the idea there? Yes, to be honest, it's mostly us just playing around with AI,

just less playing around with AI and seeing how potentially our customers could be able to contribute to our integrations down the road. And it worked. It was pretty cool. We were able to get documentation and that we would be able to generate initial mappings. It's obviously not going to be perfect. a lot of building integrations is unfortunately very subjective. It's not objective.

And so AI isn't quite there yet. Got it. But it was really good for us to be able to see, okay, maybe we might be able to expedite part of the integration built with AI. But it's not like a full-blown product. One day, maybe, but right now it was mostly just like for us to experiment with it and see if our customers could start helping us out. And did they?

Yeah, they actually did. So a lot of people started submitting different integrations. It was also helpful for us to gather interest in which ones. Unfortunately, the main blocker, for building integrations is sandbox access. It's not documentation. So even if you have a great, there's great documentation out there

and you feel like you know how to map it, if you don't have a sandbox account to like test around and like modify the UI and see how the response body's changed in the documentation, the integration's not going to work really or it might be okay, but you'll run into edge cases.

So that's the blocker, even if we get documentation. Why is sandbox access hard to get? Because by definition, sandboxes are safe environments, ergo they shouldn't be risky to let people tinker with. So in my view, the bar should be very low to get access to them.

And yet several times today, you mentioned that sandbox access is an issue. What's going on there? Because it requires a pretty advanced partnerships motion. So for instance, like, because you need to be able to issue an account to a partner and not build them and understand that the purpose is purely for testing. Isn't that just the button you click? Like, do not charge. Click.

And then, like, this sounds like something the software could solve. A lot of categories aren't self-serve. Like, for instance, like HR tech is just not self-serve. like usually you end up having to sign. And like if there's payroll information, you need to make sure you don't actually process the payroll. Okay, that would be,

that would be funny. I got paid 47 times today. Yeah, some categories are really great. Like, ticketing and help us, like it's just always self-serve. And so that's very easy for us to get a sandbox account.

But for instance, like getting our Salesforce sandbox like took a while. Like you need just like, you began to like fill out a bunch of forms. I'm shocked that sales force was a process. That company known for his bureaucratic efficiency and lack of unnecessary, layers and VPs. So really briefly,

merge definitely when you add a new integration, you make a big deal out of it. It's, you know, it matters quite a lot. So the pay loss you won from earlier, have startups come to you and said, hey, we really wanted you to integrate with us because then we will get our, you know, our name on your site, our logo will be in your materials. So are you a distribution point for companies themselves instead of just something that people add on to let them talk better amongst each other? All the time, which I, which I, which I, I, I, like, love. Yeah.

So all these companies will come to. And they'll be like, hey, like I'm, you know, like a 50 person like CRM company and I saw these customers and I really want them to integrate with us. So can you please build an integration with us so that our logo will show up on their website and then all of our customers will be able to benefit from using that product? Because if you're like a new source of truth and you're trying to compete with like HubSpot or Salesforce, it's really hard to get that roadmap time from every single vendor that is integrating with HubSpot and Salesforce. Like they might not want to spend that time integrating with you. But with merge, it's a

it's free distribution. And again, we're helping create an ecosystem around their product. So startup comes to you. Hey, hey, hey, hey, hey, you know, we are a 50-person up-and-coming CRM firm. We would like to be listed and we kind of take part in the merge ecosystem. And you say, how much? Is it free? It's free right now, yeah, right now. Yeah, we'll just do it because it makes our product better. Knowing what to build is only half the battle. You also need the right team. And recruiting is so hard when your startup. You're competing with all the other bigger players. So you might need a little help, and that's where Micro One comes in.

Micro One helps you scale your product team in days, not weeks, or months. Think about that in days, not weeks or months. And they handle the search process, vetting, even onboarding and paperwork. Imagine you could tell your AI your precise engineering needs, and it could deliver exactly what you need right away. Well, Micro One has built an AI engine that interviews tens of thousands of engineers, designers, and AI trainers every month, then they can pick the top 1%

and they'll help you onboard them into your startup while handling all the legal and compliance work. If your startup needs to accelerate its product development, but you don't have the time to vet mountains of resumes or the budget to hire recruiters, Micro One has you covered. All you have to do is tell Micro One what you're looking for and they will help you find your next engineer in 48 hours or less.

Here's your simple call to action. Visit micro1.aI-i-S-T-T-T-S-T-T-S-T-T-E. That's M-I-C-R-O, the number one. a. Slixt to instantly wrap up your product team. And Twist listeners get two weeks of free development per hire plus 10 free AI interviews. That's micro 1.a.ai slash twist.

To me, someone coming to you asking for access is usually the point of which you can extract some value. And so I presume right now there's more value in not charging than cash. I think one day, but right now I still really want to just make sure we have the most comprehensive list of integrations. And it's definitely been on our mind. Like we've experimented a few times with it. but right now I just want us to have the best product.

And that means just having the most integrations and the highest quality integrations and product. How far are you on getting to the point in which you go like, okay, cool. We have all the big integrations we need. Now we're going to go chase the last edge cases, but we've done 80% of the work. Never because there's different regions and the world is just getting more international. So for instance, like a lot of our customers are moving to different regions, like whether it's Europe or Latin Am or Southeast Asia or Japan.

And so there's an unlimited number of integrations that you have to build. So never. So I don't know, not to be rude about it, but like, are you guys getting faster and better at building these integrations? Or is that not something that you can actually accelerate, given how much you depend on other companies offering access to sandboxes and so forth? Oh my God. Yeah, it's the only thing we do.

And so that's one of the benefits of it being the only thing we do. Well, now you make me feel silly. I wasn't sure if you could speed up the process because you mentioned these other roadblocks, but okay. Yeah, because it's the only thing we do. We focus so much on like creating efficiencies with like how we test. Testing especially is the part that takes a lot of time. But yeah, we spend a lot of time on just making that faster, more efficient and, like,

higher quality for our customers. And then also whenever we make modifications for maintenance, it's also very fast as well. So it's like mostly what we do is just making sure we're able to move very quickly. But merge overall is not just like a technically difficult product to build. It's also like an operationally difficult company to build. So we from the ground up because we knew we wanted to be cross category. We knew we wanted to be like international. We just really wanted to make sure from the beginning we were building the foundation

and avoiding as much tech debt as possible. Well, and you guys had security on the mind early on. If I recall correctly, you guys had like SOC2 compliance, like out the gate. Yes, and that's all credit to my amazing co-finer. Yeah. All right. Now, one more product question before we talk about money. Merge for AI.

Build your AI features with LLM ready data. My question was, is the merge AI product for companies to ingest their own data to a single point so that they can train their own AI features? Or is it for companies that have many customers that have different integrations to bring that data in to make an AI feature for their customers? So the latter. We never ever do any internal use cases. Like the price point just doesn't really make sense for us.

But yeah, we have a lot of companies that I have gotten started in the past few years, especially with the AI boom, that want to be able to have more customer-specific data in their products. So they can provide insights for like whether it's AI search or like AISDRs or like, I don't know, you name it. And they'll use our integrations to help enhance, their product insights and also capabilities. And how big of a demand point has that been for Merge lately?

So honestly, it's, so it's been really, really prevalent in the S&B segment. Like, we just see a lot of companies just getting founded. Pretty much like 100% of us companies are all AI. It's a little bit harder to tell in the enterprise segment because you can't, like, everyone's just adding AI to their products. You can't tell if it's like AI specific or a product that has like a little bit of AI. But on the SMB segment, it's like super, super, super relevant.

The way this has been explained to me and the way that I've learned it, is the more unique data you have to build your model, the better will be, the more differentiated it will be. And the more of your own data you have inside of an AI predicated application, the smarter will be, the more you can do with it. Ergo, unique data is this incredibly important thing. If everyone's moving towards using more AI, does that mean that the total throughput of information across APIs that merge touches is going up incredibly quickly? Because it seems like it should be. I mean, that's what I think. So we've been seeing, we've been seeing a lot more companies going from using mostly just one or two categories to all seven categories

or and asking us for more. Seven categories of integrations with merge. Yes, exactly. Yeah, because the more data you have, the better than insights you have. Essentially, the AI boom,

even though it is, I mean, not exactly what you guys have been working on. It is providing a nice tailwind to the business. Yes. Yes. As my CRO says, it's always good to be lucky.

So we have, I'm very lucky. Dude, better lucky than good is one of those things that I've learned. is more... I learned that lesson again and again as I've gotten older. And so shout out to you guys for being both. All right, now, let's talk about money.

So, according to CrunchB, Merge has raised $74.5 million. Last round was a $55 million series beat October 22. And Pitchbook had that listed at a $315 million post-money valuation. So that's a couple years ago, two, in fact. The obvious question is simple. Are you guys raising more money? Not right now. We've been really efficient.

And to be honest, we haven't spent most of our money from her series. be. One thing that I really invest in on early on just because my finance and operations background was I wanted to have a really good finance team. And so one of my old coworkers joined us as employee 10. And she was just like tiger momming all of us about not spending too much money and being really efficient, especially in this current market. So not right now. We have we've a long runway right now with how much we're spending. And she's actively making sure we're keeping her cash and check. Is she like a CFO equivalent? Basically, yeah. Okay. So you have a CFO equivalent and you also have a CRO and you've only raised three rounds of capital.

That is pretty darn early for a startup of your size. We only have one C-suite member or one C-R-O. Everyone else is a VP. Oh, okay, fair enough. But I mean, you have definitely a more, forget the titles. You have a much more mature finance op than most companies do at your size. Oh, for sure, yeah. For the founders who listen to this, that's the main audience for a twist.

Just explain how you went about that costs and benefits. I think people would love to know what an early investment in this. the office of the CFO broadly does. So I hired Alexia when we were 10 people because there were just, there were a lot of operations and accounting things that I just wasn't going to be uniquely good at. And I also knew I wasn't going to be super strategic there too. We were starting to enter a phase where we might want to have more advanced pricing. At the time, we were just like flinging numbers around to customers and that that just wasn't

going to scale. And I knew we needed to be really thoughtful about it. And if you hire really good finance team, they can be revenue generating for you and really strategic in like how you expand to different products, like how you expand different regions, how you're thinking about like costs, compensation. And so we hired her so that she could really own all of that. And she ran lean through our series A. And then she started expanding her team after our series A. After your series B or series A? Series A. So then she had two people. Oh, okay. She didn't, she didn't really expand to three people until series B. But then we hired

also a VP of RevOps and then also two RevOps team members because our sales team started expanding a lot as well. In general, for a series B company, our finance and operations team is definitely pretty loaded. But I think just given how complex, like, our operations are, it's really been an advantage for us. And we have deep insights and understanding of our sales motion and we're, like, a pipeline and just like what the company looks like. When you raised that last round, that $55 million round, that, by the way, Excel led that, if I recall correctly. Yes. Yeah. Okay. It's a big round. I mean, 55 million. even for 2022, was it relatively outsized.

Were you guys at the time intending on being very conservative with that new investment when you raised it? Always. We always wanted to be really conservative. Like with our series A, we were really, we actually hadn't even touched our series A by the time we had fundraised. We've always been pretty careful about it. We knew we wanted to make some investments in some strategic areas with that $55 million,

but not definitely not spending all of it. Why were you guys so early to the, what if we didn't burn a bunch of cash startup movement? Because it feels like you raised a erotypical round and then did non-arotypical things with it. And that's an interesting dissonance to me. So the company that I used to work at Expans, they were pretty lean to. They got to, they got acquired at a billion dollars like 200 people. And they mostly focused on like really large enterprise customers. And now that I'm thinking about like how, how it ran, it was pretty efficient.

And they didn't, they didn't spend a ton of money. And I think because that we had a benchmark for what metrics might want, might want to look like. And we started in June 2020 when it was really hard to fundraise and like no one believes in us. True, true, true, true. We just knew also Merge was going to be an expensive business to build because everyone has to be technical. We were building San Francisco, New York City. And I would much rather just never have layoffs. And I really wanted to make sure that like as much as possible I could protect the team and just like hire slowly and make sure that everyone is really good and have fewer better people.

But it's also hard because we're like we're an in-person company and that automatically limited a lot of what the pool looked like. so it was always going to be slower for us to hire. Like, we would never, even if we want it's like 2x tomorrow, that could just not happen. That would just not be able to happen. That's actually the funniest thing I've heard in some time. If you are an in-office company, your hiring rate may be slower because your candidate pool is more constrained. So if you want to save money by not hiring people, return to office.

Jason's going to love that. That's like half who we've been talking about the show lately is like returning to office. I refuse, but I do respect every company's right to make their own decision. It's funny that I hear you talk about this. Merge was going to be expensive to build. lots of technical talent. We're in New York. We're in San Francisco. We're in person, which means you have office expenses. And yet you hadn't spent your Series A by the time you raised your B. You still have lots of your B two years later. It's funny to me that you manage this when most people don't seem to be able to spend less and keep growing. Here's an example of you guys pulling it off. It's just rare to hear this actually work, I guess, out in the market and not in a blog post. We have a really good team. And I think also just because like my team were so close. And that's probably from just like, to be honest, being in person, they feel comfortable calling me out, like, hey, like, we invested a lot in this. Like, we need to stop investing more in it.

Or like, oh, like, right now, like, where expenses are like this, you know, like hold up. You know what you mean? Like, they, they force me to be really thoughtful. And I think because we're so tight as an executive team, yeah, we can really call each other out on things like that. For everyone out there's listening and thinking, oh, my gosh, they haven't spent a lot of money. Surely this company hasn't grown much.

I went back to the tech runs coverage of the series B. And you guys has said that at the time, your error, annual recurrent revenue, had grown by 30x in the last 12 months and then at the time, 2,500 companies now use their service to integrate SaaS apps. So I'm curious in the last couple of years, how has growth been? It's been really good. So we've obviously tried to recompose what our revenue looks like a little bit just because when we first got started,

it was a lot of micro S&B and smaller companies building on the platform. But right now, I think we, to be honest, I haven't checked the numbers because I don't look at this as much now. But we have like, I think like 16,000 self-serve organizations on the platform, free and paying. We have around like 400 enterprise customers on the platform and also the size of them are just increasing. Yeah, it's been a really transformational past two years

and that series BWRAs really allowed us to like take the focus more on investing in that future and what we wanted the company to look like and also just making sure we could focus on product quality rather than continuing to expand the TAM. Because one limitation with our space inherently is just TAM, there's TAM constraints. Like when you first get started, you decide,

am I going to go broad or am I going to focus just on like one or two categories and you focus on one or two categories. You just, there's not enough companies out there. You can't outbound enough companies and you can't get enough revenue and then you die. But if you go broad, then your product's super shitty and then you'd also die. So yeah, the raising the money was just really great for us to be able to just double down on quality and make sure existing, like, because we'd already expand our Tam enough and just making sure our customers were happy. So essentially you would expand it to your Tam by going broad, having more capitalized you to then go deep where you had gone broad and therefore you're not spread too thin. Yes. And we had kind of like

alternated. Like, first we went really deep in like two categories and then we went a little bit broader and then we went deeper and then we went broader and the deeper. And then we also had the benefit of just being first. So we had time. Overall, like just we've been spending a lot of time on like making sure it's been like efficient growth, especially just because of the market, we don't know where things are going. And I just really want to protect the team and make sure there's a good outcome versus like overfundraising and overspending and being like, okay, we ball. Like I don't want to do that to the company. This is the only company for me. Like I, this is it. So I really want to make sure I'm like being responsible.

responsible. So, but I want to go back. You very politely dodged around my question by saying growth has been good. Talk to me about 2023. How is this year looking? I presume you didn't 30x your ARR again. Oh my God. No. Yeah, of course not. Yeah, definitely not. It's been really, really good. We've like doubled the number of logos that we've had. We've expanded a lot of our like product offerings. And there's some really big logos that we're hoping to close in Q4 that we just never would have even had conversations with last year. If those new deals land in Q4 that they're hoping they're hoping they're, like they're hoping they're, will. How much does that pull your ACV up? A lot in the enterprise segment, but probably not overall. Well, I mean, yes, because you have lots of smaller customers, but in the enterprise segment, would it be like a double-digit percentage increase in ACV? Unclear. Yeah, unclear. Okay. So, because one thing that struck me was back when you raised the series B, you guys were talking about going more up market, and I was really curious to see how well that's gone. So is the majority of merge revenue today from the enterprise segment? It's growing a lot. Yeah. when we first started, it was 0%.

So, like, none of them will use us, like, at all. Like, that'd be chaotic. Like, why would they ever use our shitty little product? But now, yeah, it's been really exciting. We've increased percentage quite a bit. And then, you know, in the enterprise segment compared to these kind of micro enterprises you were talking about earlier, are they in the same kind of overall business categories or has going to the market also opened up a new kind of like industry sector mix to the company?

Definitely different industry sector mix. There's certain companies that will just never be able to set. I can never sell to like Chevron or like Target probably unless they have a B2B SaaS component. But yeah, like financial services are now open to us and that just was not possible for us before. So, you know, right now you guys have capital. You have, it seems to be a pretty strong product market fit.

And the economy seems to be okay. Have you thought about changing up the way you guys approach cash and just trying to grow a lot faster? No, because I really want to make sure that if I spend, if I spend X, I know exactly what's going to come out of it. And it's not super clear all the time, especially with this current market. So I'd rather us just be a little bit more conservative and spend X and know, like, it's probably going to be around Y versus being like, okay, it's going to hit C. So I'd rather just make sure we're being smart with money. But thank you for saying that because I wanted to talk about your

go-to-market approach because I was going just literally reading back through historical merged tweets. And one thing that hit me is you guys are pretty active. So you sponsored product school. I'm just going back to these. You went to SaaS stock. And then if I went through, there's just tons more events and so forth. You guys are out there really. Tired. Yeah. Well, I presume you're exhausted. But like, the approach seems to be present at industry events and kind of going around. And so I presume that the ROI on literally pressing the flesh has been good. So this was a big year of experimentation. And a part of this $55 million was for us to be able to experiment with the go-to-market motion. So part of that was going to be

events and us going to different regions and us just like meeting more customers in person. So yeah, definitely. And part of the, yeah, we've just meeting a lot of people in person now. And I presume that in-person stuff does a little bit better when people understand what you're building. So I presume unified APIs have become well known enough that there's less education required in person. No. There's still a lot of people who have no idea.

Yeah. And also like every time I go to a wedding, someone's like, what is this? And then I have to spend like an hour explaining to them. So most people don't understand it. And that's totally fine with me. But I just want to make sure that it is potent. And that's why some of the in-person interactions help a lot. Wait a minute.

At weddings, you talk about what your startup does? Oh, my God. Yeah. It's like I go through like a due diligence process every single time I go to a wedding. Like you should do what I do at weddings, but just slowly find the back wall and then attach myself to it and then don't move for three hours. It works every time.

Okay, noted for the next time. All right. I want to talk about pricing really quick. And then I'm going to do just a quick reflection. So I was just trying to figure out how you guys go about pricing for Merge. And in the first of your three tiers, you get three free linked accounts. And then it's 650 for up to 10. And a linked account, I believe, is a customer endpoint integration into the API, correct?

Yes, pretty much. So it's a connection. So for instance, if your customer is Lyft and they're connecting their workday instance through Merge, that's one connection. And then you've another customer that's also using Oracle, then that's the second connection. Okay. So Lyft, connecting to, let's just say, Oracle through Merge, for $65, seems to be like it's priced at roughly one, one thousandth of the potential value for them. Why is it so inexpensive? it. Yeah, I know. So we probably need to adjust the pricing a little bit, but the thing is that we at first, when we launched pricing, it was based on volume, and it created a lot of complexity of the sales cycle. And also, what we noticed was a majority of the links that counts were

smaller. And so we were accounting for like the few instances where it would be really huge and allowing that to sway what the pricing looked like versus just looking what most of the pricing looked like, which is why we just simplified it. So it's true. It is the deal for a lot of companies that are mostly selling to enterprise. And a lot of this enterprise companies will also make money off of the integrations as well because you can. And for S&Bs, yeah, like the pricing is a little bit more expensive. We have to work with them on pricing too. But I love leaving value on the table, providing a great experience for customers. I'm actually really in favor of that. I think it's a great way to go about it. But how do you decide as a business when you might be

leaving even more on the table than you might want to? Yeah. I mean, that's always the question, right? And we do have a platform fee as well for the professional enterprise plans. And so if you are company selling to Enterprise, you can't use the self-serve plan. It's just there's not enough security features, like, you need more support, especially if you're selling to enterprise and you don't know the nuances of Workday or S-A-P, you really need more support. And so you're going to be on the enterprise plan, which has a platform fee that varies just based on the company size. And then that also helps lean out the cost, yeah. One more question about customers. There's been a trend that I've been tracking mostly through

public markets of software companies reporting pretty poor net retention rates over the last eight quarters, things have come down from 130% to 105, 160 to 120, etc. Has that compression of ability to upsell customers impacted merge's growth rate? So not that. It actually impacts us more when our customers aren't doing super, super well, and especially if you're selling mostly to S&B, then if they're not doing super well, because we're pricing based on the number of connections. So if they have fewer customers, then we can't charge them as much. Also, a lot of our customers would acquire each other, especially this year and last year. But the move-up market has helped us a lot with that.

Like the NRR for our enterprise segment is just completely different from our S&B segment. So we're just pretty lucky that we ended up making that strategic move earlier on. Okay. Now, if you're not going to raise more capital and you're growing quite a lot, eventually you're going to reach scale. So I'm thinking merge IPO Q2, 2027? Oh, my God. I wish.

Why not? I wish. I don't know. Seriously, if you guys end up some sort of horrible Frankenstein super late stage series G company, I'm going to scream. I know. It feels like you have tailwinds, you have good growth, you have good product market fit, your team seems good. Like, if you can't go public eventually, then literally startups are so broken, we need to just rip all the papers up and throw them and start over again. That's my vibe. I mean, that's the dream.

And I firmly believe that all tech companies that go public, and also just all companies that go public eventually become M&A shops where you have to strategically purchase other companies. and just like, you know, combine them in a way that is beneficial to the customer. But there are a few bets that we want to make that would be beneficial for us to do as a private company. And I don't think public, I don't think, yeah, I don't think the stock market would, like, treat it super well. So I'd want to make sure that we make this bets before we end up doing anything like that. And we're still 100 people. So we're pretty far from that right now. But no, there's actually, there's no minimum personal requirement to go public.

Oh, really? No, you can go public with two people. I mean, SPACs are essentially just zombie cash that's been floated, right? So, I mean, the question is, you know, do you have like 150 million trailing revenue? No. But will you by 2027? I don't know. I've never seen a company say, we grew 30X last year.

Here's our series B. Ha ha. Like, that's, that is a very impressive result. And I'm hoping it's kept going because, you know, I would love you to see you drop new numbers soon. I know you love numbers. I've been listening to you for a while. Well, let's explain what's actually going on. Okay, last question that I promise to let you go.

We were talking about AI earlier and tailwinds there off and so forth. Has anyone tried to buy you guys lately? No. Really? I'm shocked because I feel like some company in the AI data space like Databricks would be hovering. Yeah, no. And to be honest, I don't think we'd be super down for it either. I think the opportunity for what we're building is just too big.

And it would really be a shame. I really, I really, I really, feel like what we're building is like the infrastructure for the next generation of the internet, because just data is so important. And I don't think it would be a good, I don't think it's worth it, especially where we are right now. If you're going to become the data interchange for the internet, if you will, especially on the business side of things, do you eventually drop the API focus and maybe get more agnostic about types of data in motion? Oh my God. Yeah, Alex. That's what we're doing. Okay. What does that look like? I can't share all of it. But, you know,

Yeah, I mean, that's why we hate when people call us Merge API because it's really just merge. Because wherever your customer's data is, like, we want to be able to pull it and export it in whatever format that you want eventually. But yeah, for instance, like now we have CSV upload. We've built some on-prem integrations too. We have SFTP. So World should be our customer's oyster. Well, I should have started with that question, but Shindzy, we're out of time. I have to let you go.

But thank you so much for coming on. I really do appreciate it. And the next time, Merge does anything. You just call me up and we'll have you back on. All right. Sounds good. Thank you so much for having me. My pleasure. All right, everybody, live news coming up. Lots more twist to come. Make sure you're subscribed. My name is Alex. We're out of here. Bye.