
The a16z AI Apps team outlines how they are thinking about the AI application cycle and why they believe it represents the largest and fastest product shift in software to date. The conversation places AI in the context of prior platform waves, from PCs to cloud to mobile, and examines where adoption is already translating into real enterprise usage and revenue. They walk through three core investment themes: existing software categories becoming AI-native, new categories where software directly replaces labor, and applications built around proprietary data and closed-loop workflows. Using portfolio examples, the discussion shows how these models play out in practice and why defensibility, workflow ownership, and data moats matter more than novelty as AI applications scale.
Loading summary
Podcast Host
A lot of people think the AI story is about models. This episode argues the real story is about apps, distribution and moats. In this episode we share an AI apps overview featuring a16z General Partners Alex Rampel, David Haber and Anisha Charya, along with Jen Kof, head of Investor Relations at A16Z. They break down why the product cycles drive growth, why the AI era is accelerating faster than prior platform shifts, and what it takes to build enduring companies and AI applications. The conversation covers three core themes. Traditional software going AI native platform expanding beyond SaaS to take on labor and walled garden businesses built on proprietary data and compounding advantage.
Alex Rampel
I'm Alex Rampel on the Apps Fund. I've been at the firm for 10 years and I stole this from Chris Dixon, who published a post like this about probably 12 or 13 years ago. And the whole premise is that product cycles drive growth. And the top of the chart here is the NASDAQ from 1977 to present. It goes up sometimes, it goes down sometimes over the long run it has gone up, but there have been some very scary down points. So there really there have been four major product cycles. There was the PC and obviously before the PC there was the semiconductor. But we got to start somewhere. We'll start with the PC. There's always a infrastructure layer of companies that are building the backend. There's the application layer of people that are building things that actually are used. So Lotus was one of the first infrastructure, sorry, application companies. Adobe, Symantec, all of these companies that kind of grew out of the 1980s. But the infra players, if you were Apple and Microsoft, then you had the Internet that was enormous. Lots of bubbles along the way, but some very, very enduring infrastruct companies like Cisco and Akamai, enduring companies in the application space like ebay and Amazon that were built on top of that. Then you had cloud, so AWS accounts for the vast majority of market cap of Amazon. You've got Workday, Shopify, Viva, others that were the application layer. Mobile took all of these things that came before and now put a supercomputer in everybody's pocket. So the vast majority of humans on planet Earth have a smartphone, which is pretty amazing. That was the mobile era, which is still actually kind of playing out. Like I just bought an Android phone to test things with. It was $40 and this was more powerful than the ENIAC in 1946 or whenever the ENIAC came out. And then two years ago was this AI era is coming out as well. And the Nasdaq is higher, we know that. But the AI era really is playing out. And the cool thing is this is not a net new thing. This is building on everything before. Like if we didn't have smartphones and we didn't have cloud, but we just had the eniac, AI would be pretty cool. Like you could go check it out in a museum. But the fact is you now have 8 billion humans on planet Earth, the vast majority of whom have smartphones. And the adoption of this new technology is taking off like never before. So the AI era is here. The vast majority of net new revenue that's happening in software land is actually coming from AI, both at the application layer and the infrastructure layer.
Jen Kof
It's hard to actually think back two years ago at that point in time. Of course ChatGPT3 had launched, I think ChatGPT4 had also launched, but it was all just text and imaging and some basic reasoning. But none of the native audio stuff. Obviously real time interaction, none of that actually had happened yet. It's hard to even imagine how far we've come, even in just the two year timeframe as a part of up.
Alex Rampel
Yeah, I mean it's really remarkable like what these things have done. I mean, one of the ways of joking about this is that we had this idea of artificial general intelligence or the Turing Test, like when can we tell the difference between a computer and a human if we don't know who our interlocutor is? And the answer is if you were to take a person 10 years ago and show them, or 20 years ago, or 30, like, oh my God, this is like a fully sentient, this is smarter than, than any kind of human out there. We kind of keep changing the goalpost a little bit on what exactly is AGI. But yes, the pace of innovation here is just remarkable. And the important thing is just the opportunity set that it unlocks. So whenever you have a bull market and very, very exciting tech, there's always somebody saying it's a bubble or it doesn't work or it's all overhyped. And I think there was some MIT paper that came out. This is not a faulty mit, this is somebody who published the paper. It's like, oh, you know, most enterprise deployments really, really aren't working. In terms of AI, we're seeing the exact opposite. Two things. So there's a company called Ramp, and they are kind of credit card expense management products. And you see this giant tick up in January of 2025, which is, you know, when did enterprises. And these are much more like who uses Ramp? This is not necessarily a startup, but it's a more forward thinking company. It's not necessarily ge, it's a company with thousands of employees, maybe in the Bay Area or New York, that wants to be more tech forward. And they've just realized like, wow, this stuff. Jen, to your point, like G 3.5, pretty good. 4. I was like, wow, it's pretty amazing. I can write a new episode of Seinfeld with it. Like amazing things that I could do almost to kind of wow my friends. Like a magic trick. But now the magic trick has actually gone into the enterprise and is saving people time and money. And one of the themes that you'll potentially get out of this presentation for me is that I have this prevailing view of human behavior, which is everybody wants two things. They want to be richer and lazier, so they want to do less work and get more economic value. And this is really what Gen AI unlocks. And it's really starting to happen right now. And this has been a little bit of a flat curve, but it has been inflecting a lot. And you see this in the expense Yeti, you see it in the growth of all of the companies, both at the infrastructure layer and at the app layer. And again, whether they're overvalued or undervalued is almost not the point. It's hard to time the market on these things. The amount of value that they are generating is just tremendous. And we're going to get into this in a second. If anybody knows Maslow's hierarchy of needs, this is like this philosophical tome of what is it that humans need? At the base of the pyramid, people would joke, is WI fi. So it's like, okay, I need all these things that have been true for hundreds of years. And at the very, very top of that pyramid is the self actualization concept. But what I really, really need, if you talk to any teenager, it's like, you know, where's my WI fi? Where's my WI fi? And what's starting to happen now next is it's actually AI. So obviously you can't have AI without the Wi Fi. But something like 15% of adults on planet Earth now use ChatGPT every single week. Why are they using it? It's just part of their daily routine. Whether it's settling a bet with their friends over like, you know, how does this work or that work or I want directions to this thing or I'm really puzzled. My wife just used it to complain to the school because our kid miss bus and the bus driver said he can't open the door because it's against the law to open the door. This is the true story. So my wife had ChatGPT scan all the laws in California and the US federal system writ large. Even though our government is closed down. No, that was completely made up. Send a very, very polite note. I'm sure that the school is going to start adopting ChatGPT2 to start responding to people like my wife apologizing on behalf of the bus driver. But they did send an apology. Sorry, we made that up. Next time we can open the door for your child if he is on time when the bus has already closed the door. It's like a countably infinite number of use cases for these things and the growth of minutes per user in the US I mean this is just astronomical. And as these things work better and as they unlock more use cases, it's kind of obvious that the growth in minutes will go up. This is happening at a breakneck speed. So the key paper which was co written by this very, very smart guy, NOAM SHAZIR In 2017, attention is all you need. It introduced the transformer model. I remember we have a partner here, Frank Chen, who's been here for a very, very long time and he demoed ChatGPT or G& it didn't really work that well. It reminded me of this thing called Eliza, which was like a famous Markov chain based thing. It was basically a therapist that came out. It was an AI based therapist in the 1960s or 1970s. It's still around. You could try it. And basically you say like, doctor, I'm not feeling well. And then it just kind of says, and why is it, Jen, that you aren't feeling well? It just basically takes the words that you say, turns it into a question. It feels kind of sentient until you ask it like, hey, I want to complain to the school about the bus driving. And then it says, and why do you want to complain to the school about the bus driving? It doesn't actually give you an answer or anything that you need, you know, OpenAI. It's hard to imagine that this just happened a couple years ago, but from 2023 until now, like we really have entered the golden age of apps. And I base that purely numerically. I'm used to companies that will grow from, I don't know, like we used to talk about, like double, double, triple or triple, triple, double or all these different ways of measuring revenue growth. Because normally if you're selling a software product and let's just say that you're selling a software product to an enterprise and it's a hundred thousand dollars a year. You might sell a couple one year, a couple the next year, a couple the next year. But very, very rarely have we ever seen a software company go from zero to a hundred million dollars in revenue in a year or two. And we are seeing this right now. This is not like, oh, we're seeing it because people have too much money and they're buying these things. These are companies that are buying these things because it unlocks so much value for them. They want to be lazier, they want to be richer, and this is unlocking that. So I'm going to talk about three broader themes that we're seeing in AI applications really more broadly. These are the types of companies that we're investing in. And partially this is when we ask ourselves, what is defensible? What is it that the labs aren't going to do? Because this is a very, very good question. It's not like OpenAI just wants to be this backend layer for everything. They have a leading consumer app, like they just launched, arguably a competitor to TikTok. Microsoft is getting into the space in a meaningful way. And if you look at the history of software, I mean, this firm was started by Marc Andreessen. He started a company called Netscape. Netscape became roadkill due to this company called Microsoft that went into an antitrust case because of making Netscape roadkill and whatnot. But how do you build an enduring company? And what are the areas that potentially have the most enduring growth? And there are three that I'm going to lay out. So the first is basically traditional software is going AI native. And this is no different than like if you built a time machine right now, go back 15, 20 years and say, I'm just going to invest in every single cloud native company that pops up. You would have an incredible portfolio. You'd have Shopify, you'd have Viva, you'd have netsuite. When netsuite's a little bit older, you'd have Salesforce when it first went public, because it turned out that the incumbents couldn't really respond to that because they, they were selling on premise software or shrink wrap software for a lot of money upfront. And they didn't really know how to go for like less money every single month as a subscription. So category one is trad software that's going AI native. Category two is arguably the biggest, which is basically it's not competing with the software market at all. This is if any of you saw my talk that I gave in May. Software is starting to eat labor. You're basically selling software that does the job of what people would do before. This is arguably a much, much bigger market. The laws of business still apply. You have to build real mot. You can't just build something that's a little widget that somebody underprices your -by a dollar tomorrow. We're going to talk about that in a second. And then lastly, I call this the walled garden. But basically, really, really interesting proprietary data models where the value of this business, because you're able to deliver the finished product thanks to AI becomes much more valuable. And I'll talk about number one. So existing categories are going AI native. So this is a little. We actually have a post coming out about this in a couple days. But I'm sure everybody here has heard of bingo or played bingo. I'm from Florida. There's lots of bingo in Florida, lots of different names on this list. And one of the key lessons that I had as an investor is, you know, and Mercury is kind of a great example of the tortoise that beat and is, you know, still beating the hare. Mercury built a Neobank for startups. So they said we're going to be the better source for you when you start your company to go deposit your money with us. We're going to, you know, help you pay your bills, track your expenses, be a basic accounting system. Mercury never stole an existing customer from Silicon Valley bank in until the weekend that Silicon Valley bank failed. And it is what I would call the canonical Greenfield opportunity versus Brownfield opportunity. So Brownfield is you're selling to an existing market. So let's just take an example here. Email marketing. You use Mailchimp. I want to go sell you a competitor to Mailchimp because it has AI. That's going to be really hard. Or you use NetSuite and I'm going to say like, hey, ditch your NetSuite. I'm going to give you a NetSuite. That's going to be really hard if you're a net new company. And this is what I mean by greenfield, you have no existing product, you're not using anything, you're a brand new company or sometimes you hit an inflection point. So the inflection point, I'll pick on NetSuite here for a second. The inflection point is I have 50 employees now. I have three entities and two currencies. I've been using QuickBooks, my entire life. QuickBooks can't handle, for whatever reason, they cannot handle multi entity, multi currency support very well. KPMG says, hey, you gotta go move to a better ERP system that supports that. And now I have an opportunity to pick the better product in the market. And NetSuite is a product in the market. Or I can try this thing called Rillet, which, which is one of our companies, which is basically like Netsuite, but it closes the books for you. It has 50 AI features built in. That is a Greenfield example. Now these things don't grow like weeds because you have to wait for the new company creation. You're going entirely for Greenfield and not for Brownfield, but every single one of these spots on this bingo board, the incumbents are all adopting AI and they're going to make their businesses much, much better with AI. Like Bill.com is going to be a stronger business, or SAP is going to be a stronger business, or Adobe is going to be a stronger business because of AI. They're just going to be able to charge for new things. Work will start charging. And I mentioned this in my presentation that I gave a couple months ago. Workday will say, hey, do you, do you want us to do reference checks on every new employee that you enter into our system? That's 500 per reference check. Why can't somebody do it for 499? Because you're stuck with workday. And there's a saying that I use a lot, which is the best companies have hostages, not customers. And I'll talk about a couple examples here. So RPA, there's an existing company called UiPath, public company customer support. There's an existing company called Zendesk, it's now a private company. ERP, SAP, NetSuite, or in some cases like Zendesk charges per seat per month. That is almost an extinct business model for support software because, well, wait a minute, I don't want to pay per seat per month when 99% of all queries can be answered by the support software. I want to pay per outcome. So we've been aggressively betting on the Bingo board. Let's evaluate every company that we see in this space. So if it's payroll, if it's support, if it's erp. And the important thing is that these are systems of record. So this is the best companies take hostages, not customers. Like, we don't want to invest in hostage companies. We don't want to invest in companies that have negative 100 NPS. We want to invest in companies that still have A very, very strong moat. And that's what I mean when I use that expression. So all of the companies that we're looking at here, what is a system of record, it just means like it runs the entire business. Everything on that bingo board, like how do you get rid of NetSuite? It's basically impossible. You can enter in with an AI wedge or more often than not, a lot of these bingo categories are. We're just building the new system of record. The existing incumbent is doing that as well. But it still is a no brainer whenever you're brand new in the market or at this inflection point of do I use this old one or do I need this new one? So next here. So the second theme here, which I am personally most excited about is where new categories are emerging, where labor is software. And there is no bingo board for this at all. And the reason why is because there weren't software companies that did this before. And the predominant theme is that you have a lot of things where you would hire a person. You can't hire that person or that person that you were going to hire doesn't speak 21 different foreign languages and won't work 24 hours a day. But software can do 90% of what that human would do. Now you will pay for software, not necessarily at the same like rate that you would pay for labor. But this is not something that you would hire a software product for. This is not something that you would ever have a software product for before. So I'll talk about a couple examples here. And obviously, you know, I mentioned this, I can mention this ad nauseam, but the labor market is astronomically bigger than the software market. So next. So again, this is kind of the governing principle. Here you go, look at a job. Front desk receptionist. Plaza Lane Optometry. Plaza Lane Optometry has like they have a bingo board as well in terms of software that they spend money on. They probably spend money on Microsoft Office. They probably spend money on Squarespace or Wix. That's on the order of $500 a year. If you can deliver them a software product that does, you know, call it five out of the eight things on this job posting. They will hire that software product. What do they pay for that software product? This is the part of the market that is almost unknown because they're probably, they're almost definitely not going to pay the $47,000 a year that they're advertising for this job or whatever the rate is that they're paying for the job. They're probably not going to pay $500 for software. But the promoter, the creator, developer of this software product, an application software company might say, you know, we're going to charge you $20,000 a year. They need to be careful about how they do this. We often want to see them turn into a system of records so that if they are doing, you know, five of these eight job responsibilities, somebody doesn't pop up and say, we're going to charge $19,999 a year. We want to make sure that this is a very, very sticky end solution for plasma lane optometry. You're going to see, I believe, a lot of market cap creation on the bingo board of existing software products that have a new, better alternative that are going after Greenfield. But here you can go after Brownfield, you can go after existing companies. You could probably charge a lot more. There was a path to much, much more explosive revenue growth.
David Shute
But just maybe to take a step back, you've probably heard a ton about what's happening in legal AI. You know, just given how document intensive the industry is, there's tons of applications for LLMs in the space. Most of what you probably heard are around companies like Harvey, you know, serving the defense. And the corporate side, maybe less familiar to you, might be the plaintiff side, which is really about representing the individuals in areas like employment law or personal injury. And, you know, we spend a bunch of time looking at, you know, the different companies on the plaintiff side, in part because one of the unique characteristics about that side of the market is that these attorneys operate on a contingency basis, meaning they only get paid if they win. And so they're incredibly aligned with their clients. They don't bill by the hour. You know, they take a percentage of the, of the actual case outcome. And so as a result, you know, for every 100 leads that a plaintiff attorney gets, they often take one case. Because anytime you take a case, it's an investment in your time and your labor. So just incredible alignment with AI's impact on their core business model. Right. To contrast that, if you're a corporate attorney, you know, and you're, you know, you know, junior attorney is 50 times more productive, you just eroded some of the revenue that you can actually charge to your end client. Again, in this case, if you can make your attorneys, you know, 5x more productive, you can potentially increase, you know, your revenue by 5x or more. And so the EVE guys had a particular, particularly interesting kind of point of view from a product perspective. They really wanted to own the end to end workflow from intake all the way to outcomes. And so, you know, to Alex's point earlier around voice, you know, they recently launched a voice agent which is actually collecting evidence, you know, from their prospective clients. And it's sifting through you know, mountains of, you know, medical records or employment documents and helping these attorneys, you know, figure out which cases to take. Because it is generating sort of this data set of, you know, case characteristics such that it can say, hey, this case is potentially worth 50k. You know, this case is worth $5 million. You should probably spend time on this case over here. And then it'll just help step through all the different phases of pre litigation and litigation for these attorneys. So it'll draft a medical chronology, It'll draft a kind of core artifact of these cases, which is known as a demand letter. It'll file complaints. And ultimately, I think what's so interesting about this business and it speaks to, I think, why moats matter. You know, one is these attorneys are living in this product all day long. Like, one of the core, you know, pieces of feedback that we heard when we were diligencing the business was that literally 100% of the cases were flowing through the product. But interestingly, as EVE begins to generate data on outcomes, that data isn't public, right? That's not something that the large, you know, labs can train models against. And that data is actually informing better intake. Right. So that they can then go back and say at intake, hey, given the characteristics that we've seen in all the cases that we prosecuted across all the, you know, the platform, you know, these have these three variables that make this case potentially worth a lot more money. Or to Alex's point, it can, you know, reduce the cost of taking on a case. You know, before an attorney was only taking a case that, you know, at minimum could potentially make them 50k. And suddenly they can afford to take cases at 5k. The market expands, right? And there's a big sort of supply and demand imbalance, you know, today on the plaintiff side that. That Eve is unlocking. And as a result, it's just the market pull for this product has been candidly stronger than we even anticipated. My hope is that it has a lot of characteristics that we'll be continuously investing in where AI is just incredibly aligned with the business, both driving revenue and saving these folks money.
Alex Rampel
Thanks, David. And the reason I wanted to talk about that is I think it's really cool as Eve, but it's a metaphor for the types of businesses that we find compelling and why, you know, 0 to 30, certainly, or 2 to 30 is not normal, but it actually is normal if you're able to move very, very quickly and just deliver again, this promise of I'm going to make you lazier and richer. So let's go to the next slide.
Jen Kof
Actually, before we go to Salient, Alex, why don't we just take some of these questions here because they're relevant in the context of EVE as an example. And then also before we switch to Salient, exemplify why we find these to be particularly compelling. So there's a good question here from Brian. A lot of consumption based AI apps have found it hard to become mission critical. They're easy to switch on or off as a part of the broader suite. How do you evaluate that in diligence? Maybe, David, if you want to use EVE as an example or others that we have in the portfolio, how do you evaluate that in diligence? And what patterns have you seen around in which apps actually graduate to being essential?
David Shute
Yeah, One of the distinctions that I often draw is this notion of differentiation versus defensibility. And I think AI is an incredible tool often for differentiation. Right. So the idea that the voice agent can speak to folks in 50 languages and gather that evidence, you know, highly differentiated versus the human, right, Obviously delivering value. But that capability alone, in my opinion, is not a source of their defensibility. Right. The source of defensibility for EVE is in owning the end to end workflow. Right. It is actually in building a product that is contextual to, you know, all the work that that attorney has to do. And then I think, you know, not unique to eve, but one of the kind of X factors is that the data that that business is generating, which Alex will get into a bit in this sort of walled garden, it has a bit of these characteristics of this sort of walled garden is not public. And it sort of creates a source of compounding competitive advantage, you know, for the product itself.
Alex Rampel
Right.
David Shute
So the more cases that EVE can prosecute for all their different clients, you know, the smarter that the product becomes. And it actually, you know, kind of reinforces that loop. It becomes sort of, you know, you're showing, you're showing up to a knife fight with a gun, right. And so soon it's going to become an essential tool for any plaintiff attorney to operate with and that just becomes very difficult to displace. Right. So it's not so much the AI ness, right, in the voice or the ability to summarize documents, it's actually in becoming kind of the system record this.
Alex Rampel
Ended workflow for sure.
Jen Kof
And in fact, actually there's multiple threats to pull on it. But maybe I'll ask this question first relatedly around talking about the potential upside of market size of these companies around labor versus vertical software bucket. And how do companies in this category build defensible moats and particularly earn attractive margins as AI proliferates and costs continue to scale down?
Alex Rampel
Yeah, well, why don't we come back to that one at the end? Because I think hopefully what you'll get from it's not like we're just investing in companies that do labor and then the end, their moats matter. In fact, more than ever. Because the one thing that's happened in software is once upon a time there was a company called WordPerfect. And WordPerfect kind of like kept growing for a very, very long time or once upon a time there was a company called VisiCalc. And then whoever had the most distribution said I should do that copies it. And obviously WordPerfect is toast. VisiCalc is toast. Lotus 1, 2, 3, which was the one that beat VisiCalc, that became toast. But it would normally take five years for the bread to become toast. And there is a very, very high level of prolific speed. I mean, now Anish, David and I and Jen can go build a software product. We can vibe code, if you've heard that term, we build software very, very quickly. What makes it actually increases the peril for anybody who's built a software product that has an enormous margin pool. You know, your margin is my opportunity. Well, I can vibe code. I can vibe code against your opportunity. It has to be very, very sticky. It has to have some unique competitive advantage. And data is often one of those. So if I work with every plaintiff, law firm or you know, actually, why don't we go to the next slide here and I'll just talk about Salient a little bit. Sorry. So Salient is in the Eve mold. And I know we also had a question about like, what is the societal impact of everybody losing their job? Like, I don't think that's actually going to happen very quickly. You know, 98% of Americans were farmers in 1789 and obviously the tractor made some of them unemployed and made them do other things. But most of what we're seeing candidly is not about eliminating work. I mean, I do think that the 3.5 million people that drive trucks at some point in time, like we have a better solution than the truck driving human. You have, you know, AI doing that. But most of these things, they're really. It's like, you have cost here, you have value here. You would never hire a human where they are producing less value than their cost. It just does not make sense. But if you can now hire AI effectively, you can hire AI where the amount of value that like the cost has just gone down, the value has stayed the same. You're going to hire a lot of AI, you're not going to get rid of a lot of humans. And if anything, we never know. This is so hard to predict, but what will humans do? I mean, like, there was no job of like product manager 75 years ago at a software company or designer. Like all of these jobs that exist today, they wouldn't have made any sense to somebody in 1800. So it's hard to kind of pontificate on that. But a lot of the things that we're seeing, they're not displacing people per se. I mean, I know it sounds pithy to say software is eating labor, but really software is augmenting labor. Or it's like all of these people that I can't hire, whether there's a job shortage or a skills shortage or whatever, I can now deploy people that will answer a phone. Like, I would just never hire somebody to go answer the phone for me at 2am I would hire somebody at 4pm, but not at 2am it's just the value to cost equation is inverted. And kind of a great example of this is like the salient. Yes, they are going to people that collect. It's called auto loan servicing. So you go to an auto lender, they have to go make sure that they're collecting on their bills. Or if the person's in a car accident and the insurance carrier, is it, you know, supposed to pay you? How do I make sure that that insurance carrier is paying me on time and writing the check to the right person in this case because I have the lease, like, they need to write it to me and not the actual, you know, not. Not the person in their actual name. How do I do all of that kind of stuff? I would hire lots of people. I would train lots of people. A lot of these people hate their jobs because it turns out people yell at them all day and say, I'm not paying you back for this car. Or the insurance carrier keeps you on hold for four hours and that whole music is just terrible and you're going to want to kill yourself if you have to listen to that 12 hours. Like all of these reasons why humans don't want to do this or you can't hire humans for this. The key thing with Salient is not that they're saving you money. The key thing with Salient is that they collect 50% more. Like, this is the key thing. Because Ari, the CEO, he kept pitching like, I'm going to save you money. I'm going to save you money. I'm going to save you money. Like, people like saving money. But if you go to somebody and say, I will collect 50% more revenue for you every single month and I will make sure that you don't go to jail. Because none of these people that you hire that aren't very well trained, that have to listen to this horrible hold music for four hours a day, they don't say something that they're not supposed to say. I can make sure that AI doesn't do any of these things. Like, that's why that company is growing so explosively. It really is. It's much more about the value generation. I mean, yes, the cost is much lower. And this is one of the questions around, like, how do they figure out how to. How to charge for the product they went to their first client, had a $50 million a year call center with, I think, a 40 to 70% annualized churn rate per employee. So it's just. And not because they're firing people. It's just like, nobody wants this job. So they now say, I will do it for you with software. I will give you a system of record. I will make sure that we're scraping every single new federal and state statute. Because what you say in Missouri is very, very different than what you have to say in California is very different than what you say in Iowa. We're going to do all of these things. No human can keep that in their head. At the same time. Time. It's like, all right, I'm talking to David Shute. What do I say? He's from Santa. You know, he's somewhere in California. Oh, wait, but actually he's traveling to, like, Kansas. I don't know what to say. Like, Salient knows exactly what to say and it knows how to say it in 21 languages. And that's why the collections rate is 50% higher. So, like, this whole category of, like, we are going to make you more money and it's going to cost you less. Like, it's just. It's a very, very hard thing to move away from. The key question for us, which I think is a very good question, is how do we make sure that we're back in the right want. And how do we make sure that Salient is not. I mean this was my number one question when Ari came in. I was like, well, how are. Imagine that there's a company called Talliant and a company called Zalient. Why is it that Salient is going to be Tallient and Zalient and Ari actually had. Ari, the CEO had a very, very good answer to this. Not to like, you know, he looked up on ChatGPT. How do I answer this difficult question from a VC? But again, moats matter. We know exactly what script to say. This is an example of kind of a data moat. It's like, because we've done millions of phone calls, we know exactly what to say. We have lower latency on every single like statute that comes out from like we. They actually have like a very, very good product that ingests every single law. Like as it is even proposed as a statute in all 50 states. Sometimes it's at the county level. Like they're doing all of these things that make it so much harder to compete so that they will not lose a deal. You know, moats matter more than ever because you're able to create software so much more readily.
Jen Kof
Actually, maybe this is a good dovetail to this section, which is does this then mean software becomes way, way, way more specific in certain categories and it doesn't need to win a bunch of different categories to become a huge business? And I think that might actually be a good dovetail to this theme that you want to cover here.
Alex Rampel
Yeah, I mean this is the thing that we don't know. I mean, like, we obviously have many examples of vertical software companies that have become very big. So ServiceTitan is a vertical software company. Mindbody is a vertical software company. Toast, that's a very large vertical software company. Toast is designed for restaurateurs to run their business, to integrate with doordash, to pay their waitstaff, to do line. Like everything around operating a business, it's a vertical operating system. It's very, very hard to displace. One of those people would have doubted how big that could become. And actually a lot of people did. Like, it was very hard for Toast to raise their B round because people would say, well, I look at the restaurant space and like, you know, half these restaurants go out of business every year. I look at how much software they buy. Well, they don't buy any software. So therefore this is a bad company. I'm not going to invest in it. And you know, fast forward 10 years but the reason why that happened was it turned out the business was much bigger in this case because they added financial services. And the financial services, we're going to do lending to restaurants, we're going to do payment processing for restaurants. And we make it very, very sticky because it's an entire software platform and there's no way for First Data or Global Payments or any of these companies that traditionally do software to go append. Sorry. That traditionally do payment processing to append some kind of software solution. So that's why Toast, you know, people got Toast wrong. It's a very valuable company and a public company today. I think the same thing applies for. I'm adding in labor. Like, it's not just I do labor and then somebody does labor for a penny cheaper. I need to build some kind of system of record for you, some kind of vertical operating system for you so that you can't just go switch out for the cheaper player. And maybe this is a good way to kind of go into theme three here, which I'm very excited about. And I call this the walled garden. And this is really important today because if you look at, like, take a metaphor here, where this amazing company called OpenAI shows up and they're like, hey, we're a vegetable farm and we're farming tokens and we're going to sell tokens and we're going to charge for tokens to all these people out there that are building applications. So it plays out exactly as I talked about. Like, OpenAI is an infrastructure company. We invest in all these application companies. But then OpenAI is like, you know what, we should put some restaurants on our farm. These. A lot of people come to our farm. Let's just have restaurants here. And then all these restaurateurs are like, wait a minute, like, you're selling me vegetables now you're competing with me? Like, that's not good. The reason why I bring this up as an example is because it actually is happening. And it's a blueprint for how to potentially deal with a world where the source of the raw material is actually what is rare. So let's go to the next slide and I'll show you. Like, I'll make this a little bit quicker, but as I mentioned, this is kind of like the world's second oldest profession. There are lots of cases where I kind of construct some physical property, I build a wall around it, and I charge you for access to my property. You can do this in the data world as well. And I'll pick an example on this little bingo board here of FlightAware. I'm not sure how many people have heard of FlightAware. How do they get their data? And their data, by the way, what is their data? There's nothing proprietary about it. It's all public. You can buy an antenna on Amazon to receive. It's called ADS B Transponder Data. So every single airplane after that Malaysian plane went missing has a little transponder on it that shows its height, its speed, all of these different attributes on it beams it down to planet Earth. Antennas can pick this up and figure out, you know, this tail number is at this place. I can buy one. It's free. FlightAware, I think they have something like 100 antennas around the world. They pick up all this information and they can chart that that's a piece of data. Like, I can ask ChatGPT that they don't know that, that only FlightAware knows that. Or PitchBook does this for funding rounds. Like, who knew what the Series B price of a company in 1992 was? Like, PitchBook somehow has that. Or LexisNexis knows this. CoStar knows this for real estate data. Bloomberg knows this for all sorts of exotic financial stuff. Like, in many cases, it's all free. Ancestry.com built their entire data mode by buying genealogical records from the Mormon Church. All of this stuff is not available on ChatGPT. It's not available on Anthropic. Of course, they can license it. But the reason why I mention this is, what do you do with FlightAware data? Or what do you do with Bloomberg data? Or what do you like? I'll tell you what I do with PitchBook data. I hire an analyst. And I say, analyst, go write me a memo about this company called Eve and compare it to every other company in the legal space that had ever done something before. And PitchBook just sells us a subscription for here's every single Series B of legal tech company since 1992, okay, that's valuable. They can charge $20 or $200 or whatever they charge per month for that. What would be more valuable is saying, because they're the only ones that actually have that piece of information, they should probably charge $2,000 for that. Which might mean. I mean, maybe this makes you nervous. We might need one less analyst, because now we have a finished product. Because what we don't want is we don't just want a subscription to PitchBook Data. We actually want to do something with it. We want to somehow take that vegetable, if you follow my metaphor. And turn it into a finished meal. You know, one of my favorite examples here is Domain Tools. Domain Tools does they have one thing which is very interesting. They run a whois query which says, who owns a particular domain name? This company has been around for a very, very long time. If I want to figure out who owned a domain in 1998, there's one place to go and that's Domain Tools. So like this model has been around for a very, very long time. Before AI Very, very large companies exist in this space. When you add AI, it makes it tremendously more valuable. So I'll give you three examples that hopefully kind of hammers this point home. So there's a company called Open Evidence, which if you use it, apparently 2/3 of doctors in America use this thing pretty much every week. Open Evidence is exactly like ChatGPT. The interface looks exactly like ChatGPT, except, you know, who has exclusive license to the New England Journal of Medicine and every other medical journal out there. Open Evidence. So if I tore my Achilles, if I want to read about what I should do, all of the evidence based care out there, I can go to ChatGPT. It's moderately useful. There's no reason not to do that. Open Evidence is so much better because they're the only ones that actually have, they've built. In this case, they found all the data, they found all the unique vegetables out there. They convinced the vegetable seller not to sell it to any other restaurant. And they have a restaurant that delivers the whole thing where there's a 26 year old company called Velix. Incredible company that just got bought. The CEO was telling me the origin story of this company. He's from Spain. He bought up every single legal record in Spain. And why would you want to buy up legal records? Because I don't know. Wilson Cinni wants to know Spanish case law. In case Andreessen Horowitz goes, invests in a company and needs to figure something out. So Valex would aggregate and digitize this information, sell it to law firms and other people that need legal information. Pretty high gross margin, but very, very low scale and predominantly European in sp. Then they were like, you know what, we should add AI to this. And apparently it quintupled their revenue. Now why would it quintuple their revenue? I might love Harvey. I pay for Harvey. Amazing product. But if I want to have a finished memo for my client at 7am, I can't get a paralegal to go do this. And I know that it needs to incorporate some element of Spanish legal data like Velex is my only solution. And Instead of charging $2 a month or $2 an article or $200 a month or whatever they can charge for the raw material. And what Ask Leo does is it's a proc. So if I'm a company and every company, every employee at every company kind of hates their procurement department because on the one hand, the procurement department is supposed to save the company money by making sure that some rogue employee doesn't buy, you know, expensive widgets at an overpriced price from an unapproved vendor. But on the other hand, they introduce all sorts of complexity into the process. So imagine that I've got a contract from Deloitte to give me AI and you know, somehow revitalize my company, who has 50 other contracts from Deloitte where I can understand what I push back on. Like that that is actually very, very useful proprietary information. I wish I could go ask ChatGPT for this, but they don't have the world's treasure trove. Like what is the information they will never get? They're never going to get 50 old Deloitte contracts. Like, where would you find them? I guess you could do a FOIA request or something, but you're not going to find them. And Ask Leo has these. So it just makes the product so much better. And go back one slide here. It's hard to say where we're going to find these things, but the most compelling of the ones that we found are it's like all of the information is free, just like ads B, flight transponder data, that's free. But you find something that just like it wasn't worth that much before because like, what do you do with flight data? What do you do with who is record data on the Internet? I actually talked to an entrepreneur recently. He was like, oh yeah, you know, know what I like to figure out historical subscriber Data on of YouTubers. It's like YouTube doesn't publish like how many subscribers Mr. Beast had on August 4th, you know, 2017. Like where would you find that? There's some company that collates that collects that and that's just. They're just selling the data. It's not available anywhere else. And you know, these are some. We just published a post. I would encourage people to read it, like, you know, the Walled Gardener, we called it Fruits of the Walled Gardener. And all of these things like creative archives, logistics, like you go to like some county recorder's office and you can see who owns what property Record, but you have to go to the county recorder's office to find that. It's all free, but you can digitize that, make that available, and then add AI to that. And this sounds like, oh, just add AI. It's much more valuable. The reason why is because you're saying, I have something that nobody else has. There's a reason why people are buying this before because they're trying to create something that is of higher value at the end of the day. And you can now do this. So, you know, go to every museum. Actually, I just talked to an entrepreneur who found every old manual. This is a great example. Found every old manual for, like, blenders made in the 1980s, 1990s. Like, just, you can buy this stuff for pretty much nothing on ebay. Where would you find a manual for an old blender in 1999? I have no idea. But apparently ebay is where you find it. But it just shows like, these, these walled gardens that you can build with data. You could have built this before. You could build a ten or a hundred times more valuable company today.
Jen Kof
So, Alex, maybe can I pause you here in part because, you know, the last era of investing, you gave a great framework in the world, a great framework for thinking about the battle between startups and incumbents. And, you know, if startups could figure out distribution before incumbents can figure out innovation, that was, you know, their, their success. When, like, how, how take us through the dynamic of when you're thinking about which companies to invest into, where it's very clear that they can disrupt, you know, the incumbents in the category and where, you know, what are the examples where it probably doesn't make a lot of sense for someone to build a company like, and that has a proprietary walled garden that is going to be very difficult to unseat.
Alex Rampel
Yeah, I mean, I think there are two ways of thinking about this. Number one is in the case of the used blenders on ebay or the manuals, like, there just wasn't a company before charging for access to the subscription of, like, I'm going to sell you, you know, per data article that I've digitized by Google, where I'm going to charge you $20 a month, like, you know, probably not that interesting. But now if you have this finished product that you can charge a thousand dollars for versus, like, the raw material that you charge a dollar for, maybe now the business is tenable. So, so one category is you just find a new data source. And there's a reason why, like, you know, in venture Capital School, we learn to always ask why now? Like this is such a great idea. Why didn't this exist 10 years ago? Great answer for Uber. When it came out, there was no iPhone and no GPS transponder in every device. Once you have that, that now you can have Uber. The why now for some of these more esoteric things is it's, it's kind of like a little bit of a why now? Like, why Isn't this a 20 million business like Velix after struggling for 26 years, why is it now $100 million business? It's because you can deliver the finished product. And of course, like there are, I would argue, like a lot of the old things that were out there. Like Ancestry.com is a valuable company. They digitized LDS data and a lot of people want to figure out where they came from. And there's an NBC show that says, you know, what are roots? And people like watching that and all these kinds of things. You know, it's a valuable company. That would be one where it's like, I would be hard pressed to say, how do you make that dramatically better with AI Maybe it's like I want to say, hey, please, I'm about to die. I want to figure out which one of my errors to leave all of my money to. Please email them and set up dates with me so I can figure that out. And like that's the value add that you do with this proprietary data. This is why I'm an investor, not an entrepreneur anymore. I'm out of good ideas. But that would be something where, you know, know there is an existing data store. Maybe I licensed that like open evidence. They didn't create new medical journal entries. They were just like, hey, let's go distribute to doctors. We know that doctors are really interested in this stuff. We know that all of the information is in these old medical journals. And the back catalog is very, very, very useful. Like it turned out. Like I think of all the, the things that Michael Jackson did right and wrong, probably the, the most ripe from an economics perspective was buying the back catalog of the Beatles. Or like he bought a big chunk of that that ended up being worth a lot. Because until the copyright runs out, like Beatles catalog, a lot of people like listening to Beatles. That's going to become more valuable. So you can buy existing stuff that is already out there, that already has a business and that's like open evidence. Or you can try to create something net new, which is kind of more of the Ask Leo. So I don't know if that perfectly answers your question. But my view on everything that's happening in AI right now is it's one of these weird situations where, where it's very different than cloud, where most on prem software providers were like, cloud is stupid. Most potential customers were like, cloud is stupid. It's not safe. I don't trust it. I want to host things like, you'd have. Your entire IT staff is like, I don't trust that stuff. So the existing incumbents did not build cloud providers. Like, PeopleSoft did not say, let's go build PeopleSoft Cloud. They have it now. But that's where workday came from. They were like, we're going to build this. It took a while for the business to cat, for the, for the, for everything to catch up. I'm very, very bullish on incumbents. I, I, I hope I can say that because I don't think that, I think Netsuite is going to figure out 15 different ways to monetize with AI. I think that Quickbook Intuit has this gold mine on their hands where they're, they're just going to start charging per collections that they make to all of their existing hostages that use QuickBooks. But that still does not mean that you don't have these greenfield opportunities, you don't have these new data opportunities. Like, there's so many new opportunities that have popped up, up largely because of this value cost thing. It's like you have so many, like, it's this infinite number of things where it's like, I find something where everybody would want this at $5, but it is currently only sold for $10, therefore nobody wants it, therefore it's not a business. Wait a minute. AI allows me to sell it for $5. So it's really one of these rare situations where it's good for both. Whereas I think mobile, like Most people thought BlackBerry was great. IPhone was stupid. That's why the incumbents didn't, you know, that's why, why didn't booking.com build Airbnb? Why, why didn't, I don't know, taxicab company build Uber? It's just most people thought this was stupid. Everybody thinks that this is a good idea because of course, intelligence, like, you know, AGI in everybody's pocket is a very good idea. No, nobody can argue against that. It's more of the, the existing incumbents. This is why I'm just, I'm bearish on the brownfield opportunity on the bingo board. I'm very, very bearish on. I'm sorry, I'm Very bullish on the brownfield opportunity for the, for like walled gardens and for kind of software that does the job of labor, for sure.
Jen Kof
By the way, I thought you were going to say the smartest thing Michael Jackson did was let his family use his likeness for the Michael Jackson live show, which I, according to Ben, has now generated more revenue from that show than his entire existence as a performer. But.
Alex Rampel
But I give him more credit for like, I think apparently what happened was somebody was like, you know what, you know where the money is. It's like that, it's. It's like that movie the Graduate. It's like plastics, right? Somebody was like, like took Michael Jackson aside, you know, where the money is. Back catalogs. Good point. I have a lot of money. I'm going to. I'm going to go buy the Beatles back catalog and then I'll make money from it. Because this, you know, CDs are going to come out and streaming is going to come out and there's so many different ways of monetizing this. So smart move.
Jen Kof
Smart move by demand. Well, actually, let's cover some of the.
Alex Rampel
The.
Jen Kof
There was a question about the walled garden metaphor that Daniel had here. So the implication is that the new restaurant is direct to consumer. Why wouldn't the company sell to the end user rather than a business that is ultimately the intermediary?
Alex Rampel
This is a great question. So this is like Velix is a good example of this, right? Like, Velix could have sold their data to Harvey. Instead, they realized this exact point. It's like they should just be in this business of selling directly to. They shouldn't be selling to Wilson Cincini anymore. Or if they are, they should dramatically change the pricing of their products. They should change their pricing strategy. And instead of saying, we're going to charge this tiny subscription fee and allow so much of the value creation to occur elsewhere, we're going like OpenAI on their OpenAI chart is very, very little per million tokens. We're just going to consume that and then enrich everything that we have that is proprietary to us and then go sell that directly. So, you know, it's a good question, but I think the point from an investment lens is we, a lot of entrepreneurs are now looking for. Sometimes it's like existing companies where it's like they don't know what's going on. They can just buy that data. Those existing companies, if they're run by an entrepreneurial CEO, like, they realize, wow, I can make my business 10 times better. And we're going to go invest in those. And then lastly I'm just going to buy some antenna from Amazon and like listen to Malaysian Airlines flights or whatever and then aggregate this information that's completely free. Free, but it's not free. Past tense, right? Like the number of subscribers that Mr. Beast had five years ago. Like the number of subscribers today. You just go to YouTube, you see exactly what that is. If I wanted to see what that was 10 years ago, that's what is actually proprietary. So sometimes the proprietariness, if you will, everything is free. Anybody can go collect this stuff that's free. The value only accrues over time. And there are a lot of examples of this. Like, you know, I can go to the Mormon Church and get my genealogical information and they'll probably give it to me and I don't have to go pay for an Ancestry.com account. But it's kind of useful and easier to just do it with Ancestry.com than to go fly to Utah. So sometimes just the ease of going to somebody who's already digitized and put this information in an easier to digest form, that's one of the reasons why people go to LexisNexis. That's one of the reasons why people go to a lot of these providers. Because sometimes they're the only game in town, sometimes they're the best game in town, but increasingly today they're the ones that can actually give me a finished product. And actually it saves the end customer money as well. Because I don't really want to buy LexisNexis data. I just want to know if I should accept or reject this transaction. And there's a lot of enrichment that I do of the data. There's a lot of workflow, there are a lot of analysts. Like if I'm a financial services company, I hire fraud analysts to go tell me what's going on. And, and the raw vegetable that I need to figure this out is this LexisNexis information. But LexisNexis, like this would be kind of bullishness for an incumbent. Probably can do a lot of things if they're the only ones that have that information.
Jen Kof
Great. Alex, I feel like you paid Joe to ask this question, but I'm going to take it here and then we'll switch gears to aneesh your two sections here. What is your view on white collar services? AI rollups, ie fully verticalized software plus services companies that are popping up.
Alex Rampel
Yeah, so I read an article about this two years ago. I called it Barbarians at the gate but where the barbarians installed with an AI in homage to the RJR Nabisco deal in the 1980s and a book that was written about that. I mean, I think it's very interesting is what we're great at is like here are two people that are going to change the world. They don't know how they're going to do it. We're buying an out of the money call option. There are a lot of private equity firms out there that are like, we're good at firing everybody and moving people to the Philippines and doing this and doing that and all of these kinds of things. Like this is a big thing that private equity is looking at the same time. We do have a couple bets in this space and it's a very, very smart entrepreneur. But there's never a question of can I get more clients as an accountant because I can't hire more CPAs to do tax returns. It's like the hardest part is to get the clients. So you have to go to the Chamber of Commerce meetings. It's just very, very hard to buy one accounting firm and then by virtue of like all sorts of cost synergies, you can now onboard 10,000 more clients. Like that's just like the way that you would have to play that game is you buy one accounting firm you like integrated for nine months, then you go buy another accounting firm, then you buy another accounting firm and yes, is there value at the end? Absolutely. But you probably have to buy 200 accounting firms and then you're left with a pretty interesting business. And there's probably a big competitor called mid market PE who's done this for 500 years, not years, but has done this 500 times. And they're going to do a better job of that playbook book. On the other hand, there is a strategy that we think is very interesting which is instead of having a sales team, you buy one. So you know, take the example of debt collection. I could buy a publicly traded debt collector that has lots of people that doesn't do a very good job, that doesn't follow lots of laws. And I want to get started. Somehow I built this great tool that I believe in. I want to dog food it. I don't have any customers right now. I know I'll buy a company that has declining revenue, but five blue chip clients. I'll buy this company for three times ebitda and now I'll transform it with AI and now I don't have to buy a second one. I don't have to buy a third one, I don't have to buy a fourth one. I can just say I have better collections rates. I have five blue chip customers that love me and I'm cheaper. So do you want to be lazier and richer? You're like, yes, I already have the customers to back this up and I can now onboard a thousand customers into the existing acquisition that I made. That's actually quite interesting. So the question is, which one are you doing? Doing? And I think the we're going to go roll up, you know, a hundred dental clinics or we're gonna, and we're gonna make it better. We're gonna roll up, you know, dermatology. I have a friend that rolls up dermatology clinics. It's like, I just don't think we're good at that game. And the problem is that dermatology clinics is like, just because I bought one in San Carlos, it doesn't help me like do anything in Florida. I gotta go buy more there. Same with accountants versus, you know, debt collection. That's very, very national. You could buy one and then that is your entry point. And it's kind of an opportunity. Cost of do I hire salespeople to go sell or if the best companies have hostages, not customers, do I buy some company that is stagnant and even shrinking because they don't know how to respond to AI? Because by the way, all of these companies, like every debt collection company, they'd be crazy not to look into doing AI on their own. So it is this battle between startup and incumbent. But there is an interesting opportunity and we've done one in the, in the MSP space, managed service provider for it because a lot of it now is not, hey, come into my law firm office with 50 people and fix my printers. It's like onboard me into Microsoft Office and like all of that stuff can be done remotely. It's a very, very digital experience. It's a hundred billion dollar market. Like that's a little bit more interesting because I can actually ingest more clients that way as opposed to I have to buy hundreds of these things. So hopefully that that makes sense.
Jen Kof
Awesome. All right, should we switch gears? Yeah.
Alex Rampel
So I want to turn it over to Anish because all of these things that we're talking about, they also apply to consumer. So maybe with that, why don't we talk why and how this applies to consumer. Great.
Anish Charya
Actually if we're going to do that, why don't we skip ahead a slide and then we'll come back to this. So great. So this is the application of all the categories that Alex outlined to consumer AI. It's the exact same pattern. So the first and very important one is, is traditional categories are going AI native. This is happening. So if you look at Photoshop, it's a fantastic business, but what do you do if you're a young designer coming up in their career? You want to use the AI native Photoshop, the AI native Photoshop is CREA, that's over 18 months. So it's a fabulous product and it has all the AI primitives built in and it's the one that's being chosen by people that are adopting a first design tool and are early in their career. So this sort of transformation of existing categories is definitely happening. You know, the second is category creation. Eleven Labs is a fabulous example of this. This sort of market for voice and audio models really didn't exist five years ago. There was no, I mean, perhaps people doing voice actors and voice dictation as a niche market. It just wasn't interesting. Eleven has done something much more ambitious. They're a model, a model provider and they have both consumer and enterprise SKUs. And because they vertically integrate, they're able to really go after this opportunity, create the category in a very short period of time. Finally, proprietary data. Alex talked about proprietary data. It's actually near and dear to my heart because I worked at a large scale consumer company that was based on proprietary data, which is credit karma for many years. So I've seen this playbook and it works extraordinarily well. The area that we've actually seen it applied in one of our investments is a company called Slingshot. Slingshot is an AI therapist. How do they collect their proprietary data? Well, they actually go to existing therapists and they provide an AI AI scribe, a note taker, and the note taker takes notes while those therapists counsel their patients. It then uses the generated notes to train a foundation model. And the foundation model trains a consumer product called Ash, which is then sold directly to consumers. Of course, OpenAI and ChatGPT are formidable, but they simply don't have the data that Slingshot has. And as a result, Slingshot's able to provide a differentiated and high priced product and it's working well. So each of the sort of observations Alex made is absolutely playing out in consumer AI and we're very consistent in our approach to the three. Do you want to go back one? I think this is an important slide as well and an important concept because a very fair question is, well, why aren't either Labs or sort of big tech who have real model efforts like Google going to win it all. Well, the reason is that in many categories being an aggregator of models is actually preferable to consuming just a single model. And the metaphor that we're all familiar with here of course is airlines. It's much more useful to search for a flight from S.F. to New York on kayak because I can look across the inventory of every airline versus just going to Delta or United and looking at their inventory alone. The same thing is true in categories like Vibe Coding or Creative Tools where you really want access to all of the models. And the reason for that is the models each have their respective specializations, so they're not exact substitutes. You want to work with them all, you want a single pane of glass. And the labs and big tech companies can sort of definitionally only use their own first party models. So this is why we see the aggregators winning. And it's an important trend and sort of investing principle for consumer AI.
Alex Rampel
The key thing, I mean everybody's heard this framework before, but our job is to find, pick and win deals and then once we win deals to help these companies actually achieve their objectives. And most importantly, don't screw them up by giving them bad advice and telling them what to do. The CEO knows what to do and we're the there to advise and consent. But the way that we do this is we try to be the leader and the expert on, on every market we're putting out more benchmarks. Like there's actually a really cool benchmark that we're, we're coming out with. It's like an AI productive, AI productivity benchmark. So for all these different categories, actually this is pretty cool. So everybody on the team and the way that I would kind of phrase this is we have a process interrupt job. So our interrupt is there's a very, very incredible deal. Like incredible, incredible, incredible. Like let's go meet with them. Drop everything. This is unfortunately from my wife and children's perspective, like a weekly occurrence right now where it's like, ah, gotta cancel this. I have to have dinner with this entrepreneur who, who has discovered that the fountain not of youth but of, you know, perpetual motion or so they think. So go meet with them. That's the interrupt part. The process part is like, you know, I'll give you a good example. Somebody is going to out Salesforce. Salesforce, not for the hostages that they have, but is going to build the greenfield version of Salesforce. Because how is that possible? Everybody hates using Salesforce. There's a new company that's going to do this better, that's going to be AI native. How do we make sure that we are adept at finding, picking and winning and supporting that investment? Well, we believe in adverse selection versus positive selection. So a very inexpensive deal that has been hanging around the hoop for six months, that's probably bad. We don't want to meet with them. We want to meet with the best company. If it's the best company. Every other venture firm also wants to meet with the best company. Obviously they're going to send out their big guns to go try to win that deal. And it's very hard to win these great deals. So the best way of starting with this is to write this article. And we made a video about this as well, which has had like it's hundreds of thousands of views. It's pretty incredible. Death of A Salesforce why AI Will Transform Sales Joe Schmidt and Mark Andrusko on our team wrote that everybody wants to talk to them, but ultimately knowing what you're talking about really, really matters or you know, death Taxes in AI, we've covered the gamut on everything around taxes. What about companionship? What about like we do something that, that we just came up with, like, what are the top 50 enterprise applications, the top 50 consumer applications? You know, we often get a somewhat a pejorative joke of. I think it's a compliment. We're a media firm that monetizes with venture capital. But there's a method to this madness. And the method is it's helping us find deals, it's helping us pick deals and it's helping us win deals. So next, and this is the team that does that. So everybody, you know, again, we've, we've got process interrupted. But you know, we have a very, very prolific process calendar where we're publishing things, we're becoming experts in certain categories and trying to find entrepreneurs that are positive selection, that are building the best things here. And we always see them. And like a good example of this is Rillet, where if you talk to Nick Kopp, who's the CEO of Rillet, you know, Sima and Mark Andrusko just knew more about this category. You know, we were in a very, very competitive Series B process. Yeah, I mean, so if you look at this chart, so we have a bunch of people like, what are the two things that a lot of companies need? They're like, okay, we need help on the accounting side because it's like, yeah, everybody wants to buy our product. I shouldn't say the accounting Side but just how do I scale a business and make sure that, you know, very important revenue is more than expenses and also how do I go build out a sales team? And on the other side you have people, as I mentioned, kind of the content generators. So like Mark, Mark and Olivia and Joe and Kimberly and Gabe, you know Kimberly, I'll call her out here like there's a company called Decagon. She introduced the two co founders. Joe has written some great content and has a lot of great people coming to him. We want to make sure that if somebody leaves or somebody gets hit by a bus or you know, whatever happens, we want to make sure that the entrepreneur experience is very good. If you think about the origin of the firm, the firm originally was the only people that we will have right check are people that have run a company or started a company and actually I joined as part of that mandate because I, you know, for better or for worse run a company and started the company. But then we realized that some of the best people to find deals like Olivia is just non paral in terms of her ability to find great deals and be an expert as I mentioned in voice. Hey OG So it would be insane not to have her who's like the front of the sphere finding a lot of these great deals to be working with a lot of these entrepreneurs.
Jen Kof
Great. The follow on to that question is is there a process the case hosts want to check out? What's the process for investment decision making and is the right assumption is that each partner is given a budget to invest rather than needing investment approval or how has that changed if at all?
Alex Rampel
Yeah, so we try to be highly, highly conviction oriented and I feel like my job and David's job and Anish's job is to make sure that the right process is followed. Because the mistake, the automatic mistake in venture capital is I'm old, I don't use social apps. Why would anybody want to send disappearing messages? That's stupid. Let's pass on that deal. And meanwhile you have like the really, really smart not to be ageist, like 24 year old who actually uses this tool every day, who knows the entrepreneur and says this is the greatest thing that I've ever seen. And then the old person, I'm the old person here, vetoes that deal versus the right process is yes, we do have somewhat of a of bunch budget and our investment committee is effectively like making sure that we believe very strongly that the process was followed, that you've met every competitor, that the work is top notch and we will Often defer to the individual who is in the arena. And our job is to just make sure process and kind of turn that second key. So it's a two key process and again much more conviction oriented. I know that doesn't perfectly answer the question, but we don't have a committee where we everybody votes and then you have to have this many votes and then it's all this political horse trading. It's all right for, especially for seeds where a lot of the younger people we have them focused on doing seeds where it's a little bit trickier. But for the, for the smaller checks which we are predominantly focused on, let's just defer to the person with high conviction but make sure that our entire process is done end to end and that this is the expert. It came from the content, you know what you're talking about and so on and so forth. Great.
Jen Kof
May just generally talk about team evolution and changes, how you're thinking about the augmentation of kind of check writers on the team, how you are evaluating the path to promotion for folks whether or not in light of some of the recent promotions and also evolution of check writers if you will hire any incremental people as well.
Alex Rampel
Yeah, I think the main thing that we often debate about just very candidly is what we want the most is probably more leverage as opposed to capacity. So we have the capacity to do lots and lots of deals. But if it's the best deal in the world, we need to assume that our counterparty is Roelof at Sequoia or is, you know, a top partner at Excel or Reid Hoffman at Greylock. Like all of these people are active. But if it's a great deal, the entrepreneur wants to talk to as many people as possible and will often be starstruck by the person that started a multi billion dollar company as they should. That makes a lot of sense. Sense. So I would say the one area that we might look to add to is somebody who probably has built a quasi generational company that is still very, very hungry as an investor. This is not like you go, this is not a retirement job. This is an anti retirement job. This will drive somebody crazy to the point where they want to retire. Because you have to work 20 hours a day sometimes. And the working 20 hours a day is something that my kids make fun of. Like you just have coffee with people. How is that working? It's like you have to have a lot of coffee, you have to have a very high tolerance for coffee and then you have to switch to alcohol at like 5pm It's a lot of work to do this stuff. But, you know, joking aside, you really need to be able to meet with everybody and when it is a great, like, how do we, how do we know, like, this is the errors of commission versus omission. We need to make sure there are five. That this actually happened with the ERP space. If we get one of those wrong, not only do we lose our money because we were wrong, but we lose like infinite money because we didn't. We didn't actually invest in the right one. So we have to make sure that we are on top of all of these people and that our team is made up of experts, experts that all of these entrepreneurs want to meet with. So I don't know if that, that answers your, your question, Jen, but the. I think the only thing that I would potentially add is, you know, when it's time to go win a superpower deal, like, we all show up together. And by the way, like, you know, I, I jokingly call Mark the Air Force because, like, if we need a big strike, like, we, what do we do? We, we call in the F35s. Like, Mark's got a few of those. We'll have dinner at Mark's house. Ben will show, like, we all show up beyond just this team. But having a few other people that can kind of, you know, lead the charge on winning deals and have board gravitas is helpful. Of course, that is how we use Brian. That is how we use Andy. Like, I'm doing that too, largely. We wanted to get as much ownership as possible. And, you know, we might need more people at a senior level not to find the deals, not to pick the deals. Of course, you know, we don't want to just say, like, hey, you're just a monkey that helps us win deals. But that is a very, very helpful thing to go do. And that's a capacity perspective, by the way.
Jen Kof
I know it frustrates folks, probably on this call to no end, because we can't cleanly attribute a certain deal to a certain GP at all fronts. But hopefully that also represents how much we think about this sport as a team sport and one in which we bring the entire kind of force of the firm to bear as a part of that. And also just in case people did not pick up, Mark does not actually have an F35, but he's the F35 that comes into to win deals as a part of that. Okay, we have two kind of last questions. Maybe we can bundle them together. And this was in reference to any observations on customer retention to date on AI native companies and then just a scale of spending required for enterprise sales for these type of companies. Maybe David or Anish, you want to take this one?
Anish Charya
I can talk a little bit about the customer retention point. I mean, so far we haven't seen a bunch of sort of price shopping and switching and I think it's important that, that the companies that are selling in the startups that are selling into these companies build a rich software ecosystem around the primitive. This is what David was talking about with voice. Like it's, it's necessary but not sufficient to provide a voice capability. You've got to build a lot of things around that voice capability. So I think one is the companies that are building rich ecosystems have do a better job of retaining their customers. I think the other thing is that AI is moving so quickly. Many of these customers are looking to these startups as their small sort of AI solutions provider and they're looking to them for a much more holistic set of things. And because new primitives are being released every day, the startups are kind of helping drive them into the future and helping them sort of capture a lot of the top line gains from the new technology. So I'd say so far, certainly on the enterprise side retention has not been an issue. And then happy to speak to consumer as well where we've also seen strong retention signs.
David Shute
Honestly, I don't think we're seeing a tremendous difference from an enterprise sales perspective. I mean, if anything, we're seeing more inbound than ever. I mean, Eve hasn't had to have an outbound motion which is kind of insane given the scale with which they're operating. So there is a lot of sort of market pull for a bunch of these categories. But at the limit, I think that they will all need significant kind of enterprise sales. And I think if anything, what we're seeing, especially when companies are selling to larger corporates, is more of a forward deployed motion. On the engineering side, I think many large companies are looking to startups to better understand where and how to apply AI within their organizations. And so if anything, we're seeing people invest more kind of on the forward deployed engineering side than necessarily on the sales side.
Alex Rampel
I mean, it's a very cultural thing which is before you hire somebody, this is kind of happening in a lot of startups. It's not happening at ge. Can you use AI for this job? In fact, Ben is the CEO of Injuries and Horowitz. He's asking that before we hire people here. And I think that mindset actually, if you do it correctly, like if you're Eve and you're like, oh, I'm just going to hire people that play golf with lawyers and that's my entire sales process and I'll never use AI for anything, and I'm just going to use NetSuite and I'm just going to use QuickBooks. Like, that's not how these companies are actually orchestrated. Like, they really, they understand the transformative power both on a cost side and a revenue side, and they're transforming themselves internally.
Jen Kof
All right, with that note, thank you all for joining and talk to you all all too. Thank you.
Alex Rampel
Thank you.
Podcast Host
Thanks for listening to this episode of the A16Z podcast. If you like this episode, be sure to like, comment, subscribe, leave us a rating or review, and share it with your friends and family. For more episodes, go to YouTube, Apple Podcasts, and Spotify. Follow us on X16Z and subscribe to our substack@A16Z substack.com thanks again for listening, listening and I'll see you in the next episode. As a reminder, the content here is for informational purposes only, should not be taken as legal, business, tax or investment advice, or be used to evaluate any investment or security, and is not directed at any investors or potential investors in any A16Z fund. Please note that A16Z and its affiliates may also maintain investments in the companies discussed in this podcast. For more details, including a link to our investments, please see a16z.com forward slash disclosures.
This episode challenges the prevailing notion that the core story of AI is all about building better models. Instead, the panel argues that the most significant AI opportunity now lies in applications, distribution, and defensible business moats. Together, the team explores how rapid product cycles are driving the AI era, how AI apps are growing faster than previous platform shifts, and what it takes to build enduring and valuable companies in this space.
Three core themes are explored:
[00:44–03:11]
[02:50–06:43]
[06:43–13:20]
[13:20–20:55]
[20:55–29:50]
[30:08–40:47]
[40:11–45:28]
[49:18–53:18]
[53:20–56:54]
[56:54–63:30]
[66:47–69:14]
This episode delivers a comprehensive framework for evaluating where real AI opportunities lie in 2026 and beyond. The key takeaways are that groundbreaking value will accrue not just through foundational models but through inventive applications, sticky workflows powered by proprietary data, and deep integration into labor and vertical markets. Moats matter more than ever, and defensibility now lies in aggregation, end-to-end workflow integration, and entrenched proprietary data — not merely momentary model advantage.
The conversation offers practical criteria for both investors and entrepreneurs to assess what it takes to build enduring success in the rapidly-evolving AI era.