David Haber & Alex Rampell
Why AI moats still matter (and how they've changed)
You can watch this conversation on YouTube here.
Erik Torenberg: We’ve spent a lot of time talking about moats and how moats have evolved, and whether there are still moats in this new era. Why don’t you reflect and share some of the conversations we’ve been having here, some of your perspectives on this broader moat question. David, we’ll start with you.
David Haber: Maybe just to jump right into it with a hot take, I think moats still matter. They’re largely the same. I often think about this in terms of differentiation versus defensibility. I think AI is an incredible tool for differentiation, the idea that a voice agent can speak in 50 languages, fully compliantly, 24/7, is highly differentiated, certainly versus a human. But the AI-ness of that capability, in my opinion, is not a source of defensibility. It’s largely differentiation. The defensibility of a software product resides, in my opinion, in owning the end-to-end workflow, in the context in which it’s applied, becoming the system of record, having a network effect, and deeply embedding yourself within your customer.
And I think these were the heuristics that we would always look for when evaluating software companies. I think the thing that is fundamentally different about this product cycle is that the software itself can actually do the work. Therefore the market opportunity for software today is no longer just IT spend; it’s largely labor.
Alex Rampell: The challenge often has been that everybody can build something at small scale, and some of the defensibility moats only become apparent at large scale. Take an example from a long time ago, from the pre-AI era: if I am building an anti-fraud company and I’ve seen lots of people, am I going to do a better job than a net new anti-fraud company that’s seen a few people? The reason why this would be called a data network effect—although there’s another podcast that Martin and I did a long time ago debating whether data network effects are real—is that it’s almost like gravity. One atom actually exerts gravity on you, but you only really see it at very large scale. The Earth, you notice the gravity. The Sun, you notice the gravity. Jupiter, you notice the gravity. You don’t notice it for that glass.
It’s the same thing for a lot of these data network effects. At very small scale, when you have 20 companies all saying they’re going to stop fraud, they’re all building the same things with the same algorithms, but when you’ve seen 4 billion people and you know these people are bad, now you can sell to each incremental customer because you’ve seen more customers and can get better results.
But the challenge is that a lot of these moats only really become evident at mega scale. The same argument would apply, I’ve seen four customers and David’s seen three, pick my software. That means there are 8 billion customers you haven’t seen. There are 8 billion customers he hasn’t seen. What’s the difference? Whereas at mega scale, if I’ve seen 4 billion customers and he’s seen 1 billion customers, it’s easy to see that my product results will be better, but that’s at scale. In the zero-to-one phase, it’s hard to make the argument that I have better fraud underwriting. If it’s “AI do the work. I’ve done more phone calls to a particular type of customer and therefore I do a better job.” It’s hard to make that argument at subscale.
This is often the challenge. It’s self-evident that if you become the biggest company in the world, then you have a moat, but how do you get to the scale where you could actually show? You can’t get there if you have 9 million ankle biters, and you are yourself an ankle biter trying to get to scale and nobody can because it’s so easy to produce software. That’s the double-edged sword of AI, it’s very easy to produce software. Everybody can go do something that is a very obvious idea, because it’s obvious and everybody’s going to build it, but can you get to the type of scale where you actually could show a moat? That has gotten arguably harder because you have a larger end count of potential competitors, but if you get to mega scale, then you could show the moat. That’s the zero-to-one versus one-to-N.
Erik: Maybe talk about what’s different about defensibility for even the bigger players today in the AI era than it was in, let’s say, the Web 2.0 era. Are the companies today more defensible, less defensible, or how should we think about the strength?
Alex: I don’t know. I think the less defensible part—this is why a lot of enterprise software has gotten beaten up in the public markets. It’s two reasons. Number one is that if you’re doing perceived pricing, how do you come up with a pricing model that people feel is fair? A lot of it is just psychology. For whatever reason, for the last 20 years, it’s been per-seat, per-month with—you’ve heard my joke—the tall grande venti model of software charging. Somehow that felt fair, and whether that is fair or not, I don’t know, but people are like, “Oh yeah, it’s $85 a seat per month. Yeah, okay. That sounds reasonable.” Whereas if you proposed that pricing 40 years ago, you would’ve been laughed out of town. So this just became the norm.
The reason why, as I was saying, public software companies have been beaten up a little bit is: uh oh, maybe you won’t sell as many seats. Is Adobe going to sell as many seats if now you don’t have to hire as many graphic designers? Or is Zendesk going to sell as many seats if the software answers all the queries? The answer is no. It doesn’t mean that the companies are toast. They might actually quintuple their revenue because now they charge per outcomes as opposed to charging per seat, but that’s part one.
Part two is wait a minute, now everybody can vibe-code up a Zendesk competitor. So maybe companies will just stop buying software. This one I don’t think we’ve seen at all, but I think there are these two-sided risks. But to answer your question, does defensibility change? Well, if you now are able to code your own software, why am I paying? Your margin is my opportunity. Well, look at the margin of software companies. Salesforce has an 80% gross margin. They should have a 1% gross margin, or nobody should use Salesforce anymore. That would be the pro case of moats really starting to disintegrate, but I don’t think we’ve seen that happen at all, because it turns out, on one hand, two things are actually happening.
One is that this is Clay Christensen theory: the incumbents overshoot the market. So the amount of features in Salesforce or Zendesk or NetSuite way exceeds the feature set that you need, that any individual customer needs, because it’s meant to encompass all of these weird edge cases. And you see this if you use Microsoft Word. When was the last time you wrote a book? Never, right. I haven’t written a book. It has all of these things. They probably have 50 software engineers. But if you do write a book, guess what? Microsoft Word has all these features just for book authors to make a table of contents or something. I don’t use that. So they keep bundling more stuff in there, so they overshoot the market, and theoretically it’s going to make it easier for somebody to disrupt them.
But going back to where I started with this topic, it turns out that this concept of “I’m just going to vibe-code Microsoft Word”—there are all of these edge cases that you just don’t know about. So it’s actually, why don’t you grow your own food or weld your own aluminum or build your own house? It’s easier to use this concept of comparative advantage and just say, I’m going to buy something off the shelf.
So anyway, I think moats matter just as much as they did before. The one change is that in the supply-demand equation, there’s conceptually more supply of software because the barrier to creating this stuff has gone down dramatically.
David: I think the flip side to that too is that while there will be more software, and again the marginal cost of producing software is declining asymptotically towards zero, the way that these companies are getting more deeply entrenched within their customers has differed. Because again, the software is doing the work and therefore in many cases is actually replacing labor. And so if you’ve transitioned a team out that has now become your software, you’re now much more dependent on that product to run your business. And again, is it more difficult to replace that software with another piece of software or to rehire that team? I think it’s an open question, but again, the software is doing more of the work and therefore I think getting more deeply embedded within their customers.
Alex: Well, part of it is just the Goldilocks zone of pricing. I wrote some tweet, or whatever it’s called, X thread about this a long time ago. I call it the janitorial services problem because if I went to you—you’re the CEO of a giant company where you write your books in the future, so you have a 300,000-person company—I find you. “Eric, I can get your toilets 9% cleaner and save you 1% on your janitorial services spend.” Not only do you not care, you don’t even care enough—you won’t even exercise the mental energy to find the person in the company who does care. And that means that your janitorial services spend will never change. And the problem is it’s hard to get in. The good news is it’s hard to get out.
Whereas for something that’s like 90% of my profits go to you—I’m now 90% of your profits, says the CEO of GE—they’re going to me, your number one priority is getting the hell off of me, right, and doing RFPs left and right. So part of it is also just how relevant this is. And there are some companies that operate in this Goldilocks zone of irrelevance, like these janitorial services, where even if you have 9 million competitors, they’re just not going to go anywhere. Which is why a lot of the strategy that we talk about internally is greenfield. Those companies are stuck for good. Is there a high rate of new company creation that will not use the crappy old janitorial services company but will actually resonate with your pitch of “I will get your toilets cleaner and I will charge you less money”? That really resonates, but that’s not going to resonate to the people that are using the old-fashioned stuff.
Erik: What are examples of companies or spaces in the Goldilocks zone, and what’s an example of companies or space in the greenfield zones?
Alex: Well, payroll companies, right? ADP and Paychex. I mean, these are companies that are collectively worth hundreds of billions of dollars, very profitable. You could do your own payroll, actually. It’s a good metaphor for software in general. Why is it that I can’t just pay you? You’re my employee. Why can’t I just cut you a check? Well, because I have to withhold taxes. Well, how much tax do I have to withhold? Well, it depends, right? And there’s this super complicated lookup table. It’s like, well, you live in this county, but you spend this many days in New York and this, that, and the other thing. Oh, and you owe child support, and the IRS is garnishing your wages. All of these things that are very complicated. So it turns out it’s just cheaper to go to ADP, and ADP just charges you, I don’t know, 50 bucks a month per person. You might be paying a hundred thousand. It’s a paltry sum compared to the overall amount of payroll. So nobody really switches their payroll companies. That would be an example of one.
On the other side, I had a lot of companies coming out of 2022 where the market really went through a downturn and they’re like, “Wait a minute, I had a thousand employees. I downsized to 200 employees. I had a thousand licenses for Salesforce.” What’s a thousand times a hundred dollars a month times 12? That’s $1.2 million a year. Wow. That’s a lot of money because they only have 200 employees and I only have six months of cash. I gotta save that. And they didn’t do that for their payroll spend. A lot of companies do want to rationalize their overall software cost, especially for these things where they recognize in aggregate most people aren’t actually using the seats.
So I’d say Salesforce-type stuff, some of the creative tools—Adobe is very expensive and you might just do a wall-to-wall license saying why not. But then you look at, “How do I save $5 million?”—nobody’s using this. Well, it’s $5 billion. Whereas for things where inextricably the delivery and the payment are linked, which is very different than per-seat pricing for software like payroll. Obviously I’m not going to pay for payroll services unless you are employed here. Whereas we have 600 people that work at our firm. I think we have 600 licenses from Microsoft Office 365. I bet there are a lot of people here who have not opened Microsoft Excel in a year. So why are we paying for that? And that would be the idea of rationalizing software spend. So it kind of depends, but I think per-seat pricing, where it’s just easier to pay for the entire thing wall-to-wall, you’re in your entire organization—those are often the first to go, versus things that are inextricably linked to the actual usage.
Erik: So you mentioned earlier there was this concern that maybe instead of Zendesk, there’ll be a vibe-coded version of it, but we’ve seen none of that so far. Is your mental model that we’ll see it in examples where the cost is significantly high, or in which there’s greenfield opportunities? Or what’s your mental model for the types of software that will be replaced?
Alex: I think the greenfield one is always true, but when you look at greenfield opportunities, you need two things to be true. You need the entrepreneur to be very patient and say, “I’m not going to try to sell to everybody who’s—if I’m starting a net new payroll company, I’m not going to try to sell to GE because I recognize that they are hostages to ADP and that’s never going to change.”
So one is the patience of the entrepreneur, and the other one is you just need a high enough rate of new company creation to really make it work. Which is why I like to pick on one space of electronic health records or electronic medical records. How many new hospital systems are created every day? I mean, it rounds to zero. So if I’m trying to go build a new EHR system to go compete with Epic or Cerner, I can do that. There are a lot of edge cases there, and I might have patience as an entrepreneur, but wait a minute, I need to sell $5 million deals to big hospital systems. Every single hospital on Earth is currently using an EHR system. Going to be really hard to make that work.
So I think both of those need to be true: the right type of entrepreneur who’s willing to be patient, because it’s often a very lonely game of, “I built this great product, wait a minute, I don’t have any customers yet.” And you want to see high traction because you’re seeing in the rest of the market some companies are just going like this, and my company’s not. And I’m in Silicon Valley and I need to recruit the best people. They want to work at the company that has the graph like this, but you need this. Greenfield requires patience.
Erik: So we’re talking about how moats still matter, and in many ways they look pretty similar. Let’s steelman the other side for a second. Why are we even having this conversation where some people say, hey, brand is everything, or shipping velocity, because this era is different? What’s the steelman of their argument?
David: Look, I think this market is noisier than ever. And so I think finding ways to stand out from the crowd probably matters more today than it has in the past, I would argue. I think the other thing is that the underlying technology is changing so quickly. And so as a founder, you want to be living on the frontier and understanding what model capabilities look like, because it can dramatically change the efficacy or the capability of your underlying product.
And so one of the things that’s changed that’s been really interesting in this current wave of especially vertical applications that we’ve seen is the type of founder. I think founders today are often younger and more technical than we’ve seen in prior generations. And so they’re less often native to the particular industry, but they’re fluent in the toolset. And I think that’s really important because, to the same point, you’ve got to stay on the frontier and understand what’s coming.
At the same time, I wrote this piece that I call “Context is King”: while it is important to understand model capabilities and what’s happening in the frontier, you still need to figure out how to apply that technology. While the founders themselves are maybe less native to the particular industry, they’re still hiring for context very early in a company’s lifecycle.
A good example of this that I sit on the board of is a company called Eve. The two founders of Eve were the earliest employees at Rubrik, which is now a public infrastructure company. They built a legal AI company in the plaintiff law space. Neither of them had any particular background in employment law or personal injury, but they deeply understood how to apply document extraction technology and voice and LLMs more broadly to this very particular workflow. And they’ve hired a plaintiff attorney actually on staff. So anytime a new model is released, they’re understanding from people in industry the impact that it’s having on drafting, on their ability to reason through a case or a matter.
It’s sort of this tension of building the brand, having momentum, understanding what’s happening in the frontier, and yet figuring out ways to apply that technology in the context of your specific customer, because I deeply believe that that is where a lot of the sources of defensibility reside.
I’d love to find other examples of businesses where the technology reinforces their business model. It doesn’t compete with it. Meaning in lots of areas of legal, if you make your employee 50 times more efficient, you’re eroding your billable hour in their business. They operate on a contingency basis, meaning they only get paid if they win. So there’s no sort of limit to the amount of AI that they want to adopt. And if you can become 5x more efficient, you can take on 5x more clients. Anyway, these are sort of characteristics that I’d love to find more of, and hopefully that can be a bad signal too.
Alex: I think the other steelman is if you believe that brand matters, which it almost tautologically does, because what do I buy? I buy the thing that I’ve heard of, right? So there’s an advantage there. And if you believe that for a lot of companies and products, somehow having scale is effective—so not a network effect, but a scale effect. So if I’m Honey Nut Cheerios and I know that people are going to buy lots of my Cheerios, I can build a big factory and not hand-crank out each Cheerio. I’m going to have these compounding advantages just in terms of economies of scale.
Amazon, does that really have a network effect? No. It’s kind of nice that everything that I buy will show up the next day or in two days. And how can they do that at low cost? Because so many people are buying things. So there are some things that have scale, and those things also benefit from brand. So if you can move the fastest—if you can agglomerate capital and labor—so it’s like I raised the most money, it’s a very generic idea, but somehow, like most other things on planet Earth, if it’s the biggest and really, really big kind of gravitational scale, then it’s just going to work better.
So can I get there the most quickly? But there are 20 companies that are doing the exact same thing. And at that point, I wouldn’t say that momentum is a moat per se, but momentum has the highest chance of getting you to gravitational scale where you do have a moat. And if you don’t do that, by contrast, you’re just going to get eaten alive because you can’t hand-crank out the Cheerios. You have to get to the scale where you’re able to build a factory. And you have the biggest factory, you can crank out the most things at the lowest cost. So what is the trajectory? What is the slope of you versus all of your competition? And if you have not a good slope, you’re just not going to win that game.
Erik: One of the questions for defensibility and Web 2.0 companies was, “Hey, would Google build this?” Or Facebook, or name your incumbent. In the AI era, it’s, “Will OpenAI or will some other major company?” How should companies think about that framework in the AI era?
David: It’s funny, I feel like 18 months ago, “GPT wrapper” was on everybody’s lips. And I think it was largely used as a pejorative. And I think, to some degree, there are some spaces where the model capability and the application capability, if they’re very overlapping, I think you’re going to be in a risky spot. I think one of the remarkable things that’s happened is there are so many markets that were never particularly interesting to sell software to that are now radically interesting spaces to build companies in. Again, in large part because the market is now labor, not just IT spend.
Plaintiff law being an example. We have a company called Salient applying voice agents to auto loan servicing. Five, six years ago, would we back a software company selling to non-bank auto lenders? Probably not. The company’s doing incredibly well. Again, in large part because the capability of being able to speak in 50 languages, fully compliantly, with customers in 50 states, working 24/7—it’s just so differentiated versus the individual. And they’re finding that their ability to collect is meaningfully higher than that labor. The cost-benefit tradeoff is so dramatic. The company is getting a lot of revenue from those customers who may not have had millions of dollars of IT budget historically and are now very willing to pay for a product like that, given the impact on the business.
Alex: The way that we used to talk about this a long time ago—and this almost had a pejorative slant to it—are you building a feature, a product, or a company? And what’s the difference between the three? Well, a feature is like there’s an existing product and you tweak that product to make it marginally better. A product is, hopefully, some system of record or something that keeps track of something. And then a company is probably the most defensible of those three, where you have a product and maybe you own a platform. The platforms tend to be the most valuable companies.
But a feature is, I’ve built a Chrome plugin. And that doesn’t mean—and there, by the way, were are lot of Chrome plugins. Honey was a Chrome plugin that got bought for $4 billion. I wish I had done that, right? That’s a good feature, but that was a feature. A product would be like, oh, I built my own browser. And a company is like, all right, well, my own browser company actually makes money. You don’t actually have a company, even if you have 10 products, if you don’t have a sustainable path to have that company be around in 10 or 20 years.
And I think another way of thinking about what David just said is that now the features—the feature was the most pejorative and seemingly small of all of those three—almost obviously, some of the features can be incredibly profitable. Because it’s like, wait a minute, this feels like a feature because it could get added to Salesforce or it could get added to one of these other things. But the amount of money that I can charge for my feature is orders of magnitude more because it’s like, “Hey, I’m going to be the front-office receptionist for your orthodontic clinic.” That’s my job. That’s the feature. And it sits on top of whatever software you currently use, but the feature I can now charge $20,000 a year for because it is doing the job of labor.
Uh oh. Will the existing product that my feature is riding on top of, will they go build those pieces of functionality? And/or will another company show up that just says, “Hey, we’re going to sell the greenfield with a new product that has this feature set embedded”? And feature, product, company—it still is out there, but I’ve just never seen a world where the features, if you will, can get to revenue scale as quickly.
By the way, you often have to start with the feature because a customer isn’t—think of it from the customer’s perspective, the customer being the business buyer of software. It’s like, “I know I want to be locked into a piece of shit software company for 20 years. That’s what I’m looking for as a buyer.” No, it’s like, “Ooh, I have a problem to solve. My problem is I can’t hire a front-office receptionist for my orthodontic clinic, or I can’t call people in Mandarin or Cantonese to get them to repay their auto loans. What do I do?” Oh, something shows up and it offers that functionality. Boom. I’m a buyer, and then that functionality has to—that feature has to backfill product, backfill company as quickly as possible. So that’s still true today as it was 10 or 20 or 30 years ago, but the difference again is that the feature, the revenue for the feature, is just so high, and the demand for it is so high, because again, in many cases you’re just responding to help-wanted ads effectively.
David: And so I think the effect of that is that there’s been sort of a Cambrian explosion of interesting markets to go after. I think it’s unrealistic to believe that OpenAI is going to go build the front-office assistant for the dental clinic as their core business. They’re not going to do that across every single market. I think the other dynamic is that for many of these companies, part of the product value is actually orchestrating the work across lots of different model companies. And so I think having one foundation model business going up the stack, I think, limits the actual impact of the application potentially as well.
Alex: Well, I think if you think about this versus other platform companies—so Facebook was the preeminent platform company of Web 2.0, so call it from whenever they opened up Facebook Platform, which I think was like 2007. People built their businesses on top of Facebook. Facebook would never do those particular things. So Facebook is never going to show up and say, “Hey, you know what? We should build a farming game.” They were like, “No, we’re going to have a platform that allows companies like Zynga to build these farming games.” But what the platform normally does, if they don’t actually go compete with the underlying products, is they say, “I’m going to tax it, but I’m going to tax it in ways that are kind of at my fancy.” So this week it’s 10% taxes. That’s my promise. Oh wait, I changed my mind. Now it’s going to be 40% taxes. So that’s why it’s always dangerous to build on somebody else’s platform.
So I think the two things to look at are, number one, will the platform owner compete with what I’m doing? And that’s also another Goldilocks zone question. Because why is it—I published this graph of VisiCalc versus Lotus 1-2-3 versus Excel. So VisiCalc invented the spreadsheet in 1979, had a hundred percent of the market because they were the only player in town. Lotus built a better version of that. Lotus got to, I think, 70% market share by 1985, which was when Microsoft released Excel for Mac. And then by 2000, Microsoft had 96% market share. And why is it? Because they owned Windows. The platform owner normally wins, but that’s because it was just so huge. Why do I buy a computer in 1997? Because I want to use a spreadsheet. It was just so intrinsically linked. That was one of the main use cases for computers in business use. It’s like using spreadsheets. So that was a violator of the Goldilocks zone.
Whereas other things where all you have to worry about from the platform owner is that they’re going to tax you, but they might tax you in very bizarre ways. But part of what David was saying, in terms of there are multiple model companies, which is great. The problem with Windows was that it was 95% of the market. 95% of your customers used Windows. So if I’m going to go build a competing spreadsheet, I’m just toast because the platform owner is just going to drown me. Now there are five model companies, or more when you include all the Chinese models and whatnot, open source. I don’t have to worry about that, but I do have to worry about them saying, “Wow, this is so relevant.” Why is it that OpenAI got a public company CEO to quit her job just to become the CEO of applications at OpenAI? Maybe because they have a huge application opportunity. But this is the nice thing, a lot of these things are so obscure, but they’re still big. But I don’t think OpenAI is going to go do them because it’s like, are they going to do dental care management? They could, but if they’ve done that, then I would be short OpenAI because they’ve run out of good stuff to do. That’s something that they should do in 2029.
And then this is—I think I told you this story before. This changed my outlook on life. When I pitched this guy Dan Rose at Facebook, who was running business development there, I was like, “This is a huge opportunity. You should use us for payments. We’re going to do this. We can make so much money for Facebook.” And he was so patient and nice, and I love this guy. I’m on a board with him to this day. He was like, “Alex, that’s such a great idea.” I was like, “All right, I got the deal. Yes.” He said, “It’s a great idea, but we’re not going to do it. Because you’re pitching me a gold brick, like we have gold bricks all around us.” He was right, I mean, Facebook in 2010. Facebook has grown their revenue; they have more profit every quarter today than they had revenue per year in 2010. It’s just such an incredible company.
He’s like, “You’re pitching me a gold brick that’s like a hundred feet away and it’s real. I love that gold brick, but we have hundreds of gold bricks where I just have to stoop down at my feet and pick them up. So I’m just not gonna do that one right there.” That’s how these big companies think, but the nice thing is that these are gold bricks. These gold bricks are bigger than they’ve ever been because you have software that can do the job of labor.
Erik: On that note, if you were running OpenAI and you were thinking about which gold bricks or how do you even—what mental model to think about what are the things that you should be doing first versus things that, hey, maybe let other people do it. How would you be thinking about that question?
Alex: I mean, I think a lot of it is where, well, it’s two things. Number one is we want to be the backend for everybody, like the platform. I think it’s two things. Number one is, can we be the platform for pretty much everybody who’s building anything? So we’re not going to go into these obscure spaces like orthodontic care, at least not until 2045. So let’s make sure that every single developer is using us.
This is part of why Microsoft crushed Apple in the 1980s: because Apple made it really hard to develop software, and what’s actually kind of interesting is that Microsoft started off as a compiler company. Their very first products, they were not Microsoft Office, it was not DOS. They built a BASIC interpreter for the programming language BASIC, and they had a big business. Their biggest competitor was Borland, which only made compilers. And the early rallying cry, if you talk to any early Microsoft employee, was “Beat Philippe.” Philippe Kahn was the CEO of Borland. So Microsoft was focused on that, made a lot of money on that, and Apple was like, “We should make money on that too.”
They had a product, it was called MPW, Macintosh Programmer’s Workshop. I remember, I used to use it in the 1980s, and it was like $2,000, I think, in 1980s money to buy this IDE or programming thing. How do you afford that? But it was like, “We have to make money on that. Microsoft’s making money on this.” And then lo and behold, there were like 10,000 times more DOS and Windows software products than there were Macintosh software products. And of course, Apple corrected that mistake when the iPhone came out. Now Xcode, which is the way that you build products for Mac and iPhone, iOS, it’s free. So they kind of corrected that mistake.
I’d say two things to answer your question. Number one is, can we be the biggest consumer brand in the world? So ChatGPT has 800 million weekly active users. Get that to 5 billion, right? Is even if Gemini 3 came out today and it might be five times better, are people that are using ChatGPT just as consumers, are they going to switch? Maybe, but it’s unlikely just because they’ve kind of made that their default, and then be the backend for everybody who’s building anything. And that way, all the gold bricks come to you.
David: I think the other thing that we should anticipate, we’re already beginning to see from some of these big model companies, is, what are the big horizontal applications that they can likely sell to every large enterprise? I think you saw today with Google’s launch, the IDE is going to be one of those things. I think if there’s product-market fit for LLMs, coding is definitely one of the top categories.
So, thinking about what are the big horizontal applications in the enterprise, I think there’s also, to some degree—and I think this has been earlier to sort of play out—it’s sort of the Palantir opportunity. I think we’re still very early in the proliferation of this technology into large enterprise. At the same time, unlike prior product cycles like the cloud, if I’m the CEO of a large public company and I’m asking myself, “Do I need to be in the cloud?” it was sort of an esoteric idea. Today I can plug a prompt into any one of these models and intuitively understand the impact that it could have on my business, the efficiency gains in my customer support organization and my engineering organization, in all of my back-office functions.
At the same time, many of them don’t know where to start, and so I think you will see sort of this consultative, sort of forward-deployed, Palantir-esque sort of sale into very large enterprise from some of these big model companies. Again, I think we’re early in that, but you’ve heard inklings of this with Anthropic talking about wanting to build into financial services and other markets.
So I agree. I think the biggest opportunities are the ones that Alex is describing, but I think you will see them selectively try to build applications that cut across every one of those. And then they’ll probably choose a few sort of lighthouse customers to build largely bespoke, custom integrations into these bigger enterprises, but where the ACVs just really make sense.
Erik: In Web 2.0, there was a lot of winner-take-most. You were talking about one of the benefits in AI is that there are multiple winners. To what extent is consolidation inevitable, or how do you think this plays out?
Alex: Well, I think if you have 20 companies that are all doing the same thing, what has historically happened is that it’s a bad market if there are 20 companies doing it, but then, the bottom 15 just go bankrupt. Then maybe there’s some consolidation where number one buys number two, number two buys number three, and assuming that we have a functional FTC and whatnot, all of this is approved because it’s not like you’re taking—this is orthodontic clinic answering software or something. Then what was a bad market becomes a good market.
This kind of goes back to why momentum is important. Because if you have 20 companies that are all at the exact same scale, then it’s actually great for the customer, which is that the prices go to zero. Or they converge on the price of electricity. This is not saying you want to go build a monopoly in orthodontic answering software or something, but rather you can charge more if you get to a certain scale because whatever the quality of the product that you’re delivering at the end of the day is just higher, and you have to get to the critical scale to get there. And sometimes you just need these markets to work themselves out.
I mean, when I was running my company TrialPay, we had, I don’t know, 20 competitors, and it was tough because everybody would be pricing their product at a loss. This loss-leader strategy only works if you end up leading. You have to make money at the end, and nobody really had a plan for that because the venture capital dollars were really subsidizing everything. And that does not make a good market.
What does become a good market at the end, and sometimes this is what Vista, the private equity firm, would do, is, “We’re going to buy one as our anchor. We’re going to go lowball and put the other five out of their misery.” And now we end up with actually a pretty good product at the end, or a pretty good business at the end, or a pretty good company at the end. So I think that will probably play out the same way here because you just can’t have a market where you have everybody loss-leading and nobody’s big enough to get any kind of scale effects.
Is there going to be a world where the 19th player survives? I mean, Jack Welch would always say you have to be number one or number two, and there’s no value to being number three through a hundred. I don’t think that’s changed.
Erik: Right. Even in the model provider example? And I’m almost curious if prices go down.
Alex: Yeah. There actually are, people know xAI, Anthropic, OpenAI, Gemini, or Qwen. They know the big ones, but there’s a long tail of things that people haven’t heard of, where they’ve raised lots of money. It works fine, but how can you survive? The model company game is the most cutthroat because if you’re state-of-the-art minus, minus, minus, and you’re trying to earn a living, that’s just not going to work. So that game is super cutthroat.
David: I think the one area where that may have diverged, and Martin talks about this a lot, is when markets are growing so quickly, you end up having specialization. And so I think in other modalities, in some of the creative tools, people have specialized to serve the upmarket. “I’m producing movies.” Okay. “I want to create social-quality content.” These are different markets that the models can kind of specialize in. Time will tell how defensible those become over time. Maybe that’s the optimistic take that early on, everything looks overlapping and competitive, but the market is growing so much that everything can kind of expand and people can specialize over time.
Features, Products, and the Danger of Building on Platforms
Erik: Earlier when we were talking about feature versus product, didn’t Steve Jobs once tell Drew Houston that Dropbox was just a feature?
Alex: Yeah. That’s why it’s always been this pejorative thing. But that’s kind of the point that I was getting to, is that nobody wants to—it’s like, “Oh, I need this company.” No. It’s like, “I need this feature.” Every now and then you see a product that is not a feature because it’s just so far out of left field. Nobody was anticipating ChatGPT dominating their daily workflow in 2022 in October. But then once it came out, it was this like, “Holy crap, this is incredible.” That’s not a feature. You could argue it’s a feature on top of your iPhone, but no, the iPhone is the delivery mechanism. That’s a product, and they’ve obviously turned that into a company.
Whereas other things, it kind of is like, why is there antivirus software? That almost doesn’t make any sense. Shouldn’t the operating system stop you from getting viruses? Why do you need a third-party tool to do synchronization between devices? But it turns out the reason why Dropbox has survived and thrived since Steve Jobs made that comment is it’s really hard to do well, and there’s a lot of other things. Once you’ve built that feature, you can backfill with all sorts of other product, which is what Dropbox has done a pretty good job of.
But it is hard because this is the danger of building on somebody else’s platform. I’m going to build this thing that they should have had, if they had the foresight, and if it doesn’t operate in the Goldilocks zone—it’s like, wow, this will triple Apple profits. Let’s just say that Dropbox would’ve tripled Apple’s profits. Would they have focused on building that versus the iPad or something? Whatever Steve’s last gizmo was—sure, but if it’s kind of in this Goldilocks zone of irrelevance, like janitorial services, yeah, they should do that, but platform owners get lazy.
This is why half the things on my iPhone don’t really work if they’re built by Apple. Any parent that’s listening to this, if they’ve tried Screen Time, it’s just an embarrassment upon humanity, because they don’t have to go sell, They don’t have to compete on feature. They compete on the fact that they don’t even compete. They’re the platform. They roll it out, it’s going to be bad, and that does create an opportunity for somebody to come up with a feature and actually outcompete the platform. But you have to be careful because obviously the platform owner is going to go compete with you.
That’s why often what I find very compelling about entrepreneurs is when they know this. They’ve studied, how is it that from every single platform shift from—we were talking about AC versus DC current—there have always been these battles for who’s going to be the underlying layer.
The best entrepreneurs have studied this and they have a plan. They’re like, “I know I have a feature.” Drew knew this. He was like, “I know that there’s this stupid comment on Hacker News. It’s like, oh, this is just rsync with this, that, and the other thing.” Yeah, of course Drew knows that. But he built this into a $10 billion company because he had a plan.
And the best entrepreneurs, they often—okay, I know it’s not this naivete. It’s like, “Oh, I’m going to build this. There’s no way that they’re going to build it because they’re too dumb and stupid.” No, they’re not. These companies, if they get their act together, they will marshal a lot of resources to go compete with you. It might take them five years, but they will 100% do it. You have to backfill your feature with a product, and you have to have a moat for that product, as opposed to, “Oh yeah, the big company will never figure this out.” That’s not true.
David: I wrote this piece a while ago called “The Messy Inbox Problem,” and it was sort of a wedge strategy that we’ve been observing across lots of different industries. And it’s just this idea that you hook into a bunch of your different unstructured data sources. Could be email, could be fax, could be phone. Tennr, as an example, has trained a model to be able to extract all the relevant patient information from those data sources to plug it downstream into some system of record, in their case an EHR. But this exists in a CRM, an ERP, what have you.
I think that that wedge for that feature is interesting in large part because it lives upfunnel from software. You are replacing the kind of human-level judgment of the individual. Often that admin, the secretary, is sort of collecting the physical facts and then plugging it into the EHR. So now a bunch of AI companies can kind of wedge in and then eat away at all the downstream workflows that might have been point-solution software companies.
Tennr is no longer just doing the messy inbox. They’re now doing scheduling and prior authorization and eligibility and benefits. And they’ve used that wedge to try to become the end-to-end platform. Eventually, maybe they’d become the system of record. Again, because you can replace the human labor now with software. I think it’s creating opportunities for these features to actually become products and, in their case, I think, become whole companies.
Alex: Well, I think this is the thing that in my mind is very dramatically different than every other platform shift, is that it is just so consensus. Cloud was not consensus. Mobile was not consensus, and that’s why the incumbents kind of screwed up, where sometimes it was just completely, I’ll use the Silicon Valley term, orthogonal to their business model. Because it’s like, “I sell $5 million a year products, and wait a minute, I’m going to charge a hundred thousand dollars a month? That’s just hard. How do I pay my salespeople? How do I make my quarterly numbers?”
So that’s why Workday beat PeopleSoft. Or that’s why Salesforce beat Siebel. So all of these things played out, but behind it was this concept of, “That new thing—that iPhone—is stupid.” There’s the famous Steve Ballmer clip of him saying “Nobody’s going to buy an $800 phone with no keyboard.” There’s no version of that for AI. How do you find a big CEO or even a small CEO who’s like, “Nobody will use that tool that makes you a hundred times more productive”? Of course.
This is why it’s kind of a bonanza for most of the incumbents as well, because anybody who has a system of record will add a button or a feature, to use our parlance, that will make them more money. So there are just gold bricks everywhere, and the challenge though, is that there isn’t this kind of white space to occupy in the same way that there was for cloud or for mobile or for a lot of the Web 2.0 things where the incumbent screwed up, they weren’t paying attention, they scoffed at this new technology. Nobody’s scoffing at this new technology. Everybody’s just trying to embrace it, but the opportunity often exists where a lot of the areas that just seem too small, that don’t have an incumbent at all, those actually might turn out to be trillions of dollars of value. And that’s kind of what makes it much more exciting than last gen where it’s like, “Oh, I’m just going to copy everything that was on-prem and make it recurring billing cloud. And I’m going to do that at a time when the big guys say that’s stupid and I don’t get it.”
Erik: Some argue that mobile was ultimately sustaining, in that although there were net-new companies and use cases that were hundred-billion-dollar companies like Uber and Airbnb, et cetera, the incumbents, some of them became trillion-dollar companies, got to buy mobile. When we look at the business impact of the AI era, what’s your mental model for thinking about incumbent versus startup, or net-new company, in terms of value capture?
Alex: I think a lot of it is the same. Unless you really screw up the pricing model, or you’re all per-seat pricing, it’s very hard to just get the market to adopt something that is just violently different, and you’re operating in the public eye and your technology team is bad, there are a lot of “ands” that need to happen. I have a hard time believing that incumbents will really suffer.
I mean, there probably are some things. Take one example of, and this kind of goes back to distribution versus technology, all of these business process outsourcing companies, these BPOs. They’re the largest employers on the planet: Tata, Wipro, Infosys. So if I’m JP Morgan and I say I need a call center, and this call center needs to have access to customer records and it needs to be safe and everybody needs to be trained, I need to have a hundred thousand people that can answer the phone. You know who can do that for you? Infosys. Or Tata.
Tata has already done the integration with JP Morgan. In this case, they might just add AI and now they don’t need a hundred thousand people, and they maintain that JP Morgan contract and they operate in the area of the Goldilocks zone, where they’re going to make a hundred times more money. That’s one case. That’s the bull case for Tata.
The bear case is JP Morgan’s like, “Wait a minute, we should partner with a startup to do this, or we should do this ourselves.” Now Tata loses that relationship altogether, and it could go either direction. I think a lot of these things are really up for grabs. But I think the default is that the incumbents probably will do well, but you can pick a lot of these cases.
This is why you see the public markets kind of don’t know what to do, where there is a case that is very bad for a lot of software companies. But there’s an alternative case, which is, if you operate in the right Goldilocks zone, and you have the right momentum to actually build these things and embrace these new technologies, you’ll maintain all of your customer relationships, and you’re just going to have a more profitable business. And it’s not that you’re going to do this—the most compelling thing I think about AI that almost everybody gets wrong is, “Oh, it’s going to destroy all the jobs. Our beloved representative from Silicon Valley is trying to eliminate AI. It’s just so crazy that our elected representative wants to turn us back to farmers of tangerines and whatnot in Silicon Valley.” Which again, I think is crazy, but it’s not like all the jobs will go away. I actually think that’s not going to happen at all.
What’s going to happen is there are a lot of things where, if I could hire somebody for a dollar to do this task, I would a hundred percent do that. I cannot hire somebody for a dollar. I’ve never been able to hire somebody for a dollar. Now I can hire software for a dollar.
So a lot of these tasks—look at how many people took taxis post-Uber, and did you hear—you probably took an Uber to get here today, right? Would you have taken a taxi 20 years ago? No way. Because where would you find the taxi? How would you arrange the taxi? It’s just way too complicated. Whereas once you make it very abundant and less expensive, everybody’s going to use this.
I think that’s what Ro Khanna and his ilk are missing, which is, it’s not like, “Oh, I’m going to go and say I’m going to eliminate all the jobs.” Think of it in that JP Morgan example that I just mentioned. Wouldn’t it be cool if every single customer of JP Morgan Chase could have their own personal friend that they could talk to every single day, that would help them with every single element of their financial life?
Or, “I’m stuck downloading the app. I can’t figure out how to get it set up. Oh, talk to somebody in real time that will help you about that.” Why don’t they do that? Because the cost is known, it’s high, and then the value is probably low. As soon as you can bring the cost down to zero, now you’re going to start hiring AI in all of these different areas that you just would never bother hiring a human for, because you can’t train the human, you can’t find the human, and the human’s too expensive.
Erik: I think that’s a good place to wrap. Guys, thanks for coming to the podcast. Moats don’t matter.
Alex: Yeah.
This newsletter is provided for informational purposes only, and should not be relied upon as legal, business, investment, or tax advice. Furthermore, this content is not investment advice, nor is it intended for use by any investors or prospective investors in any a16z funds. This newsletter may link to other websites or contain other information obtained from third-party sources - a16z has not independently verified nor makes any representations about the current or enduring accuracy of such information. If this content includes third-party advertisements, a16z has not reviewed such advertisements and does not endorse any advertising content or related companies contained therein. Any investments or portfolio companies mentioned, referred to, or described are not representative of all investments in vehicles managed by a16z; visit https://a16z.com/investment-list/ for a full list of investments. Other important information can be found at a16z.com/disclosures. You’re receiving this newsletter since you opted in earlier; if you would like to opt out of future newsletters you may unsubscribe immediately.