Episode 855 | The Eight SaaS Moats that Still Work in the Age of AI

You're listening to startups for the rest of us. I'm Rob Walling. In this episode, I welcome a A&R volset back on the show, and we record what I think might be the most important episode of the year, maybe of the past couple years. You'll notice that the runtime on this episode is longer than usual because I just let it go. I let it take the time that it needed to take. And this episode is a collaboration between A&R and I, where we talk through the eight SaaS Moats that we believe still work in the age of AI. As you'll hear once I bring A&R on the show, we are taking this very seriously and really trying to think about it from all the angles, and trying to take an outside information from folks other than just the ex-Twitter Bootsdropper ecosystem. And as investors in hundreds of B2B SaaS companies, we have a vested interest in figuring out what will make SaaS companies defensible as AI gets better and better and takes over more and more of the software ecosystem. This episode is not an outline we put together in 30 minutes of brainstorming. A&R literally spent weeks of research doing the lit review, putting thoughts together, has run it by our internal team, then ran it by me, and he and I went back and forth and collaborated on it to create not only the Moats, but there's these vectors of attack that we see AI is having on SaaS companies, most of which you haven't thought of, but that we are seeing and are aware of, as well as thinking of examples of a lot of these, because that's the hard part really is not just having a concept out on its own in an academic way, but having real life examples, maybe of big companies like Stripe or small companies that you've seen on the show like Sinwell or Jim Desk that fit into these buckets of the Moats that we discuss. This episode is a combination of two microconf talks that A&R and I gave just a couple weeks ago in Iceland, and this episode is so dense in a good way. It's action-packed and information filled that we have brought the slides from our talks as well as producer Ron's going to put together a detailed summary of what we talked through in this episode, and you can get access to that plus the Q&A that happened in Iceland that we're going to cut from the YouTube version of it, but there was, gosh, there was what, 20-something minutes of live Q&A after the talks. If you go to microconf.com/motes and enter your email, that's microconf.com/motes and you're going to get a ton of stuff. Again, video of the live Q&A after this concept, you'll get the slides from our talks and a summary, a pretty detailed of this podcast episode. I know I mentioned at the top of this episode that it might be the most important episode of this year or the past couple of years, and I think that could be true. I think that is likely true, actually, in the way that it can help you as a sassfinder think about how to shore up your business not only to stand business, not only to keep growing, but to have a great exit valuation if and when you decide to sell. We thought about taking all this information and keeping it for tiny-sea companies, putting it behind a paywall, or however else we could try to guard it, but realized that it will actually do so much more good if we are able to distribute it in this way for free to the community. I hope that you listen this episode, gosh, maybe more than once, and try to internalize the ideas that we're presenting here. And with that, let's dive into our conversation. In our full set, welcome back to the show. Thanks for having me. We are having maybe the most serious conversation on this show of the past year or two. There's a lot of episodes show up, I answer a few listener questions, but just so folks are prepared. It's going to be obviously a longer episode, but it's really our take on the state of sass in the age of AI. Correct. And this has come about for a number of reasons. One is, as folks know, you run discretion capital, which is a cell-side M&A advisory for sass companies between two and 20 million. And so you have a pretty unique viewpoint into the sass M&A market. You and I run tiny seed, 217 investments, and we are seeing things on the ground that are some are good signs and some are not good signs. That is accurate, yes. And we're trying to distill that into something actionable. Now, today's episode is kind of a, it's a version of us talking through the microconf talks we did in Iceland a couple weeks ago. And I did a 45-minute talk. You did like a 15-20-minute talk. AI could have been the same continuous talk. It's all about like how, if you're running a sass company today, how can you potentially survive AI? What are, what are some modes that are still available? We're definitely going to talk through the threats. I want to, we're going to kick off with AI is more of a problem than you think, because I think most founders, especially in the, in our space, the microconf, tiny seed space, are underestimating the threat that AI is to their business or will be over the next couple years. Yeah, I mean, I think like it's a very common thing with a lot of technical, particularly back mostly technical founders, right? And like, there's always some technology. It's like, oh, mobile's here or something's here or whatever. And like for the longest time, like taking the stoic view, like, yeah, yeah, but I'll just keep doing what I'm always done. Like that, that was the right approach versus I'm not, I'm not convinced at this point. Yeah. The Sasspocalypse happened in February and we're recording this in October. And Sasspocalypse was the public markets dropping because there was a lack of confidence in public cloud companies, Sass companies. And that rippled through the, the M&A markets. But what's interesting is we haven't really seen, it's not like we've seen our portfolio in mass like plateau or go down. That's not happening. And so that's where they're conflicting signals. And this is it's cloudy. I would agree with that. We actually have spent the past several months thinking about this topic of like how we have this massive portfolio of Sass companies, tens of millions in ARR across our portfolio. How do we advise them? And that's big part as well as, you know, when you're advising someone who is going to sell to private equity or a strategic, that's what we're trying to do. And so as you're listening to this, realize the dog and the race we have is this is advice that we're going to give to companies that we are either selling or that we are that we're invested in. You want to talk a little bit about your process for this because you and I talked about are there modes that still exist? Are there some things, you know, they can do? And I came up with several modes and you looked at them and you're like, none of those are modes in the age of AI, bro. And I was like, so, so you went off and did some, some craziness with cloud are giving the chat GPT for weeks. So there's a couple of different things here. One of the things is like, from my perspective, I was interested in like, how does the buy side think about this? Like, what is the, what are the thought leaders saying on the people that buy these kinds of businesses? So there were a bunch of commentary that was coming out of places like Gartner and Morgan Stanley and all this stuff. And so I spent a bunch of time thinking about how do they think about it? What do they think is still durable? And there was a lot of opinion writing. And it's sort of like, it's interesting because it's sort of bifurcated into pre and post, you know, February, pre and post SaaS apocalypse. The underlying sentiment was very much up until February was very much like vertical SaaS, vertical AI, whatever you want to call it. Like, AI is basically a tailwind. Like, it's going to be a great year. SaaS is the same, nothing is changed. It's maybe even just a tailwind and then you're done. Then after that, like when, and really, like, what's the actual triggers for this has apocalypse in the public markets was these releases from anthropic on the capabilities that they released where people all of a sudden, like, oh, this isn't just like a chatbot, it can actually do things. And so why ended up doing basically was sort of synthesizing do sort of a literature review of like how the buy side was thinking about it. And there was mostly writing on that side on the investor side. That's what, you know, there was some of it on the founder facing side, but most of us was on the investor side. So I was sort of synthesizing that and trying to get a sense for just map out all these various modes. And I think one of the things I realized, because I part of the reason I ended up going the investor side is because my sense is that most of the time founders up until this year hadn't really thought about modes, not really. Like, you know, they were probably relying on it and, you know, they definitely were relying on it. You know, like one of the most and we'll talk about it that's no longer a mode is just like, I've wrote a bunch of code, it took me a while and they're like, you know, it's difficult to write code and it takes a while. There's a surprising number of founders that just relied on that, not knowing that that's really the only mode they had. And so really mapping out, like, no, what are the various different ways in which why do people stay? Why do people keep paying for software and then mapping that on to the new reality that exists with AI and like which of these are still actual modes versus like kind of drawing or eroding modes or how do you think about it? And I saw someone it was a bootstrapper in IndieHacker on X, Twitter saying it was in a conversation I brought up that I was going to talk about modes in Iceland and he said, why are we talking about modes. There's like VC. things like we can succeed without modes, right? And it's like, yeah, maybe, maybe a year or two ago, maybe that was feasible. You find a niche, you, you thread a needle, you get some SEO, you get whatever, and you can grow business. AI is coming for a lot more than people realize. And I do think that it's helpful for even if you're a bootstrapper to think about modes. You're seeing indie hackers and small bootstrappers go to zero. They're businesses go to zero all over the place on Twitter, because either the models themselves, we're gonna talk about people, people think there's like one or two threats. AI is one or two threats to my business. One is that my customers are going to vibe code. My app, the other one is that the model is going to get so good it can do it. We have eight more. This was 10, 10 of them. And this was the shocking part. I think when you brought the AI vectors of attack to me, maybe they were like seven, six or seven or something. And I hadn't thought of half of them. And that was, I think, a big shock. So with that said, we're going to talk about those vectors of attack right now. Then we'll talk about modes and our best assessment of the modes that still work. Are we 100% accurate on all of this? I don't think so. But are we relatively close? Are we directionally correct? I do think so. And I haven't seen anyone else talking about these in this capacity. And so this I think is something as a bootstrapper that I would be thinking about. I guess we'll say this a lot. I'd be thinking about how I can implement one or more of these modes. Because even if AI hasn't had a big impact on you yet, I do think it's coming. I think that's really that one of the big unknowns is like, I think it's undoubtedly true that a lot of these things are is coming. The question is like, how quickly does it spread through the economy and into your niche like that? I think will depend on like what your customers are like and all that stuff. Yeah. And I've been saying I think between 10 and 20% of all SaaS companies are either going to zero or basically plateauing declining. That includes big company, small companies. And again, we're already seeing it with very small. You have a tiny little utility online that converts a file from this to that like that's done. You had a little thing that transcribes or that I don't know podcast show notes. For example, used to be like an app. And I mean, it's still an app, but like it's easy enough to implement. But we're also seeing it like my prediction. My baby Ruth calling his shot is that square space. Unless they do something, they're not going to go to zero tomorrow. But they're going to plateau. Like I'm moving all my sites off there on the static sites. And I think a lot of other people are there are now services. I mean, I interviewed Nathan from Blink pages, a tiny C company and that he's starting up a whole service to do that. Yeah, there's no doubt about that. I think fundamentally a lot of these things are changing. And I think like how these companies react is going to have a huge impact. And it's almost like it's the first time in like at least 10 or 15 years that leadership really matters, you know, in companies large and small. Because like for the longest kind of time, like once you get to a certain size and you had a size and you had this retention and you had this, you couldn't really screw it up all that badly. Now you can't like, you know, like you really can. And people will for sure. And unless a company like square space comes up with a whole new kind of AI first approach with static sites that it supports. If it does that, they'll survive. But if it doesn't, I just don't, I don't see how they do it long term. So let's dive into this. There's a lot of, we see it in tiny sea companies, migrant companies saying, AI is amazing. I'm shipping three, four, five times faster. And my customer base, they're not technical. So they are not going to vibe code my product. Then there are folks saying, well, the models will never do what I do because I'm so niche. And this was the thinking I had even three, four months ago, right before we started talking about that. So those are two attacks. There are 10. We're going to break them up into supply side attacks and demand side. And what we mean by that is if you look at economics 101, if you manufacture widgets or you manufacture a SaaS app and the supply of widgets just in the overall market shoots through the roof or the supply of competitive alternatives to your SaaS app shoots through the roof, you have a problem because you're commoditized, your price and goes down, your, just your ability to charge goes down. Similarly, on the demand side, if the demand goes to zero or goes gets cut in half, like that's a big shock, right? So that's what we're going to walk through. We have six supply shocks. The first one is the vibe coding argument, right? Is that your customer builds it? Keep in mind. People say my software is super complicated. It took us 10 years to build. And there's two things to this. Number one, it won't take anyone else 10 years to build it. Usually they also, if you have 100 features that you've built and they only use five, they only need to vibe code five features and they're set. Attack number two is that a competitor, because all your competitors get AI, a competitor reaches parity, right? And this is the thing that I'm seeing and I had a slide in my microcomptalk where I had eight different AI note takers, including, I don't know, it's like granola and otter and six others. If I have them, yep, they're all kind of the same now, because you just get the feature parity real quick. And so otter, I believe launched in 2016 and granola launched in 2024 eight years apart. And they're kind of the same product, because AI just gets you there. So this is something that even if you have a lot of, you know, the old mode of, hey, we have a bunch of code, it isn't. Yeah. And even integrations, you know, that used to be a big deal and all this stuff. Like it's a well documented API, you're integrating with it's like, well, I mean, what is it a day, two days? There's that. And there's also like different, there used to be, you know, this like stratification, it'd be like, oh, there was like the cheap SMB option that just kind of did things and competed on price, but didn't have all the complexity. And then once you, as a, as a client or a customer, once you got large enough, you would graduate into the larger mid-size solutions and then maybe up in an enterprise, the challenge now, if you're the mid-size, then potentially your lower costs, upstart type competitors, all of a sudden have every feature that you have. And so there's really no reason to switch as they, as clients grow. Yeah, it's, it's crazy. And for those listening, there's a lot of information in today's episode. So I'm going to have producer Ron put together a PDF summary of this of all the info. And that will be at microconf.com/motes. This is hundreds of hours of research. And frankly, you and I could probably talk about this for six or eight hours, which we will to tiny sea companies, right? This is the basis. This is like a 101 version of what we're going to dive into with our own investments. So tack number three, on the supply side is that open source gets good enough. So we're seeing more projects launch, we're seeing them move faster. And usually if you're a buyer, you know, open source is free like a puppy for that expression. Well, now non technical buyers can deploy open source, not it's not that much effort. And are any one of these things going to wipe out all sass now, but each of them is another few percentage points, right? That's what we're getting at here. Attack number four is a platform that you're integrated with. Ships your feature or features set for free because they already have the customers and they move five times faster than they used to. And they were never going to build your little niche thing until they decide, Oh, look, here's a spec for how to build it. The spec is your app. Now, I'm going to crawl through. I'm going to screenshot it, right? There was a big public market fallout thing with something related to this pretty much. I think it was service now or one of the big like field service engines. All of a sudden overnight decided like not no longer accepting integration with one of their like long term partners, which was a lead management tool. My read of that is like service now is like, no, we want to own that part too. So we're just going to build it. That's going to be faster and faster. Attack number five is interesting. It's custom software at SaaS-ish prices. So you have a generic SaaS, charge 99 bucks a month, agencies also get AI. And so agencies, we've seen firsthand 30,000 to $50,000 agency projects. What used to be that? They're just like, look, we'll build it for a couple of grand up front and it's completely custom to you and we'll charge you 500 a month, 1000 a month, but it is custom to you and we'll maintain it on that monthly basis. That's actually a great business. If I was getting started, go after once. Yeah, why not? Think about it, right? Like, you know, particularly the mid-size enterprise side of things where maybe you're kicking out, I don't know, you're a business, you're paying 50 grand a year for this software. And here comes some along and says, you know, for 25K a year, you'll get a super custom with an SLA and I'll monitor it and integrate and do everything you need. Yep. The mode of SaaS being complicated to build and host and maintain, it's not going away, but it is being dramatically impacted, right? Exactly. So attack number six is that the model itself, like the LLM, the foundational models basically absorb your product. They kind of turn you into an obsolete wrapper. It's not going to happen to everyone, but we've seen it already. It happens to sound. I mean, Christ, do you remember we had some early investments in Tennessee where someone really smart and gone long and they like fine-tune their own model and it was like perfect for this. It was like state-of-the-art. And then like, I think Chatchee PT released a new version and it was like, oh, out of the box, zero shot, it does a better job. It's like, well, then you're screwed. Yeah, we've seen a handful of companies within our portfolio experience this. So now let's look at demand side and these are the ones there were a couple on the supply side that were surprising to me, but the demand side was the one that I hadn't thought of. So there's four here, right? So attack number seven is will your customer still exist at all? And the idea of, let's say there's an agency out there that used to either translate words text audio or they would transcribe words text audio. So since AI has come out, the number of words translated and transcribed in this world have what would be 10x, 50x, whatever. I transcribe everything now because we can. And it's really good. The agency revenue for those businesses is effectively going to zero. If it's not going to zero, it's going to a 10th of what it used to be because everybody knows we can just use AI to do it. So the activity itself, the volume of transcription and translation has gone up. But if your customers, if you were building software for translation agencies, transcription agencies, you're done. Like there's no market there anymore. - Yeah, I mean, this sort of reminds me of my favorite conspiracy theory that dentistry is basically just a fraud. They fixed it in the 1970s and then they decided, you know what, we can't get all these dentists unemployed so we're not going to distribute the solution. This is something like if you fixed dentistry, if you gave it to people like a pillar or a shot when they're born all of a sudden, like you had perfect breath and no cavities, like if you're selling to dentists then that's a bad, that's a bad industry to be in. And it's sort of similar, right? Like an example, you know, a slightly less 10-poly example is like, okay, AI receptionists are not bad. I know good enough most of the time. To the point where like, actually, like I wish my airline would like offer me two options. One, call into the usual regular call center and two, here's a phone line to the AI voice agent that picks up instantly. And so that's an example too, right? It's like, okay, if you had the world's best SaaS that was selling to call centers and for human receptionists to click around on, okay, that's a tricky business to be in at that point because you're just going to get replaced. The people clicking on your buttons aren't there anymore. They're not going to pay to do that. - Yep, and that leads us into the next attack, which is like, there's two that are very similar, but this one is are your seats going away. Meaning, like Zendesk is a great example of this, right? If you're charging per seat for customer support software, you used to have 6, 10, 15 customer support reps. And AI's handling a big chunk of that. Not handling 100% of it, but are they handling 80%, 90%. And so now you need what, two reps, three reps that are just monitoring the agents. Like, if you charge per seat, you don't go to zero, but you lose 70, 80% of your revenue, right? - This is sort of the problem with like, and HubSpot has other problems too, but HubSpot's similar thing, right? Like, you used to be per seat for human writers to write things and SEO, optimize and all this stuff. And now it's like, it just, no, like, the thing is AI slop. You just generally don't need as many writers. So if you're counting writers and you don't need as many writers or you're counting receptionists, you don't need as many receptionists or customer support agents or whatever, that's a contraction, even with the underlying industries doing well. - And then the ninth one is what you already referenced, which is like if you cater to receptionists and they truly just start going away. It's all agentic, which there's a chance that happens. Really good one. With like phone receptionists, you know, specifically, I guess we'll talk about, 'cause there's a pivot here, that's an attack, you're gonna go to zero, but the way to get around this, of course, is to build the agent yourself and start selling that, right? - On the offensive side, for sure. - Yeah, so like, you're selling $50 a month per seat, software for receptionists right now, do you charge 500 a month or a thousand a month, frankly, build a full-on AI agent and just go all in on that? So we'll talk about that. A little later, the 10th attack is, does the step you manage still matter or does the artifact that you create still matter? - Yeah, the interesting thing here is like, there's a lot of things, once you start to think about it, like there's a lot of things that only exist in this world because that's how humans coordinate work around something. If AI is gonna eat part or all of that process, you don't need those artifacts anymore. So if your things are managing those artifacts, then-- - Yeah, and the example in my talk was, take home coding tests for developers. And it's like that, just can't, that's just not gonna be a thing moving forward, right? Because everyone can use AI to do it. You're not gonna be able to tell. And so there's pair programming. I mean, there's other ways to do this, but if you had a business built on that, that's a time to pivot. So I wanna talk about old modes very briefly. I wanna spend less time than I did in the talk. 'Cause we've already started touching on it, right? It's like as builders, makers, we think things like a unique feature set or like clever, complex code, years of coding, even integrations you already mentioned. High switching costs for some, I think, are gonna go down. I don't know if you've used Manus, but I've had Manus click through some stuff, and it is, but Manus, when it can figure out, you know? And it just figures it out. - Yeah, I mean, browser use, yeah. I don't think the average person doesn't realize how wide an AI can do with a computer. They will soon, but they don't at the moment. - Right, and even like UX familiarity has been a moat, right? I've like, oh, whether you're a whole team of non-technical folks at the, in the office knows how to use this construction management software. You don't wanna retrain 50 people. And it's like, yeah, maybe. But what if agents start doing a good chunk of that work? What if they start doing 40%, 50%, suddenly you don't need to retrain an agent, it'll just figure it out, right? So again, none of these things are, this is where people who say SaaS apocalypse are like, oh, 100%, everything's going (beep) and everything's going to zero. It's like, no, no, no, no. But also, I'm not gonna polyanna and say up. My customers won't vibe code my app, so therefore nothing's gonna happen. Neither of those is true. So let's talk about moats. The eight moats that we hypothesized still work. And as I said, you did this big lit review for weeks, dude, I was like, you had like a snorkel on. I would talk to you and you're like, I'm so knee deep in this (beep) I don't even know. You were talking to me like I was clawed. You're like, dude, my friend clawed, told me. And I'm like, ain't our, he's not a real person. You gotta get out of the building. You gotta go touch some grass, bro. You were, you were deep in it, yeah, go fishing. And came up with a list of, I think it was six moats. And then as I started digging in, I came up with like four or five more and then some of those weren't moats, but we wound up with this list of eight. And I think this is pretty close. I think it's close to reality. When I knew we were on track was when you and I then kept coming up with other ideas of like, ooh, what about, you know, if you have suffer plus hardware and it's like, we missed a moat, there's gonna be nine. And then we came back the next day and we're like, aha, fits into this category. That's when you know like a taxonomy is starting to hold up when it stops changing so often. So let's dive into this first one. This is where buyers know and trust you, all right? So it's not just having a strong brand, but it's you know the buyer and you know how they buy. There's a partner, potentially partner ecosystem where they've built their businesses around recommending you. And hopefully there's a champion inside your account that is fighting for you. Such that a competitor with better software still has to build all of it from scratch. And I wanna point out that these are all on a spectrum. I don't think this is a binary one or zero. Like I just listed four things, right? The brand, you know the buyer, you have partners that recommend you, you have a champion, like if you have four of those things, amazing, if you have three, you're less on the spectrum of the moat, you know, and it's like this lower score. If we were to score it on one to 10 or one to 100, you can imagine that being there. - Yeah, no, I mean, I agree with this for sure. Like the example I think you used from the talk was most QuickBooks, which is a Hey QuickBooks with a passion of a thousand sons. But it's still like, look, this is like what everyone, all the accountants use and that's what they recommend. And so it's not easy to replace them for that reason. Like you could, you could buy code a QuickBooks alternative. I don't, I don't think that's super hard to do, but I'm gonna distribute it. - Right. I mean, accountants, my accountant has used it for years and he would recommend it if I wanted to look at the books. I just don't want to, right? The other example I use was Shopify. They have an amazing partner network, right? Thousands of agencies that if you go to hire an e-com agency to build your cart, the odds are good. They're using Shopify. And so, you know, there's a, we have a question at the end of each of these modes. And it's like, it's a thought experiment of like, to see if you can fit into this mode or implement it or where you stand with it. And it's like, who's built their business or their career around recommending you or your product, right? Let me give that some thought. The second mode is that you're a system of record. And you've been using this term for years 'cause it's, I think it's a popular one, right? And kind of in private equity circles. I've always thought of it as being like the operating system for a business, the underlying OS, the whole business runs on, right? So daily operations are built around you. Approvals, statuses, balances, everything lives in your software. There's rules configured in it and other systems are wired into you. They integrate with you to either pump data in so you can act or to pump data out so they can act. - Yeah, it's like another way to think about it is like, is there some canonical state that's just recorded just in your system? You know, it's not elsewhere. It's not a copy of something that's elsewhere. It's like, okay, the decisions we made or the definitive record of it, like the example we use here is like Jim Desk, which is a tiny C company. And it's like, yeah, like who's running those classes? Like is on a Tuesday at 6am is recorded in the gym, like there and nowhere else. And that's like the canonical state. Like who has paid? Who has paid for membership this month? What is like, who has access to the key fob? That sort of thing is like, yeah, system of record. And yeah, that is remains one of the strongest things. It's like, that's where like the canonical state is stored. - Very hard to try to leave that. The switching costs are still enormous. It's not just pointing Manus at it and exporting data and moving somewhere. It's everything, like what can't you export? What breaks down stream all the integrations you have? What do you have to retrain your human? to use. What did I have to rebuild? You got to reprogram the key fob. You got to ask people to add your credit card in again. Like, are you going to do that? No, you're not going to do it. Number three is that this was one when you first set it. The next two, actually, when you first set them, I'm like, I have, I've never thought of these at all. So these were a surprise. But number three is you're allowed to act. And I had to really think this through. So a tool that recommends a payment, like, hey, you should, here's a CSV. You should go make $20 payments to all these affiliates, right? That's one thing. But then there's a tool that is authorized to just make the payment to actually move the money. Stripe Connect is an example of this. Like, it moves money around very hard to leave. Brembe is a tiny seed company that sends reimbursements to job candidates. Someone, you fly them into your office to do a job interview. You don't want to send them up as a vendor. So you input their Venmo PayPal cash app. And it just sends the money. It actually moves the money. This was not something that I would have thought of as a moat before you brought it up. Yeah. I mean, it's just one of those things like, can yourself do something in the world or is it just an advisor? Is it, if it's just an advisor and the human still has to do something, that's much easier to replace, right? Because then it's like, oh, I still got to look at and review. It's like, okay, but this actually is wired in. It does the thing. It's part of the plumbing, kind of. Yeah, people don't want to pull it out because it's autonomous in essence or it, yeah, it gets worked on. Fourth mode, again, another one that I was like, I hadn't thought of this, is that they don't check your work. Which sort of relates to the prior one, right, for sure. But it's this notion that like, you're allowed to act, you do something, and then like the people who pay it, they trust you enough, they don't check it. And I think like the classical example here is like, which like, you know, like HR software, I use gusto for some things. And they file documents that like, I could apparently go to prison for it for a run for various state agencies that says, this is your worker comps, this and taxes and that, and they send money around and like, do I, do I go in every month, every by week and make sure that the quarterly tax repayment that they're doing to the state of California and the state of America else, no, I'm not. And like, that's, that's a big deal. Like, I have, I have bestowed upon you, they're not only the right to act, but I trusted you as the software, not the screw this up. And to the point where I don't have to check it, if I have to check it every time, then you're easier to replace. Yeah. And so when you vibe code your gusto replacement to save 49 bucks a month, right? I mean, that's the, or like, when, when an indie hacker vibe codes gusto, are you going to do it? No. Or like, they're like, you know, like accounting software and tax software, so similar to it's like, do I really want a vibe coded solution to save 50 bucks a month so that the IRS is going to like, come right up my ass. Do I really want that? No, probably not. Number five is an interesting one. It's some, I have it, I have it worded. I mean, we have different wordings for these all over the place, but the one that I have is someone official already said yes. And this is around regulation licenses, accreditation and certifications because a competitor can clone your code, but if they need to get approval from a board to do something, it can take months, if not, longer. And I want to be clear here that serving regulated customers isn't the same as you yourself needing to hold sock to ISO 2701 approval from the ABA training or the psychological, you know, the APA training board. What's your, what are your thoughts here? I think again, like this is a scale, right? There are some things that are heart are like seem like hard to get, but aren't all that hard to get. And then there are some things that are actually, yeah, like this is a pain in the ass to deal with. And it, it's like, it runs the gamut, right? The fact is, like, if you are an organization that needs some stuff to be hyper compliant because you're dealing with personal health records or medical health records, then, you know, like, if you have some software that's dealing with that and you know that that's hyper compliant and like, you're not going to get into trouble with the, with the, whatever. Am I really going to vibe code like a HIPAA compliant replacement? Like, is that what I'm going to do with my life? Like, I probably not. Like, the hurdle is higher. This is what I mean, like, there are no perfect modes. Like, eventually, like, we're all dust in the universe and they, you know, the heat death of the universe, etc. But fundamentally, like, this is one of those things that are like, yes, I'm not just going to do it. Like, yeah, Tom down the road down the city, he vibe coded a new version of our EHR. And he says, like, we're going to save 500 bucks a month by doing this. Am I going to take that risk, that compliance risk? Probably not. And there's a fair amount like that. Like, actually one of the examples is like, just from the tiny seed portfolios, it's like sign well, which is like a docuSign competitor, eSign competitor, right? And it's like, okay, that seems easy. Like, like, what is eSign? Ah, it's just figuring out where to stick the signature in a PDF and to store it. No, that's like not, that's not an interesting part of that problem at all. It's like, okay, you signed a contract. Now there's a dispute on that contract. Okay, is it valid? Are you going to vibe code an eSign solution so that you're a million dollar contract? Like, is it in question because you vibe coded the eSign solution? Yeah. Another example from tiny seed is sidekick learning, which is ABA training, which I believe is applied behavioral analysis. So it's affiliated with psychology and therapy and stuff. And their courses count towards a credential and CEUs and all that. That would be, I get it, that's niche, but that would be, so nobody's vibe coding that a competitor to that. Yeah, exactly. And it's just like, look, is it super duper hard to have AI generate some training material? No. But like, do you really want to find out that you're, you know, you're, I don't know if this is specifically the case, but like, oh, it turns out that the 200 nurses that use your software aren't actually, you know, legally registered nurses anymore because you didn't do exactly the right amount of, you know, compliant training or whatever. Like, that's just not, not something you want to deal with most of the time. Yeah. And mode number six is, I call this everyone's already in it or who has to move. Really, it's network effect. It's like, let's say if you have, if you have software where one person needs to start using a different app, then they can switch in an afternoon. If you have to move a whole team, everyone has to agree, migrate together. But if it's like multiple teams, a whole company, or a whole community, and I know what example you're going to bring up. So I'm going to pass it to you. Like, it is a mess. Slack is the perfect example of this, right? None of us really likes like that much. It's getting kind of bloated, kind of painless, kind of expensive, but we just keep coming back. What's your other canonical example of this? School apps, man. Like, if you have kids and, and anywhere in the Western world, you've been tortured with this like every start of the school year, it's like, hey, get on. The parents view and grade Tom and this. And it's like, you're like, what the hell? Like this like four different logins. It's like, what is this complete dog. And you have no doubt, like, you have to do it. And I'm like, and you log in, you're like, look, my kid could like vibe code a better solution than this in a, in a couple of days. Why is this also bad? And the answer is because the school doesn't want to move a thousand students and 4,000 parents or whoever many, they just don't want to do it. It's just, it's install base. It's like it's there. It's like they're just not, they're just like moving software means telling an entire community that, hey, you know, all that stuff we told you about, like, we've been through the sign up. Now you have to go sign up for this other thing and do this whole other thing again. And that's just not, that's just not going to work. You know, so it doesn't, almost doesn't matter. Like it just, the threshold for functionality is so low. It just has to be barely better than dog. And it still gets paid. Yeah, that's what network effects get you. Moat number seven and eight is that every customer makes you smarter. So this is where Stripe Radar is an example, right? Every merchant on Stripe is protected by the fraud patterns seen at every other merchant. So no single merchant could ever build Stripe Radar. There's a ton of value in the collaborative in the community, you know, and the data Stripe has. No one else has. So you can't just come in and replicate this. Yeah. And like, you know, the, the sum of the stuff, this is sort of related, some of the stuff we'll talk about on the sort of offensive side, like, what are the opportunities here? But fundamentally, like, yeah, there's, there are certain things that like, if you have software that learns from actual use of your customer and gets better, yeah, in general, like, it's almost a similar issue to a network effect, you know, like some, some networks are like, like, Twitter is better, however, even more people are on it. And this is kind of the same. It's like, you have more customers and because of that, you're learning more premium customers than your product improves, like, ideally automatic. And someone who just goes and codes this scratch doesn't get that. And if you leave your product, you lose that value, right? You lose the learning. And so, yeah, as I said at the beginning, these are defensive modes. These are how to defend. And at the end, we'll get to going on offense, which was, was your talk in a sense. I guess to put a finer point on the learning loop aspect of it, I had that it only counts if it clears three gates. Every customer makes it better for every other customer. Customers can take their data, but not what you learned from it. And you keep learning, your app keeps learning because last month's data isn't sufficient, right? If it stagnates and you're not constantly improving, that doesn't matter. So I started this agreement a little with this. I think there are still, you know, it doesn't have to necessarily translate across every single customer all the time. Like, it can be like, look, in the best case, like, you can, like, just having if you are a single person that over time. it learns about your habits. It's going to get better. That's good for you. Even better is if it goes a class of team or an organization. If a whole organization comes in, it becomes better. It fits that organization better. It learns inside the organization. It doesn't have to transfer across organizations, but that's the idea. The idea is more. And another thing here too, and we'll get into more details on this on the offense, it needs to be something that's not easily stealable or not easily exportable, easily learnable, easily simulatable. It has to be things that are like you're observing things in the real world on a real-world clock as opposed to something that you can just generate a crap ton of synthetic data and then go out and just very quickly learn what you need to learn. All right, eighth final note is it's hard to keep running, and I was referencing that Peter Levels had a tweet where he's basically like, "I'm vibe-coding everything, all my apps. I'm not going to pay anybody else, I'm self-hosting everything." And he's obviously one extreme. I'm not most of us aren't going to do that, but it's just an example. But he said the only thing he's paying for now are cloud flare and some bare metal hosting somewhere. Maybe it's VPS's, I don't even know, and scraping B. And it's like, well, why scraping B? And he said, he's been repeatedly having AI try to build scraping B, and it's like, what's not the code? It's all the proxies and the capture and all this other stuff. And AI's just like, we can't build this for you. And so it's that thing of like web scraping isn't hard to build. But to keep that working, it's brutal. There's a big, big difference between like, I'm going to, I have a single page to scrape for my own purposes, and I can just hack that sh*t together, and I'm just going to scrape it and lo and behold, I'm able to get it through to like, yeah, I need to like, overtime manage proxies and keep track of that and like, avoid the anti-scraping technology and like all this stuff. And fundamentally what's going on here is like, there are certain things that it just does not make sense for a single client to do themselves. The cost of it is such that you need it amortizing across a number of customers. And like, you'll be familiar with this from your drip days. It's like, look, send in some, look at this face, send in some goddamn email, getting that sh*t delivered, not calling spam and like, not having really angry customers. Like, staying ahead of, and staying ahead of the folks that want to use your services for spam. And like, all this sh*t's like, that does not make sense for a single, you know, you have to be a pretty sizable company for it to make sense to have, you know what, I'm going to hire 25 dudes and they're going to be looking after my email infrastructure. And that is the sort of similar deal to scraping be. And honestly, a lot of businesses are like this. It's like, I figure out a world class at this, the cost of doing it, there's just a minimum hurdle of like, complexity of running it, figuring it out, maintaining it, SLAs, whatever. And you just can't compress it and that's just the way it is. Yeah. And that cloud flares another example of this. It's, is it, is it hard to build this to code it or is it hard to like, keep it running? Like, is it, there's a DevOps load, there's a, I mean, the other spam one you bring up is just, we had a full-time person plus just doing this full-time. Yeah. And someone has to like, sh*t goes down. Oh, you're like under spam attack or like, some of your spammers in Nigeria started sending like 10,000 emails out of your one good IP that your number one customer uses. You got to wake some dude up at 2 a.m. and it's like, look, like, do you want to do that? Or you want to pay $249 a month? That's, that's why this is a moat. So those are the eight. And I won't re-kept them here because folks can go to microconf.com/motes and download a list of them. But as we talked about, and there's, you know, we're going to create curriculum like a whole playbook for tiny sea companies around this about how to evaluate themselves where they do. They have any of these modes, maybe scoring potentially and the idea of how, what modes are you closest to? If you don't have any or if you have three and you want to get four, like, how do you do that? So there's a lot more to this conversation than just this identifying of them. But that was us going on defense, right? How do we, how do we not get sideswiped by AI? You've talked a lot about going on offense. Why don't you take us from there? So going on the offense side, like, basically, there's sort of three moves effectively, I think, in this world. And so the obvious one is the one that people are kind of already doing, which is like, keep improving the business. Use AI to make, you know, to go out and just build more software faster. It's sort of similar to like kind of what you've always done. It's like, okay, for me, like, maybe you identify that you have a bunch of these modes, actually, your businesses in great shape, no problem. And actually, like, it's just an opportunity. And you just go out and like, you just have to build more features. And AI gives you that capability. And great. So that, to me, is almost like, you know, table stakes as it relates to what you could do here on the offensive side. And I think everyone can do that. The second part is, you know, one of the, one of the things we identified is this notion that people will have more agents. There'll be more agentic things happening. And so like, agents are going to be interacting with your software most likely and somewhere shape or form. And there are kind of two stances to do here. One is to try to lock them out, which is sort of the Amazon approach, I guess. You saw that this week with the Muse, the Facebook bot, like, you can't order things from Amazon using it. That's one approach. Another approach is basically say, like, look, I want to make it super easy, you know, for the agents to come through and offer an MCP and maybe gate those people and like, get paid based on based on that. And that can be, that can be a really good solution, actually, for a lot of things. Because, you know, AI doesn't seem to always think it should build everything from scratch. You know, the way that I think about this is like, AI makes a decision on complexity. There's like, it might decide, you know, what, like, the best solution here is just to use this third party thing that will just get us what we need. Let's just do that. And actually, like, I thought about this when we're in Iceland, it's like, it's almost like there's two different kinds of SEO now. There's like the SEO to get your human customers to use you. But then there's also the SEO that's like targeting agents and figuring out, like, how do you get the agent to pick my solution over all the other solutions, which might be actually be a different question than the SEO for humans. And so if you have something here where it's like, it's clear to you that people are going to be using agents to do that, making that ideal and actually charging for it. And like, you know, doing that, agentic SEO or whatever you want to call it, like, that is definitely a play. So those two things, I think are sort of like semi-obvious. I think most people have done it. The really the main thing, I think, is you want to kind of become the worker. Like, you need to start doing the work. If there was one thing I could do in terms of portfolio or just a wider community, it's sort of a kick up the ass in the sense that, like, look, an awful lot of SaaS, BDB SaaS is, I have a tool, some kind of a tool. I'm selling that to a, my customer, a business owner of some kind, probably. And that business, they basically hire somebody or they have an employee to click on that on your tool in order to have some outcome happen. That's a lot of SaaS. And I'm here to tell you that, like, if that's the case, somebody is going to come along and they're going to say, okay, you don't have to click on this thing anymore, just like, we're just going to sell you the outcome. Like, we're just going to give you the outcome that you actually want. And my view here effectively is like, if you think that's likely to happen, maybe not immediately, but like, in the next year or two, you should do that. You should be the one to do that. And the point I made in Iceland is like, it's not, and like, I actually see this also, like, if you just look at, like, launches for a wide combinator, like so many of them, like, 40-50% of those are this. They're basically agentic SaaS. I think it's sometimes called. There's notion that, like, yeah, you sell the outcome. Like, don't sell the tool, sell the outcome. But on the flip side, if you are a bootstrapper with existing customers and things like that, and you have industry know-how and connections and customers and all this stuff, it's not obvious to me that you are in a better shape being a Silicon Valley startup with a 10 million open AI token budget or whatever, versus if you're listening to this most likely, you already, you have a SaaS in the industry and it's already doing something. Like, you might be in a much better position because the hard stuff to get is like, trust from the customers. And like, there's a lot of noise around this stuff. If you come along as the trusted vendor who used to sell the tool and now you're like, here's like partially or fully the outcome and you can charge 10 times that. Like, that might be a much better solution. So I think for most people, 80%, I would say, that is the play. That is your opportunity. And I think like, there's a lot of like angst about SaaS at the moment. You know, there's almost this sense that like, oh, like Chris, there we go. You know, it's like, why can't I sell my SaaS like five years ago and then you do whatever. And I'm like, yes, but also like, the opportunity here is much larger than it's ever been. Just because what you're doing is like, much more aggressively than before, you have an ability to go in up into the labor budgets. And the labor budgets are much larger than the software budgets. If you look at your average business in this world, they spend a lot more money on salaries than they do on software. And if you can access some of that, then you're essentially in a great position. And one of my like favorite like, books is called 10x is easier than 2x and I think that's the mindset to have on this. It's like yeah like okay you can build it a little faster so you can build some things so you can charge some more for your software. That's great and I'm not saying you shouldn't do that but it's always worth taking a step back and being like all right I'm charging $75 a month. What do I have to do so that that same customer will give me $750 a month. That as an exercise. Like thinking about like okay what are they even using my tool for. Like what other things does this person that's clicks on this. What other like what are adjacent workflows that I can go after and just offer as a solution so that I don't have to use as much labor in order to get the outcome that they want. That is the opportunity and I guess I'm a little surprised that the negativity in some cases about you know AI on this and I'm like yeah I get it like it's there are certain ways in which you know we're starting to hear people say I forget it I buy quoted a solution or I just use cloud for this but equally I think there's an opportunity here to 10x a lot of businesses that prior to AI may have really struggled to grow two or three x in the same time frame. So the opportunity there has become the worker and I can already imagine all the objections that I I would hear from this from bootstrappers and such. To me it if I were a founder still if I was let's say running drip I'd be like god that sounds like I'm that sounds like a lot of work I'm already working full-time just trying to keep this business afloat now you're telling me to like spin up this whole thing with a like a worker that I then have to monitor that we know is not going to be 100% AI is going to be 70 or 80% so then how do I put the safeguards or whatever else in place and how I sell that at 10x and I need a sales force where I was self-served before or what I mean I can come up with a litany of reasons to not do this or at least objections right but I think what you're saying is well do you want to survive right that is it I mean I think I think that's it and and like one of the things that it allows you to do is that if you're going to do the worker the stuff then then kind of naturally this mode of like this sort of learning loop that we talked about this this like notion that you're observing how things are going and based on the actions that are happening and the actions that got taken by your AI worker thing you're able to improve over time that mode is always like if I'm thinking about modes that I think are likely just to to keep forever like ahead of the capability that is the one mode that I think over time will will just get stronger stronger this notion that like it acts on the world your AI worker does some things it resolves a ticket it gets a refund it does whatever it does and then it learns from that and because of that it gets better and that then transfers across into different you know across your customer base so that you as a as a business you're learning more and more and more and getting better and better and better at this thing that is like the only mode that I can see that's likely to stay ahead of sort of this expanding capability frontier that the LLM's have so yes it's a lot of work but fundamentally I think like your alternatives are good but not great right like so like person who thinks I'm just going to stick with this and like I'm just going to like use what I call the old school learning mode which was always like you always kind of there was always a learning mode right and it's like it's this it's you observe your customers you see what they need or your your prospects or whatever and then they tell you what they want and then you build that feature and then you ship it and then you see if they like it and then like that sort of thing the problem with that and that's sort of the classical manual loop as they were but the problem with that is like the outcome there is visual UX like it's code that is you know maybe take very long time in order to get right in terms of designing and learning from your customer but once it's built it's like five minutes of screenshots from Claude and it just replicates everything and it just pushes it out there versus like a learning group where like it's knowledge that exists now within your organization that people don't just take with them when they leave or you can't just screenshot in order to copy that is the one loop that I'm that the one mode that I'm pretty confident is going to stay ahead for the years to come like that the more I think the most valuable businesses will be like that there'll be things that you have learned over time within the real world from actual customers and like you add a more capable AI to that or whatever yeah everything gets even better but it doesn't negate the fact that you have learned all this stuff that an LM just can't learn very easily yeah and I want to call out to folks listening like the first offensive move is to become the worker agentic sass and related but a second and different offensive move is to implement a learning loop like you've said right not the old school learning loop but the actual like you have data and figure out how that data helps and predicts for all or most of your customers that's another thing that is of incredible it's not just a mode but as you have it it should like draw in more customers and build momentum the thing that I struggle with with it because I totally agree with you obviously thing I struggle with is I know that if let's say we built that again if I was still running drip and we built an amazing learning loop to keep spammers out and to keep everything deliverable there would be a bunch of schmos who would come along and vibe code stuff and say but our deliverability is great it's 99.9% and they'd just be lying and so building this is not enough it's like you have to you have to kind of build I think build your brand around that for people to believe it like just saying it on a marketing site I don't think that's enough and that sucks let's sex it's not fair it's not your fault it's your responsibility to get this done right it's it's that type of thing where I wish that it was easier I wish you could just build a learning loop and be like look our tech is the best and everyone believes you right or our learning is the best our engine is the best you know that people are going to argue over this and try to discredit you and say that they're better and all that stuff so just be prepared for that fight doesn't mean it's not worth it cool man so we covered a lot of ground today about the vectors of attack the 10 10 vectors of attack the supplied and demand shock eight modes that we think will survive in the age of AI as well as how to go on offense as a reminder you can go to microconf.com/motes if they want to download a summary and potentially the slides if producer Ron gives me the the thumbs up to give away my own slides you know you know what I love dude I love going to microconf and people come up to me and they're like hey Rob so where should we be at what time and I'm like you know what I go where producer Sonya whatever something my calendar is where I'm showing up so please don't ask me information about logistics because I just genuinely don't know and yeah Sonya is my boss at the events it's similarly with with Ron I'm always like so what are we giving away around because I don't actually know what with the giveaway is so all that said if folks want to keep up with you on the internet you are posting daily about the San Francisco Giants fishing and sometimes even SAS M&A and MOTES you are in our volset EINAR VOLLECT on xTwitter and of course discretion capital.com so you run the cell side M&A advisory for SAS companies doing between two and 20 million and obviously if you're even if you're doing seven figures and your SAS company and you're thinking about selling here in the next whatever six 12 18 months A&R is thinking a lot about this stuff about how to make SAS valuable and how to make it still a desirable asset to acquire we are also thinking separately like I'm thinking a lot from the founder perspective and operating businesses in tiny seed you are too but you're also then thinking about it from private equity and how to market these companies to make sure that you still get good multiples. Yeah I think it's fundamentally changed like you know the positioning as fundamentally changed in that like being knowledgeable and like really understanding how to position basically how to argue that like look AI isn't going to kill this business next year that that has become a fundamental part of a cell side M&A work for sure right I just didn't exist two years ago we were such young spring chickens why didn't I we should have just sold it our covid was bad here we are my sweet sweet summer child yeah totally it's like woof so we don't I mean this is the thing like you and I it's almost like day to day week to week we'll talk and be like how you feeling today and it's like man I think I think we're good I think things are going to be great like we have modes and we're trying to educate him and then the next week it's like I don't know this is really maybe I'll just go fishing for the rest of my life you know I'll become a fishing guy yeah and what's funny too is like you can see how my thinking has evolved certainly on this podcast like I recorded an episode two three months ago about like SAS in the age of AI I think was the was called and that's still generally valid like I still say I think 10 to 20% of SAS companies are going to be impacted what I didn't have was these modes at all I had three or four vectors of attack I think is what I listed and those were the ones that I had come up with and and then with new information we formed new opinions right because we're not politicians that are accused of flip-flopping when we actually change our opinions on things right so you're watching it happen in real time so in our thanks again for joining me today thanks for having me thanks again to anar for joining me on this epic adventure through SAS defensibility and offensibility which isn't a word in the age of AI again if you you want to download all the goodies that come with this episode, microcom. Thank you for listening, this week and every week. This is Rob Walling, signing off from episode 855.