Don't start SaaS right now — with Mike Hill

There are just un-leevable numbers of competitors in each vertical now. And if you're not in the top three consideration, you're in trouble. I don't understand how SAS can die, right? Any business where the underlying expenses are ongoing, the revenue needs to be ongoing, right, to match the expenses. That's just basic economics. So where do you call that SAS or something else, whatever? The one piece of advice I do give founders, whenever they contact me, is get a technical co-founder. All right, but hang on, why do you need a technical factor? Just trust me, get a technical co-founder. We're not there yet. How do you come about idea? How do you, you know, then launch it and etc, etc. And my advice to them now, which is different to a year ago, is don't. Hey, Mike. Welcome to Sounds Good. How are you doing? Why are we still being the building sauce? Hey, Anna, how are you going? Well, a lot's happened since we last chatted. I know. Well, why are we still. That's a big question. I don't know any different at the moment. Why are we still building SAS? That has so many sort of background connotations to it, which had a lot of conversations about SAS Pocclips. Is it still a viable model? A lot of talk about vibe coding and not just vibe coding, but AI making it easier to build a SAS. Or even not build a SAS. Builds just some sort of direct connection into the data where you don't even need a SAS. You know, it's just an MCP into the data and it's just a chat interface. I think the SAS Pocclips was happening long before AI. And it has nothing to do with AI. And I'd SAS Pocclips is such a weird term. I think that there was a proliferation of businesses out there, copycat businesses that started to happen long before AI came about. And we're just starting to see that happen. And I don't think it's anything to do with the ability to create a business using AI. I think it's just that more and more and more people have caught up to the facts that you can build a microsis business. You can shave off 5% of the addressable market and have a very profitable bootstrap business. That started to happen five years ago. Well before that day that chat GPT launched and sort of amazed us all. So no, I don't think it is SAS a little bit in trouble. Yes, I think SAS is a bit in trouble. I think the overfunded SAS businesses are in trouble massively. But I don't think it's because of AI. I think it's because of the proliferation of businesses out there. AI has from what I can see made good people better. And bad people worse. Not bad people. Junior people worse. A climate mask. Yeah, it's good for senior senior and junior is probably a better term. So AI has made senior people better and junior people worse, I believe. We've seen it in so many aspects of our businesses where if you don't know what you're asking, the AI, the AI, whichever AI you're using, and you don't know how to tell when it's giving you the wrong answer, it'll make your work much, much worse. But if you know how to work with it, you know how to brief it properly, it's genuinely making our senior people, especially our senior developers way, way, way more productive. But it's a problem if you haven't got to that, that seniority, not seniority is again the wrong word. It's more that that's level of capability where you understand what is right and wrong, what you need to get. It's causing some real issues where people are just going, yeah, okay, I'll take that for granted and moving along. Yeah, we were like, we were talking before we started recording and that's exactly what I thought, because you said like your whole growth strategy from day one is always content, right? It's content let grows and you know, I also just, I'm just, I'm deeply supported because I think, I mean, content just lives out there, right? And then people just find you and it compounds, it's, it's a great lever. But you said you used to always outsource it, right? And before Claude AI, you didn't know much, okay, how do people actually do the magic, Covescio, how do people do that? I do. Yeah, and then you said in the last three months with Claude, you, you're still outsourcing it, but you know, so much more. And you know, I thought, yeah, okay, it makes sense. But then again, you have this massive knowledge compound over, you know, 15, 20 years. So you know, what's what you've seen good content and bad content. So when people are just thinking, watching TikTok and thinking about launching their first sauce with the help of Claude, chat GPT, whatever, like where to take that knowledge and where to even have an opportunity, a chance to win with content. Is there a time? Is that, can we game the system? So, so I've done a couple of interviews over the last year or so and had a lot of people contact me to say, I want to get into SaaS, looking for an idea, how do you come about coming up with an idea? How do you, you know, then launch it, etc, etc. And my advice to them now, which is different to a year ago, is don't, don't just jump in. Hey, here's a quick pause for a shout out to SaaS group who sponsored the shout. Full disclosure, I worked there for three years before going independent, so I know exactly what I'm recommending, not just getting paid for it. There are B2B SaaS acquired with around 20 portfolio companies and they offer some of the most flexible deals on the market. Founders who exit to them, either staying, keep running things, or hand over to someone the team already trust. The diligence process is fast and if that finance and figured out before they make an offer, if you are somewhere between $1 and $10 million in ARR and you started thinking about what an exit might look like, or you just want to know what your business is worth without committing to anything, they're worth a conversation. SaaS group, find the link in the show notes. And it's got nothing to do with AI, it's just because so many people are jumping in and trying to solve the same problem, whether it's, I don't know, a CRM for horse riders, who knows, like whatever it is, it could be super, super niche. More and more people are getting into tech and starting their own side hustle, that it is becoming harder and harder and harder to find ground where somebody, where there isn't a really, really good tool in the market. So six or seven years ago, it would be quite easy to find a vertical, whether one or maybe two solutions in the market, and you could come in and be a third or fourth and make really, really good money. In most verticals that we find now, they're at 10, 20, 30 solutions, and it's really, really, really hard to find new ground. Not impossible, but it is harder. I'm working harder now. I'm better at my job than I've ever been because of AI, but I'm working harder than I ever have. And I don't like, well, I totally, I don't like working hard. Can't stand it. And it's becoming harder and harder to find that point of difference because of the competition. And it's a bit of a nightmare. You look at people are now using chat TVT or or applaud instead of search, and that is killing bootstrappers because you used to be able to go on to Google and say, okay, what's the best tool? What's the best digital signage solution? And you go on to Google and you find three or four articles or list of pulls or whatever it was that had 20 or 30 examples in each of those articles. Immediately, I had 30 tools to start. So you go and do that on chat TVT or applaud and they'll give you three and they'll be the biggest three. The guys that have been around forever. And it stops you right there from your research, whereas Google lends itself to you going further and further and further and further and going down a rabbit hole. There is no rabbit hole to go down in research phase when you're doing researching toward chat TVT. Does that make sense? So all of the really small bootstrappers businesses are getting stomped on. And so we're going to have the work harder and harder and harder for those first few customers. And so I think it's just also the fact that with Claude or chat TVT, we're also allowed Two, two basically two. talk to it and say, "Hey, okay, if I am this and this and this and this and that, and then I need all of these problems solved, then it's so sure, it's so confident giving you those three. That you were like, "Okay, it knows better. Who am I to ask?" So it just goes in. Yeah, and you don't go any further. And so people don't go to Reddit. People are not going to Reddit as my people are not going to wherever it was that they used to be finding all these niche products or either cheaper or better than the incumbents. It's getting harder to find those products. There are more of them, but they're not being recommended as much. And Google doesn't recommend listicles as much anymore. They've changed their algorithm two or three times over the last big way over the last three or four months, which I don't think has helped the smaller companies at all. So it is way harder. We are working harder for every single sign-up than we were two years ago, much, much harder. And it's kind of good for us because we've got through the stratosphere as it were and we've got customers on all of our businesses. And yes, we are bootstrapped, but we've got through that no customer just passing out phase, which is harder now than it's ever, ever been. It's been easier and easier than ever to build the tool and it's harder than ever to get a first customer. Right. And yet you're still playing around for that, right? It's like, where you go on LinkedIn, everyone is trying to rebuild a helps bottom of a ball or what not, right? Everything's possible. We ditched this. We ditched that. And you went on and you launched Smile. And to this, you don't know it's Myles. It's a super funky. Like I loved interacting with it. A super funky, tiny product that sits on Slack and allows you to send very funny kind of door, kimmelennial style postcards. There's cards, things like that. It's kind of, it's the antidote to those, those really crappy birthday cards. It's just got a little bit more personality, a little bit, you know, a little bit more fun, basically. Yeah, look, we started that. It's not a pain. We're not all dying. Exactly. It's totally different to all of our other businesses. It's just, it's not, you know, if Smile died, our customers, it's not going to change their business, but it's fun. Right. And that sort of, it's totally different to the other businesses that we've got, which is why I love it so much because there's, it's not essential, but it genuinely does make people laugh. And I've seen people when they get it and that, when they get one of the cards and I've seen how much they love it. And that's kind of cool. It sort of bridges a gap between an entertainment sort of B2C products and a B2B product. We've been building that one for bloody ages, by the way. We didn't just go out and that one, that one, that is looking at ages. But yeah, that's, that was a greenfield kind of, not a greenfield idea. That was an idea that didn't have many competitors. When we first started looking at it, there are lots and lots of competitors now. I mean, you try to rank for birthday cards on Google at night, mate. But, and so I don't, I don't think we're even going to try. But that has a nice mechanic built into it in that it's one to many. Right. One card goes out to 30 people or 100 people or whatever it is. And so the product has that biomechanic built in. So we're not relying as much on the O as some of the other businesses. And yeah, look, the other thing, we just keep, we'll just, until I can't be bothered anymore, we'll just keep building new products and, and testing the market. And a lot of people just give up as well when they don't get within the first three months, they don't get accustomed, they give up. And, you know, that's great for us because we just keep going. We've never given up on any of our businesses and we just keep going, keep going, keep going. It's almost, it's unwavering belief that it will work. This belligerent kind of personality that we all seem to have. And eventually it works. Or at least has done so far, whether it will continue or work, I have no idea. But yeah, so it's sort of, but, but, but, but, as I said, when people ask me, should I get into SAS, my, my genuine answer is no, at the moment, because there is so much noise. Now, some people are just bloody good at what they do, right? And they're always going to win. And so there are some people that come to me with an idea and I'm like, wow, you know, just do it. You don't need to talk into me, just get on with it. You know, they've got a team where they bloody amazing designers, incredible engineers. And those people are always going to make it because they're the top 1%. So, but yes, as I said, it's changed, it's changed massively in the last year. It does kind of worry me. I'm getting older now. And, you know, well, I, it wasn't until last year and knew what the term, what an MCP was. I mean, I'm embarrassing that like kids growing up these days just seem to know what everything is well before I've even heard it. And so, you know, tech is still a young person's game. The way I would build a product is very, very, very different to how a 20 year old is going to build a product that they will use different tools. They'll use a different process. Our process is changed massively, right? We used to have a very traditional kind of going to weigh and do your research, you go and speak to people, you build an MVP, make it look good. You get it out in front of five or six people, etc, etc. And then you go back and forth, back and forth. Nowadays, we just build it. Just build it, put it in front of people, if they use it, keep it, if they don't take it out, done. And design has become, it's no less important. It's just moved. You know, designing something has moved to the back of the queue, rather than the front of the queue, UX and coding something has moved into the same bucket. You can generally do that in one go, build something and then design it last. We don't always do that, but it's just a totally different way. That's only really started with us in the last year or so. And it's much more important as well. But you get stuff in front of people and you can test three or four different ways of doing the UI and the UX really, really quickly before you really craft something up. Yeah. I think that's the best thing about AI. It's not that you can whip up a new HubSpot. It's not that you can potentially go after this huge idea. But, you know, I asked if there is space for tools like Smile, but what I genuinely honestly think is yes. Because we've been in this data automation, this kind of, you know, all this lingo on LinkedIn for wing to long. And I feel like, you know, this is the time to unclench. To unclench and have some fun. Yeah. Who actually built some cool products that you're passionate about with great design, you know, kind of, you know, make internet fun again. One of the things I've noticed is that I know that I have a limit, right, in terms of the number of people that I can manage. It's a, it's a, just mentally, the number of people that I can keep in my head that I go, okay, that person needs that person needs that. I need that from that person. That's three. It took me a long time to figure it out. I can only manage three people. Some people can manage 10 people. Some people can manage five people. Everyone's got a different number. I can manage three people. It's the same with functions that you'd get AI to work for you in a business, right? So I can only manage, I still need to outsource SEO, tech or get someone else to do it as it were. Because if I get AI to do those functions, I still need to keep in my head that I am managing the AI to do those functions. And that becomes really, really tough to do if you go beyond three. So it's a little bit like a junior developer suddenly becoming a senior developer because it has to manage all these agents or manage all these workflows. It doesn't work. And so there is still a cognitive limitation on people, the amount of work that they can manage in their head, no matter how much, how many credits they've got. I'm getting a little bit off-piece to it. But so it hasn't really, you know, I see all these people on LinkedIn saying, "Oh, I manage, I now manage 38 agents." Or whatever. Can you imagine managing 38 agents? When they're all coming back to you saying, "I've done the work in three seconds." You say, "Okay, just hold on, I can't check it." Check it, it's crazy. Right, and you've got 38 pieces of work and people go, "Oh, I might, you don't have to check it." You do have to check it. You know, unless you want to be so incredibly average, you need to check everything. Otherwise, it will be just a lot. So the bottleneck is still people, right? There is still, I haven't, some people that keep telling me how they're literally pushing out 10 times the features, 10 times the work that they've done. Really, I'm skeptical. Is it 10 times the amount of work? Or is it 10 times as good as well? Is that work as good as the work you were doing before? I guess it isn't. To this point, like honestly, talking to so many founders is a blessing and a curse. But I know about amazing things that are going on, but I also know that some of the things out there on LinkedIn, Twitter, whatever, is complete crap because just without, of course, naming any names, but it was just shocking to me very lately to realize two companies I knew just completely collapsed because of that, because the founder basically fired everybody and said, "You know what? I'm managing this with 138 agents." And then at one point, he collapsed because he could just not keep up with all that work, like exactly what you're saying, because it was more, it was way more and way more and way more. And because with AI, you can do more. Like you're just sitting there, "Okay, good thing." Just kind of playing this slot machine. Back while it builds up and builds up and builds up. It's crazy. And it's not just backlog and development, but it's everything. And eventually everything they were doing was, again, to your point, so mediocre and going so slowly into the wrong places that it just completely killed the companies. And it was so wild for Mutant to understand. I mean, everything is a bet, right? The girl with content, it doesn't work. Well, the product dies. We do product let completely 100% whenever it has sales, it doesn't work. So everything is a bet. But this was such a massive one affecting people in the company you got laid off. And I thought, "When is this craze going to just die down a little bit?" Do you see the all birds that the shoe company decided they're an AI company? Do you see that one? Yes. Oh, God. Yes. And they stopped. They stopped. They stopped. My stomach. What? Oh, yeah. I thought that was an April fools gag, but clearly not. But doubtful. Yeah, look, it's sort of, it's a weird, weird one. I try not to like talking about AI because it makes me, because I'm a little bit of a skeptic in how it's going to change everything. I feel like I'm just, you know, an old guy, chown at the moon. And they're with you. It's changed everything I do. And at the same time, it hasn't changed anything. The fundamentals are still important. It has, you know, let's say this again, it has just allowed us to fail a lot quicker, right, to test a lot of other things. It hasn't allowed, it just allowed us to explore 10 routes, 20 routes to find that one route. But we're not necessarily getting to that one route any faster. It's just allowed us to make sure that that one route got one feature of that one article, whatever it is, it's better because we messed up 20 times. And that's where it is, it is really cool. But yes, it can be, it can be very frustrating when you're working with people who just rely on AI to do absolutely everything and, and, and don't check their work. It can be not frustrating, it can be destructive. We've had two incidents in two of our companies recently where AI has hurt us really, really badly because we just let it, because we didn't check the work enough. Never again, never again. Yeah, it's like sometimes it's frustrating. Like I don't want to call myself a founder or anything, even though they, I just create a couple of things. And I have a friend who helps me. He's a pretty brilliant developer, super serious about security and everything. And every time I show him anything, and I say, Hey, you know, it's a yes, but I did, I did check it, you know, I did, I did give it to another AI and then another AI. And I think it's all clean. And every time he looks at it, it's like, no, that's shit. That's shit. And I'm like, Oh, God, it's all humbling, but it gives you, you know, it's so great on the circles. Yeah, the one, the one, the one piece of advice I do give founders, whenever they contact me is get a technical co-founder. All right, but hang on, why do you need a technical fact? We're not there yet, right? We are not there. And then people will cite, but this founder, you know, is not a technical person. They managed to say, yeah, you can always find an outlier, but the vast majority of people fail because they haven't got a technical co-founder. I promise you, you will always find out outliers, but one swallow does not make a summer. Right? It is so, so important, especially now, more so than ever before that you have a technical co-founder. Otherwise, you're, you're just working with a maniacal villain technical AI that won't, you know, that lies to you. And everything's put together, held together by rubber bands. It's more important. So last big scandal on Twitter, what was the company? Have you seen it where AI completely deleted their entire code base? Or some pocket book or some pocket something? Something. Yeah. And just it was a combination of a developer letting cursor and Claude run riot on their production code and deleted terabytes of customer data without asking the person has terrifying. I shouldn't laugh, that is absolutely horrific. But look, again, those are, again, those are outlier cases, right? That generally doesn't have. For the most part, AI is really good at working with senior developers to come up with solutions, especially solutions or especially solutions that have been done before. Right? If you need to build a dropdown, if you need to build a calendar, if you need to build a website or anything, right? There's been done billions of times before. That's why AI is really, really good at helping with those things. And the structural for me. I go from, from pillar to post on this discussion, 30, 50 times a day. And probably if you interviewed me, we have this conversation tomorrow, I'd be saying something different. Yeah, yeah. Absolutely. But it's just so interesting. One year's time and see how I feel then. I'll probably be out of it, Jill. Somebody has rebuilt a spot on Loveable After all. But you know, one more thing that I want to ask you. And I feel like I'm, I've been more of a skeptic publicly than somebody who loves AI. And I do. Don't get me wrong. I think it's a brilliant, brilliant tool. And it's still a lot more magic to me than anything else. Right? But I just, I hate to jump to conclusions at anything. It reminds me of Bitcoin and all kinds of, you know, NFT, whatever bubbles we went through when, you know, 15 minutes after the thing was born, there are experts and people telling you what to do and how to do it. And you're behind. And, ugh. It's just, you know, it's horrible. But I still-- I'm less worried about that because you can muddle your way through that, right? People are going to come out and say their experts on absolutely any new technology that comes out. And they probably are genuinely now real, real, real experts out there in what we're talking about. The bigger worry I have is reliance. The other day, I was sitting at work, and this is such a minor thing. But, you know, if you compounded it out, I was sitting at work, and Claude was down. It was only down for an hour. And-- or maybe longer-- I don't know how long it was down for, because I just went home. I got screwed this. I can't be asked to work the old way. Like, what's the point? So I just came back the next day. Claude's back on. Great. We can work again. It's a little bit like, can you imagine trying to get from-- I don't know where you are at the moment. I'm in Sydney. Can you imagine trying to get from Sydney to Brisbane without Google Maps? And a bit where you just follow the coast? You know what I mean? Get from one place to another without Google Maps these days. It's been nine and possible. I just-- I rely on it. I can barely work out how to get to the beach. And that kind of level-- And you don't even like the beach. Yeah. That kind of level of reliance is terrifying. Absolutely terrifying, because what happens when it's not there? Worst. What happens on what's worse? What happens when it's three times as expensive? What happens when those credits that you're using, the developer that you replaced, end up being $6,000 a month, probably could have just had the developer, which would be much better. And that reliance-- I've hated having reliance on-- in our businesses on external APIs. I mean, sometimes you just have to-- I've hated it because as soon as those APIs go down, your business is exposed. And I've seen a number of businesses who rely on, for example, the Facebook API. They just got kicked off of the API. This was years ago. It kept off the API. I'm not going to name the company, but they were a competitor of ours. And they were doing something wrong with the Facebook API, the meta API. And they got booted off. And they went from being a $10 million ARR business to zero in one year, just because they lost access to the API. Because it was a miscalculation in what they were using. It was a bit of a gray area, turns and conditions. They had everything invested in that one API. And that's kind of what's happening myself included. We're relying up that we have fundamentally changed the way we work, not just internally, but the products that we use rely on these LLMs. And that is terrifying because they could change their cost structure, they could get more and more inaccurate. These have privately held businesses. You never know who's pulling the strings behind the scenes, GROC, not to name anyone. But Chatchy PT and OpenAI as well. So yeah, it's pretty terrifying. I don't have the answer because I'm relying on it now myself. I have drunk the cool aid. What I will say is it's no replacement for humans. Amen. So that's an interesting point. And I think that's one of the last things I would like to ask you here. Because we talked about that as well. But it was actually like I think a red thread just free out every business that you've been building. You've been doing quite a bit of lifetime deals offering a lot of lifetime deals. And somehow now because distribution is a bottleneck, I mean, it is difficult, right? We all build the next grade up on the weekend, but then where do we sell it? And people kind of went back to App Sumo of all places. And my, you know, great for you, right? Like it's a good way to kind of jumpstart your business. But it's AI companies. And I'm wondering if you are selling a $50 lifetime deal. When are you going to 10,000 X, $100,000 X that in token cost over the lifetime of your product? Is it still viable? No, it's not. And that's why so many of these AI, LTD businesses have gone out of business. Or they've had to change their terms and say, AI is no longer part of a lifetime deal plans. I've seen that a lot as well. And look, that part of our strategy-- so part of our strategy, one of the things we do is to get a leg up early on and get a little bit of money through the door is to sell a lifetime deal, which is a single price for an account for life. It could be $50, $200, whatever it is. That use-- again, that used to be really, really good. It used to be a great, great hack. It's not as good as it used to be because every single business in your vertical has launched a lifetime deal. Probably five-- there have been five of 10, 100 CRMs that have launched a lifetime deal. Everyone's got one. They don't need another one. And so the number of buyers out there are decreasing and decreasing and decreasing. As evidence by, I'm reading between the lines here, but absolutely, I used to be 140 people. Then they're back down to 40 people. The LTD market is not as big as it once was. And so it's not as much of a sure thing for founders to go out and do. There's been an increase in people want more and more and more and more for their LTD. So you can no longer launch an MVP early on and say, hey, come and buy my product. And they're like, yeah, but it's missing 50,000 features that this other platform has got. And so you have to build more and more and more and more before you even get to LTD phase. And so that's getting bigger. And the buyers market, the number of buyers who want your LTD is shrinking because they've already got a lifetime deal with your competitor that launched a year ago. So as I was saying, going back to what's killing SaaS. So I think it's AI. I think it's just the fact that there are so many competitors. And that happened before AI launched. There are just unleevable numbers of competitors in each vertical now. Because I don't care who you are. If you're going out there and researching, for example, signage digital signage businesses, right? We've got a digital signage business. If we're not in the top three, we're not going to get bought. Nobody's going out there and researching 10. It's impossible to make the choice between 10. So they choose three. And if you're not in that top three, game over. So yeah, the proliferation of products out there is making it harder and harder and harder. The good news for consumers is it means that good products win. And the products that are winning are getting better and better and better. For the most part, every now and again, people found us shoot themselves in the foot and go, hey, let's add more and more and more and more and more features. And then they go, oh, god, I can't use this product anymore. Because it's got too many features. And then it's got so many features. We're doing so well. We hire more people. And then we've got to raise our costs. And so the cycle starts again. Monitoring on that, to be honest, because everyone's asking everyone, right? What is the future of SaaS? How is it going to look? What do we want? What do we need from SaaS? Is it going to survive? Because I think yesterday I saw this photo. Or picture on LinkedIn, where I think it was eight times already that SaaS has been declared dead. And let's measure it in cat lives. Is this the ultimate last one or it's gone forever? And I thought everyone's talking about how products get bloated by lots of features. And I was thinking maybe that is what we want. Because then we will pick and choose out of basically every product becomes a platform of features. And then you just pick and choose whatever you want. And that's what you're paying for. Yeah, but somebody still needs to. And until there is a time where I have my own server in the corner of the room, and I'm creating my own energy, and I'm running everything off of my own server, I don't understand how SaaS can die. There's that the business model, if I go in use Trello, or I go and use like cloud, or whatever it is, somebody has to pay for it. the ongoing power and energy that that assassin is consuming, whether it's storage, whether it's just server power, whatever it is, you don't go to a restaurant and say, "Can I just have some food?" And then come back tomorrow and say, "I'm going to have some more food. I paid for it yesterday. I'm going to go get it. This is new food. We've had to pay for this." So usage-based pricing might become more and more common rather than just fixed or seat-based pricing. Sure. That's still sass though. Any business where the underlying expenses are ongoing, the revenue needs to be ongoing, to match the expenses. Maybe if you call that sass or something else or whatever, but a subscription-based service, I cannot see how that is going to change. Because unlike my metaphor of getting food, in the time between I've got breakfast today and getting breakfast tomorrow, I'm not using their power at lunchtime and dinner time and what have you. That's usage-based. Right? Whereas with software, stuff's going on in the background all the time. I can't see how that's going to stop. There's no other. Nobody's proposed a new model. For sass to die, there needs to be a new model that's going to work and nobody's proposed it yet. We could go back to the old Photoshop model whereby you buy Photoshop version 4 and you have that for a year and then you buy Photoshop version 5 and then you buy Photoshop, you know, as the features come out, that's a viable model. But you have to keep all your photos on your computer. You can't keep them in the cloud. You manage everything. So there's no extra costs for Adobe or whoever it is. It's a viable model but if you want anything kept in the cloud, someone's got to pay that and that's just a description of the model. There's no getting around that. And until somebody proposes a new paradigm, nothing's dead. Just, you know, I'm just going to see it. I don't know why it would be dead. So touring publics. Everyone says it's dead for a different reason. That's the other thing. Everyone says, you know, some people say it's dead because it's not growing at 50% a year. That's not dead. That's just slowed down. I'm like, oh my God, that's still an insane experience. Some people say it's dead because you'll be able to build your own hotspot. Maybe that'll happen one day, maybe. Like, I don't know. I tried to build my own tools and I'm not doing it again. I tried to bait my own bread, fun, once, never doing a little one. I don't even make my own coffee because I can screw that up. I just, you know, just because you can do something doesn't mean you should. There are still web design agencies out there. Web design agencies, right? People who make websites. Like, and people think that SaaS is dead? Websites aren't dead, yeah. Web design agencies would go away first. Websites are so easy to create these days, but they haven't. But I get called to make websites. I haven't done that for, you know, my friend, I haven't done that in 10 years. So just go and do it yourself these days. No, no, just rather someone else. I don't know. Exactly. I once was talking to a founder on a podcast who, who told me their mission was to help with digitalization in Germany. And at first I laughed and was like, what, excuse me, what do you mean digitalization? I was like, we are all digital. Come on, we order whatever underwear for three bucks that comes from China. Two Germany in two days, nobody knows how everything is digital. What are you talking about? And then I realized he's he is right. There is still so much that is not it. I mean, there's facts. So people ask you to send facts for your visa application. And you're like, why? Yeah. Yeah. I was just like, we, we, you know, you look at zero. I don't know if like zero global products. It's like my ob. It's accounting software. It's just accounting. No idea. Apparently, people 10 years, even 10 years ago accounting used to be done manually, like on files on computers. And that's just mind boggling. Absolutely mind boggling. That was only 10 years ago. And you just can't, you know, and zero is a SaaS business that runs like they have all that infrastructure that they need to keep alive day in day out. But just the idea that that would not be a subscription service is just bizarre. So yeah, look, I haven't, I haven't seen a viable SaaS is dead argument yet. I, I think that, I think it will change a hell of a lot. I think it's becoming bloated and harder to get cut through, but has nothing to do with AI. It's just because the model has grown up and there are more people in SaaS than there ever has been. There, you know, the whole industry is growing as well. So yeah, it's just competition. If you, if you, if you're doing a PhD, right, in the, I don't know, the white blood cells of a rare type of spider in North Africa. And there's someone else doing a PhD in exactly that as well. But there's only one job, right? Competition becomes absolutely brutal. And that's what's happening in SaaS. Like there's all this competition. There's lots of people trying to do a CRM or whatever it is and it's eating itself. So for that reason, I think it's becoming harder and harder for bootstrappers to get cut through because there's so much noise. And I can't see that changing anytime soon. I can only see that getting worse for bootstrappers. Now we have to read another pitch deck VC era. Maybe, do you know what I hadn't thought of that? Because I hear a lot of people saying it's never been easier to build a product, never been easier to launch business. Yeah, it's never been easier to build a product, never been easier to launch business. I reckon it's never been harder to build a successful product that gets customers because of the noise. Big, big difference. Like there are so many ghost companies out there, you know, that, that have built a great little SaaS business that has all the features you need. And it's got no customers. The hard bit just is different now. Yeah. So to recap this, there is still SaaS, will the eye or without? Yes, yes. But if you want to do it, maybe don't. No, don't, don't do it. Just understand that the bit that you need to really, really conquer is not building a product anymore. The bit you need to conquer is go to market. The thing I didn't realize is I used to spend, this is as, as, as, this is two years ago, a year ago, I would spend 80% of my time, 90% of my time on the product, 10% of my time on the website and everything around the website, the top end of the funnel. I'm now spending 60, 70% of my time on the website and the marketing and 30% of my time on the product. Because the product has become the easy, easier bit and getting cut through has become way tougher. So that's where everyone needs to skill up, I think. Unfortunately, I'm really bad at that bit and I don't enjoy it that much. So I still find myself, I still find myself going back to the product. It's my happy place. Yeah. But, but yeah, that's where Claude comes in. It does teach you a hell of a lot of stuff about, you know, if you really delve, delve deep. But don't underestimate the power of getting an expert in to help you. Absolutely. I think that's such an important thing. Every time I see anybody say fired this marketing person and we're not hiring again because we want to do it with AI. Yeah. Okay. Well, good luck with that. You know, there should be an adult in the room. Like, it's like you said, you may not be junior. Okay. You might be a senior product person at a company, right? And you're really good at product and you were working with this amazing marketing person that was taking the product to the audience. And now you think, okay, well, they did a good job. So now I'll take whatever they did, feed it to Claude. And Claude is going to continue. - Yeah. - One of the things I like the way I like to think about it is everyone's just got a sidekick now. That's it. The team doesn't change, but everyone's just got a sidekick. Everyone's got a junior version of themselves. And nothing else needs to change. Nobody needs to be replaced, but the person, everyone's become a manager now. And if you can think about it that way, then productivity should in theory go up, quality should go up. But everyone has to be accountable for their own work. Otherwise you just get the same slop that everyone else is churning out. - Yes, yes. That's paying for it to watch, honestly. So thank you for saying that. - Yeah. - Anyway, I mean. - On LinkedIn, oh my God, some of the posts, and I'm like a moth to a flame as well. - I just started. - I just, I can't help myself. And I tend to throw in a bit of abuse every time I say it. And I shouldn't. - Yeah, I know. I know it's like every time there is a certain structure, I'm like, nope, not reading it, not even the second sentence. - Yeah, yeah. - Be better, do good, do good work. - Yeah. - All right. - Well, the one thing that I still haven't found, actually that AI can really help with, and it still gets worse is design. It's getting film and animation and stuff like that, really like it's getting much, much better. And, but just pure good old-fashioned design on the page design. I haven't found a tool or a way to get the most out of AI yet. I've found a, you know, a way to work with human-created design systems. But every product that I've worked with to try and come up with better design specs, better design, better design overall. It's pretty gross. It still is not there yet. So if anyone out there has found a good product that does good design, please let me know. Don't say cursor, don't say love of all. Good God. - Is it not a good time to ask what you think about my website? - I don't have a look, I look, as long as you're not a little honest feedback. This will be all for record. - Okay. - I'll look, I'll look. - Yeah. (laughs) - All right, but thank you so much. I mean, always super fun talking to you. I actually sent all my colleagues invite to Smile and said, okay, when I'm leaving, can you please use that? (laughs) So there's your VARL component. So thank you for that. It's real fun. And yeah, well, we'll talk next year. See what you think that. - Yeah, yeah, if I still go to job, I guess gross. - Yeah. (laughs) - All right, friends, bye. - Cheers. Take care.