So, Justin, what exactly do you do here in in your job description? What exactly do you do all day?
Justin:Oh, man. Welcome to the panel where two bootstrappers talk about building a better business and a better life. I'm Justin Jackson, cofounder of transistor.fm.
Brian:And I'm Brian Castle. I'm building builder methods, and I think I'm I think this is day one of a new cold for me.
Justin:Oh, you you know what? I'm suffering from something. I can't tell. It's like in my sinuses, but I get really bad allergies. So, yeah, my sleep score has been not great.
Justin:Yeah.
Brian:My my wife has had it, and I think one of my kids had it. So now it's my turn.
Justin:Yeah. So what's going on? What are you thinking about? You you we have a few ideas here in chat. You had one about distinct phases of startup in year one.
Brian:Yeah. Okay. So I guess just like before we get into it all, a couple like, a little, like, agenda for for today. We don't know what we'll actually get have time for and stuff. But Yeah.
Brian:Yeah. Yeah. Like, in terms of, like, business talk, like, the thing on my mind right now is, like, distinct phases of a startup, especially during year one, the fur like and and I I really think of it in, like we'll get into it. But I I think of it in, like, very distinct, like, okay. Now I'm turning a corner, Mhmm.
Brian:And this has different implications, different mindset, different activities, and stuff. Okay. And then and then, you know, just like in in general, like, the other topic, I don't know when we can get to it, is like, you know, obviously, I'm fascinated with with AI in general.
Justin:Mhmm.
Brian:Not just on, like, the development side and building stuff, but, like, in in the world. I'm I I I can't stop consuming content about, like, news items and, like, the big companies and the and the positioning of OpenAI and Claude. I listened to another episode of Cheeky Pint. I went back to the one where he interviewed Dario Amade, CEO of of Anthropic. Yeah.
Brian:And really, really great episode. And and then so it's it to me, what's really interesting in that whole world we can get into is, like, the pretty big differences in my view between how OpenAI is establishing themselves and how Anthropic and Clot are are establishing themselves. Yeah. And sort of like where we think the big players are. And I guess the the the question that that is on my mind that we could talk about is like, which one do do we personally feel a little bit more loyal to?
Brian:Mhmm. You know, like like many years ago, I sort of became a Mac person, and now I have an iPhone and a Mac and an Apple Watch and and AirPods. And like Am I going to be an open AI person? Am I going to be an anthropic person? Are we both?
Brian:How do you think about that? Yeah. Anyway, so what else do you have?
Justin:I've been wanting to experiment with building a marketing site with vZero, and I had a kind of a I'm helping a guy out, and I was able to test that experiment out. So I've got some pros and cons of building a marketing site with Vercel's v zero. Interesting. I think about this blog post all the time. Derek Sivers has this blog post about singing the counter melody.
Justin:And I I I've just been thinking about that in all sorts of phases of business and marketing. You know, when everybody is kind of playing one melody, what's maybe sometimes that's an opportunity to play the counter melody to do the opposite thing. So I've been thinking about that.
Brian:I like that. That actually another thing I I listened to earlier today was we can get into it when when we get up to that, but like who's the author of The Big Short and Moneyball? Oh. Michael Lewis.
Justin:Michael Lewis.
Brian:Yeah. And he was talking about, like like, can can AI replace someone like him, like a big time author? And interesting answer to that.
Justin:Yeah. Alright. Well, let's get into it, man. What's your first topic? What are you thinking there?
Brian:Yeah. So I I think that, you know, like with Builder Methods, I'm I'm still well within year one of this new business. And if I And I really think of it in, like, very distinct phases. Each phase has different lengths of time, usually several months each. Yeah.
Brian:And there there was the first phase, which was like, should I even start this business? What what business am I going to start? Is this the right direction for me? And so that first phase is like a thousand different questions, so much uncertainty. Nothing even exists yet.
Brian:I might not even do this. Maybe I'll do something completely different. I'm I'm probably weighing two, three, four different ideas for for what I'm gonna sink myself into. Then I start to, like, play around with okay. I I think creator style business that speaks to builders who are building with AI and and getting up to speed and, you know, building around a YouTube channel into a membership.
Brian:I think I like this direction. And then I have a 100 more questions about like, all right, how do I dial this in? What do I call it? What's the funnel gonna look like? What can I charge for this?
Brian:What should the shape of the product be? So all those questions are floating in the air. That's like phase one. I don't know what we call that. Maybe we call it like the uncertainty phase.
Justin:Yeah. You know what it reminds me of is the that hill chart that Mhmm. 37 signals often talks about. Like, that metaphor, I think, is so, like, the first phase, if you can imagine, like, going up a hill, that's the figuring out phase. Totally.
Justin:And then you get to the top of the hill. And then as you go down the other side, that's the making it happen phase. Right?
Brian:Yes. And and I was and, like, I'm not as good as you with, like, speaking while, like, drawing pictures. But but it the the picture in my mind is something like that hill chart, plus it's like two charts overlaid on top of each other. One is like one is like the stress level. Okay.
Brian:Or like how calm I feel or how secure I feel in the business. Right?
Justin:Does it also go up and down like a hill chart? Like, it's like the same
Brian:Let's see. So I I guess in the first phase, in that, like, uncertainty phase, it's like I I guess that Well, first of all, like, the two lines, like, the two charts would be, like, maybe like busyness factor. Like how much hustle is required to do this right now? How busy am I in the day? And then the other one is stress level or like, you know, feeling of security or
Justin:Yeah. They're probably pretty similar. Like, as you're ramping up something, a new project, it's like your busyness and your stress.
Brian:Actually, I think in that very first phase, it doesn't actually take a lot of time. It's really just more questions and thinking. You know, like I'm probably doing other stuff to make money and whatnot, but like it's the actual business hasn't really started yet. This is like pregame, right? Yeah.
Brian:It's like just So there's still like a high level of stress, I would say, because it's like, I don't know what I'm doing with my life. Right? Yeah. But the actual output and the activity, I'm not really hustling quite yet. This is still pregame.
Justin:Just figuring things out.
Brian:Yeah. So I would say it's like high stress, probably low ish activity. Got it.
Justin:And then
Brian:you get into like the second phase, which is, I would say, building. Yeah. It's like starting to build. And this is pre revenue. So you haven't made $1 yet.
Brian:Yeah. And so so I guess in my case, just to to paint a picture for builder methods, that that first phase uncertainty was probably if I start from the time that I that I started zeroing in on what builder methods became, it's probably like three or four months of of uncertainty. Maybe less than that. Two two, three months. Then I would say there's like another two or three month period of, like, building Mhmm.
Brian:With with no with no revenue whatsoever. Right? So for me, that looked like doing a lot of YouTube videos, making the website for builder methods, starting to do, like, weekly newsletters, learning a lot more about building with AI, And I guess, like, tinkering with what would be a product, but I haven't sold anything yet. So so this this is when that that activity starts to go up. Like, the hours, I'm I'm starting to hustle.
Brian:I'm I'm working on weekends. I'm I'm getting excited about what I'm building. But also the stress is sort of high because it's like I'm working a lot for $0, you know? So so there's that phase. Okay.
Brian:So then then we get into a third phase. And what should I call this? This is sort of like it's it's like a it's you you started making money. You you started selling something. You have the first dollars in the door.
Brian:This is like a chaos phase. Okay? So this is like the what what defines this phase, and I think for me, it was probably another six weeks or so, is you've started selling some sort of product, maybe a little bit of presales. You started getting customers. You're out in the market.
Brian:You you have traffic. You're getting exposure. You're getting leads. You're selling things, but you don't you haven't even come close to fulfilling the product yet.
Justin:Mhmm.
Brian:Like, it's everything on the site says coming soon. People are getting, like, discounted pricing, but they are paying something. But this phase is also defined by a lot of inbound messages. Like people saying, when is this gonna be ready? When is that gonna be ready?
Brian:I know you keep saying that's coming soon. Or will it do this? Or will you offer that? And so there's this, now this phase, like, introduces a new element of, like, an external pressure. And people are usually, like, very understanding of, like, okay, this is a startup.
Brian:He's hustling. Like but a lot of it is, like, self inflicted. Like like, I feel the pressure because I know that there are people who have already expressed interest in what I'm doing. Some of them have even paid as the earliest paying customers. So that adds another level of stress and pressure to actually deliver on the promise because now I'm actually holding like dollars in my bank account.
Justin:Yeah. And even and that phase is tough because you're also still there's no proof yet, like total proof that this is even a good business. Like, you're still holding all of this uncertainty.
Brian:Totally uncertain. You're starting to see good signs. People have actually taken out their credit card for something that you're offering. But that that to me is not validation. That's a good sign.
Justin:Yeah.
Brian:Right? So again, now it's very high activity. I'm I'm still hustling, still working nights and weekends, late nights, all day, you know, building stuff. What's what's really hard about this is that it you're still so so early that you can't just build the product that people are buying. You also still need to do the early marketing stuff.
Brian:Like, for me, that's like YouTube videos, email newsletters, this podcast. Like, just doing stuff that, like, yeah, I know I have, like, an incredibly long to do list of things that I have to build and ship. Yeah. But but I can't go do it in a cave. Like, now is the time to to keep that keep that buzz going.
Brian:Right? So so, like, that's literally, like, at least 50% of my hours just go to, content and and marketing. And then, like, 50% is, like and and now I have, a loud inbox. Whereas the the the phase earlier, it's like, yeah, I'm busy. I'm building stuff, but it's still quiet around here.
Brian:Nobody's actually knocking on my door. Now I have inbound messages adding pressure, and I have to, like, put stuff out to the world, and I have to start building the thing. You know?
Justin:Yeah. It it it's like there's multiple hill charts. It's you're because even, like, sometimes you figure something out and you're like, okay. I've got, like, you know, the uncertainty. It's like there's uncertainty around, can I and should I run a Discord as a part of this community?
Justin:Like, that's uncertain. And then you kind of push the, you know, the ball up the hill and then you're like, okay, now I know that part. But with marketing and making the product in the early stages, it's like you go through multiple phases of uncertainty and certainty all the time. Like Yeah. It's still, on one hand, can't is there enough, like, marketing interest and market demand out there for the direction I'm going?
Justin:And you're still trying to figure that out? And then on the product side, is there is this what people want? Will they pay me consistently? Is there enough revenue here?
Brian:And also, like, does this have legs a year from now? Like, is this is this just a thing that's, like, hot right now, but is this actually gonna stick around for a year? Like, is is the market still gonna demand this going forward? Am I going to continue to deliver on it? You know, a lot of those questions.
Brian:But the other thing that happens in that phase where it's like, I haven't fully delivered on on the promise. There's a lot of coming soon labels on the website. Like, this started happening to me in the last two weeks or so. It's like the first phase of like, the first batch of customers in Builder Methods, they were sort of, like, well aware of this is totally new. We're like the earliest members.
Brian:Of course, there's not a lot of content here yet. Mhmm. But as you get into like four, five, six weeks into that phase, now I'm attracting totally new people through YouTube who have no idea about the backstory of builder methods. Yeah. Like, yeah, my website has a couple tiny things that say coming soon and and we're still in like the early phases of this.
Brian:But like a lot of people just skim over that. And I literally got messages from new paying customers saying, What did I just buy here? There's no content. There was no library. There was no You log in and everything says coming soon.
Brian:So I did start to create private video content for the members in the last couple of weeks.
Justin:Yeah.
Brian:But I still haven't yet created I did now, but up until last week, I did not have the library video The video library section on the website. The only way the members were getting it was I was dropping Loom links into the Discord. Just to get it out there for people to chew on while I continue to build. Right? So anyway, now, this past week, I finally shipped the actual searchable video library.
Justin:I saw this. Yeah.
Brian:Video library is up on the website. I I now have four member only videos, like, totaling over two hours of or three hours of of footage, plus the YouTube videos are up there.
Justin:Yeah. I'm looking at it now. It's I mean, it to me, it felt like all of a sudden, we got a big drop of, like, what is this? 10 videos? 12 videos?
Brian:In the week leading up to this, like, I I had a I felt a lot of pressure to, like, finally deliver on, like, like, put the the infrastructure of the business actually in place. So I got the video library up there. There's also the members directory.
Justin:Yeah. Yeah, I saw that.
Brian:So now members can create their profile. It's got this cool thing where you can list your experience and rate yourself. Like, I'm a Talend expert, or I'm a Rails expert, or whatever. You could add your interests, then you can filter and find other members who match your interests or find people with certain expertise.
Justin:I thought that was really slick. I went through that process, and I built something similar for MEEPS and the Mega Maker community. The way that you did it was was cool. Like, yeah, there's this little experience thing here. I I should actually put less experience for Ruby on Rails.
Justin:And then interests, contact info, social media, like yeah. It's sweet. Like, got all this.
Brian:And then, like, the other thing, existing members probably didn't see this, but new members now have, an onboarding.
Justin:Oh, cool.
Brian:So, like, along with it because, like, the the member profile, it's not activated by default. You have to sort of like opt into it. Mhmm. So so I now, like new members who purchased today, they're gonna see a welcome page. First of all, it's like a a new survey.
Brian:Like, hey, why did you join Builder Methods Pro? I like to collect that information. And then and then the next page in onboarding is like, hey, activate your your member profile. Let's just get that situated. And now and then, like, third, like, head into the area.
Justin:Yeah. This feels great, dude. Like, this feels in terms of, like, how you've built it and how simple it is, it just feels really well done, this whole setup.
Brian:Yeah. I I you know, it's nice owning the whole app. Like, it's my own Rails app, I can design it however I want. And and I'm building with a lot with AI. So, like, each of those features only took me, like, a day.
Brian:It's just like I finally got around to doing it. You know? Yep. Now, with the launch of those and I sent a big email blast to all the members and wow. I feel old saying email blast, but anyway.
Brian:So so anyway, the now I feel like that marks the beginning of this next phase. The the next and and what this is is like, okay. I can start to breathe a little bit. Mhmm. Stress level comes down a a little bit.
Brian:If it was high before, now it's like mid level.
Justin:Yeah.
Brian:And so also like the hours in the day now can start to calm down. I still have a lot of work to do. But the nice thing now is that like I don't have that feeling of like being behind Yeah. Or or trying to play catch up to, like, all these people bought this thing. There's literally nothing there yet.
Brian:I'm I'm playing catch up. But now it's like people who are buying and coming in today, it's like they see something. They see a few hours of of private content. They see a video library that they could search. They see a member's directory.
Brian:They can join our Discord. And so now my job becomes just like get into the normal routine of this business, which is YouTube video every week, private content as I build my own tools and document that, and engage in the in the Discord, and and just get into the rhythm of that. And so, like, literally this week, to feel like, I started to work less hours this week. Know? Like, I can feel like, okay, I'm settling into what my routine is gonna be.
Justin:Yeah. That's great.
Brian:I mean, it's still like, now now it's also like a pressure of like, okay, now what's next? What's what's sort of interesting and the is that this is such a different type of business from all my previous ones. In that, like, this is a content creator business.
Justin:Mhmm.
Brian:But it's also in a space that is changing so insanely fast.
Justin:Yeah.
Brian:And and I'm and right now, I'm grappling with that big time. Like about like literally what what's my workshop gonna be about in November? What's what are my next four YouTube videos?
Justin:You feel a lot of pressure to make, like, content now? Is that
Brian:I the always have the pressure to keep the schedule up, but it's but now it's also it's just it's just I need to start to adopt a slightly different mindset with this. And that is like, that's part of the deal with this space of building with AI is that it's gonna change a lot really quickly. So like this week, Cursor two point o just came out. Yeah. Which is a pretty big update.
Brian:Like, they literally launched their own model. There's Cursor two point o. There's ClawdCode web version. These things are going in a very agentic delegate to agents, have them build stuff in the background. So that's like a sort of a new paradigm that's really starting to become the norm.
Brian:And it's like in the last couple of months, the idea of spec driven development, which Agent OS is heavily built around, I think that has a lot of value still. But I could also see that, like, things are shifting again. Like, maybe the idea of spending a lot of time and a lot of tokens upfront on writing specs is not as relevant as as it was literally three or four months ago. Really? Yeah.
Brian:Because, like, because the tools and the models are getting so good now.
Justin:Okay. You know? Like, you don't need as many specs
Brian:think there's a place for it, especially for large builds and also for teams and organizations who need more process and more standardization.
Justin:Mhmm.
Brian:But I think for a lot of us, like folks like us sort of like building new products and MVPs. I don't mean like vibe coded slop, but Mhmm. Moving faster than a than a 500 person team would move. Right? We can move really fast now with the current tools and especially Cursor two point o.
Brian:I'm I'm really impressed. I'm just playing with it today. And the the other thing is that these tools, like Cursor and Clod, they both have, like, plan mode, and that that can be a version of of specs. Like and the the the LLMs themselves are much better. Now you can, like, literally spin up, like, three or four agents to build the same thing and pit them against each other.
Brian:Like, you can do things like that that you literally couldn't really do six or seven months ago.
Justin:Yeah. I mean, you this to me, this is the advantage of you building a community, though, because now it feels like you have the ability to move fast and respond. Like, you could post a video today saying maybe you already have. I don't know. But you could say, hey, I've just looked at Cursor two point o, and here's a bunch of thoughts that I think the Builder Methods community needs to know about.
Justin:And honestly, if you're just getting started now, here's how I would do it. I would go into plan mode. I'd do this, this, and this. And, you know, you don't actually have to do as much heavy lifting anymore. So to be able to get that quick feedback, that's even better to me, like in a community format.
Justin:And you think that the other members are also sharing real time information with each other. So it doesn't have to all rely on you. I think the value of your community that you're building is gonna be the big thing. Like, I think the content will be a part of it. But I think your ability to respond really quickly is gonna be a really big.
Brian:I think you're totally right. Like that there's a couple different mindset shifts that I have to get more comfortable with, which is like like, one of them is I I started inviting a couple people onto a new, like, YouTube show podcast that I wanna start, like, through through like, I'm gonna probably gonna call it the Builder Method Show
Justin:Okay.
Brian:Which is which is like interviewing guests to just share your screen, show me your workflow, show me your tools, what are you using today. And I expect every episode to be completely different because everybody is doing things differently right now. And so I have the next four weeks booked out with guests. Nice. And so those will start to those will be like additional YouTube content probably starting to drop in December.
Brian:And so like that'll be one angle where it's like, I wanna learn and see what all these different people are doing and try to get the best ideas in terms of like workflows and then curate those and share with the community and also see what's sort of like working for me. The other thing is like, again, like, maybe the last four or five months, the big thing was spec driven development. Maybe the next three, four months is gonna be this, like, move toward agentic coding, agentic development. Probably three, four months after that, it's gonna be a totally new thing. And and the tools are gonna look much different than they do even today.
Brian:So, like, this idea that, like, what it's just gonna keep changing. Like, that is the thing that, really all professionals in our industry need to start to grapple with, but it's like but most people are gonna continuously be behind. That's that's the big pain point that a business like mine right now speaks to. But it's sort of my job to stay ahead of it. Yeah.
Brian:Everyone else is gonna be behind it. That's normal. But I'm here to try to stay ahead of it. So even if my website is promoting this thing, Agent OS, like, I'm already sort of thinking about the next thing.
Justin:Yeah. And the and again, you can see the potential here for, like, your weekly emails to either the big list or the builder method list. Like, hey, friends. I know you're all struggling to stay afloat. Here's three things you need to know about this week when it comes to AI and building.
Justin:Number one, Cursor two point o is out. If you've been looking for an entry point, this is your best entry point. I made a quick video here. And you can just see the the if if people are going to feel behind that this is exactly like people hire other people to solve their problems. People hire products to solve their problems.
Justin:It's like, what am I hiring Brian for? He's my buddy that goes out ahead and he chops down all the tall grass. And then he comes back and he shows me the way through the maze. Like that's the product. And I can see that being very valuable.
Brian:It's literally today. I spent all of today up until now, like, playing with Cursor two point o, just like throwing up, like, practice projects just to to play around with it. And then, like, between that and then, like, building some some real new tools as, like, small projects that are that be that create teachable content. Yeah. That's kinda what I wanna be doing every day anyway.
Brian:So, yeah, that's that's that's sort of the thing. Like, I guess the the one, like, mindset thing that I that I wanna take away from that is that, in all my previous businesses, it always started with one big product.
Justin:Mhmm.
Brian:And then the whole business for years after that was defined by that product. Yeah. And it's like once you establish what that product is, you can't just change the product every four months. You know? And like yeah.
Brian:Like, builder methods in general, like, big umbrella is like it's about building with AI. It's about staying ahead of the curve. Mhmm. But the reality is like every quarter, every season, the the workflow, the content, the teaching, the focus is going to change. And so like, even so like if all my SaaS businesses, it always started with like a product, and then it's like you're hammering like the product and use case and the job to be done.
Justin:Yeah. Yeah. Yeah.
Brian:Even back in the day when I did the productize course, day one, it was like productize services. You know, hire a team, sell it as if it's a product, but it's really services, build processes.
Justin:Mhmm.
Brian:That's the message. And I'm gonna keep preaching that for five years straight, you know, to sell that product.
Justin:Yeah. I mean, the thing it feels like you could help people with This just came into my head is that if the way things are going, if the direction things are shifting is that every product is going to be like this. It's going to be this organic thing that gets updated and shifts shape way more frequently than before, whether it's a SaaS business or something like Builder Methods, you're going to be able to be like, kind of, again, a trailblazer, as Emmett says in chat here, a trailblazer in terms of this new paradigm for building products, which is in the old days, you built the product, and then you could kind of like sit back. But now, no, things are shifting. Things are always like every day organically, your product is way more of a natural ecosystem shifting rapidly.
Justin:Like, it's it's going to be adapting and shifting way more than it used to. And I've, you know, I've been building this thing called builder methods. And I will just show you. Here's how it had to morph this way. And then it had to morph this way.
Justin:I think that's going to be an interesting discussion.
Brian:I'm always really fascinated with the content that am talking about. Like, right now, just today and yesterday, I started to key into this new way to think about the early design phase. Really, the whole focus of builder methods is like picking apart our development workflow, how we take product concepts, build them, and ship them. And taking that apart and redefining what that means in in our AI age. And that And again, that redefine That redefinition happens every three, four months now.
Brian:And one of the really interesting things, I've had my eye on how do we rethink the design phase. That's the part that I haven't really figured out. It still feels way too manual to you know, create wireframes manually in Figma or whatever, and like drag and drop these boxes and lines and type in and fiddle around with like stuff with the mouse all day in a design file.
Justin:Mhmm.
Brian:And and, yeah, we can like prototype stuff, but then it's like you're sort of vibe coding and it's not getting what you want. And now and I then I started to check out some some content about like, you can use Claude to create a JSON sample dataset for a new feature or a new product. Start with a JSON sample dataset.
Justin:Oh, yeah. That's a good trick.
Brian:Just start with that, and then feed that into whatever you want. V zero, Claude, cursor. And like now design a prototype based on this sample dataset. Yeah. And like that's it's such a new way to think about taking a concept and designing a prototype that's actually pretty accurate to what you wanna build.
Brian:And and like like I'm really excited about this. And like it's a it's a totally new way to think about the early phase of a product. And like that's gonna be a series of content coming up, you know? Like, just stuff like this just keeps changing. It's so it's so exciting to me, but it's also, like, so fast.
Justin:Well, talking about that makes me think I think we're going to go back to we're gonna go the the concept of dynamic content, like content that is constantly being updated is going to also be a way bigger thing. So instead of just having like, oh, here's the book I published on, you know, like marketing for developers. Here's the book. It's published. It's a PDF.
Justin:It's paper book that you you buy. This idea of like, hey, now we got to dynamically update content way faster. It's got to be live. It's got to be what's new today. And, you know, that kind of brings me to this thing I wanted to talk about, which is like building a marketing site with vZero.
Justin:Very interesting experiment. And I think it it it kind of goes to this thought of like dynamic content. Like now, can we make changes really way quicker? Can we go from, oh, this needs to be updated to updated? Even now, like, we just had our Thursday transistor team meeting.
Justin:And a lot of what we talk about is, oh, this support doc is out of date. We should update that. This marketing page is out of date. We should update that. This feature copy is out of date.
Justin:We should update that. And then that becomes a to do. And then, you know, eventually we do it. And then we put it up on staging or then we just say, is this look okay? And then we deploy.
Justin:Like, it's it's a fairly long cycle. Even though we're, you know, we're a thousand times faster than a lot of other organizations.
Brian:Mhmm.
Justin:But you can see where it's gonna get even faster. And and I mean, I think that was my interest in trying out vZero for this marketing project. So I'm helping out this film director who's built a sorry. He's he's done a film on the podcast history of podcasting called the age of audio. And basically, the film is about to go to distribution, and I'm gonna help him get the word out.
Justin:I'm gonna help him build up his email list, gonna help him get some more sponsors. And so I was like, okay, I'm gonna I'm gonna build this marketing site from with vZero, go out and look at a bunch of other documentary websites and see what I like. And I don't like by the way, most websites are so terrible. It's it's incredible. And
Brian:Wait. So that looks really cool for those who aren't who aren't watching right now. You just sort of scrolled through through this site, you know, for this film called Age of Audio, like black background. We've got some, like, logos. We've got some feature people in there.
Brian:Like, the layout looks super professional. Looks looks pretty dynamic. Yeah. How long and and you're showing it in v zero. So from first prompt to what you're showing there, how how long does that
Justin:take? I mean, I've been working on this a few days, but I would say it's certainly it's probably maybe actually I don't know. There's some things that took longer. Like I have this one part of the website where I feature a bunch of people who are in the film, but I manually went through the film and took these screenshots of, like, their faces so I could use actual, like, photos from the film in this feature section. So that took me a long time.
Justin:But in terms of, like, actual design work, like building this, it's a couple hours, maybe. Like Mhmm. Not very long. And some parts were very frustrating. For example, I initially said, you know, I've got he has all these people who are in the film, Mark Maron, Kevin Smith, Ira Glass, Fred Armisen, Kara Swisher, Roman Mars.
Justin:I was like, okay. Here's all these people in the film. I want you to go out, find their profile images on the web, and then just put them in this site. It said, well, we can't do that because those images are copyrighted. So what we'll do instead is generate AI avatars for each of these people Right.
Justin:Which were terrible. So I said, no, no, no, no. Okay. I'm gonna go get these screenshots and update it. And I gave it file names with each of the people's names.
Justin:And I said, take the file name, match it up with the person, and then put that profile image for, you know, their associated with their name. It works perfectly for 90% of the people. And then 10%, it just stubbornly kept the AI version of the person. And I could not like, get it to replace it.
Brian:I feel like that kind of thing feels like a v zero, like, shortcoming. And I feel like if you were to do that in Claude code with, like, Sonnet 4.5, like, I feel like it it would have nailed.
Justin:That's what I wanna try next.
Brian:You know, it would have nailed that.
Justin:Yeah. I mean, it it was still a very interesting experience. Like and I think along with the dynamic kind of nature of this, like, I like the ability of eventually, I could see a marketing website just living on the web. But with a model constantly running a little buddy, little AI agent that's and I just tell the AI agent, listen, I want you to, like, constantly, dynamically update this website based on the you know, what's happening in SEO, what's happening in the world, what's happening in terms of how people are talking about us on Reddit. And maybe all I'm getting all day is in Slack.
Justin:This little AI bot is saying, hey, we think that we should change the headline on this page based on some trends we're seeing. Do you want to do that? Yes or no? And I just reply, yes, no in Slack. And then it just goes out, builds new version, gives me a staging link.
Justin:Does this look okay? Yes. Should I deploy? Yes. And then it it just, you know, deploys the changes.
Brian:Think I do think that CMSs could be a thing of the past
Justin:Yeah.
Brian:In the very in the very near future.
Justin:I mean, I'm sure people are building the AI marketing CMS or whatever. I'm sure there's, like, 20 startups doing this.
Brian:But, like, you could you could definitely see like, I I could see where the friction would still be for for, like, a the like, the only reason I ever implemented something like Static or or WordPress back in the day on my sites would be if I was gonna hire other marketers or other writers or VAs to make edits on the marketing site. If it's just me and and it's just solo, I don't need a CMS. It's easier for me to just change the HTML. But that's when I would have reached for Static or something like that. And so if I think about teams who have marketers or, you know, assistants or interns going in there and posting blog posts or changing headlines and stuff, do they need a WordPress or a Static anymore?
Brian:And I I think probably the answer is still yes. But the only reason for that is probably like the git workflow and like the pushing changes live workflow. You know? Like, if that is still a little technical to like It's easy enough to to prompt any even if it's cursor or something like prompt, say, I wanna change the h one to some new text. That's pretty easy enough for anyone to do.
Brian:But if you're really nontechnical, to then commit that, like, create a pull request, check it on staging, and then deploy it to production. Like, you still need to know those mechanics. And, like, I I I do think that, like, professional marketers are gonna need those mechanics in their tool belt, even if they're nontechnical like that. Either that or the whole thing needs to become even easier than it is today for technical folks. Yeah.
Justin:Yeah. And it feels like we're on
Brian:Or, like, I should say, like, you can have Claude do that for you. You can say, like, just commit this and and push it live, like, and it will. Yeah. But for a large company with with a high level of traffic, you can't just let a marketer do
Justin:that. Yeah. You'd wanna test things out. It does feel like this is the we're in an age like, I I think of RightMessage, Brennan Dunn's thing. And the the hard thing about RightMessage is you had to so the idea was personalize your website.
Justin:Right? So you can based on where people are coming from, you create new content. So if you notice that they're coming from Facebook, you can give them content there. Or you can show a little survey and say, you know, are you a manager or a developer? And then you could dynamically change the content on a website.
Justin:But it was it required a lot of heavy lifting on your own. Like, you had to manually create all this stuff. I think now, I mean, something I would love today is, you know, you've got your analytics and all of a sudden it could be like, holy crap, you're getting tons of more new traffic from Product Hunt or from Blue Sky or from Hacker News. And it creates a little alert saying, hey, we've noticed we're just getting an influx of traffic from Hacker News. Would you like to publish a banner on that page?
Justin:And then it could just automatically go, hey, we see you're from Hacker News. Welcome Hacker News or whatever. Here's a special coupon or here's some special content or whatever. These things now could just be these little AI agents could just be watching your website as and especially your analytics for what's happening today. Like most of the time, I'm looking at analytics and I'm like, oh, I had a bunch more traffic on Friday, October 17.
Justin:Okay. Like, that's past now. I can't really deal with it in real time. But Yeah. In the future, you could see all of a sudden, it could be like, hey, we've we've noticed something happening in your analytics or out in the world, or a trend or people everyone's talking about, you know, how this one company messed up and they're looking for people to switch to, or this company is closing down, like, you should probably act on this right away.
Justin:And not only does it give you the notice, but then it also says, I've prepared a web page for you. Like, would you like to deploy it? That will be insane.
Brian:You know, like, on on my end, like, I have a thing on my to do list this month. I I I wanna set this up, and I also wanna create a video about it, which is like, you can and we should be we should have exactly that for the tech stack. For the right? Like, I have bug like, I have Honey Badger installed, like, tracking bugs, tracking errors on production. Right?
Justin:Oh, yeah. Yeah. Yeah.
Brian:Like every time there's an error, it should file a GitHub issue, and then that should automatically fire off an agent to go fix that issue. And I should just wake up in the morning and see a PR that I can merge. You know?
Justin:That would
Brian:And like Yeah. Like, that that's absolutely possible today. I just need to set it up.
Justin:It's just the glue. Like, the is what's kind of exciting about this stage is it does feel like it feels similar to that whole API boom, where it was like, there's this big API boom. And it was like, okay, we've got all this functionality and apps that's now exposed to the world. But you need to be a developer to figure it out. And then Zapier came along and said, oh, yeah, look at all these APIs.
Justin:Let's just like link these all up.
Brian:Connect them.
Justin:And connect it so that, you know, this all works. And now, you know, that they're worth billions of dollars. It it it feels like a similar moment where there's all this stuff you can do with AI and agents and things. But it's the it the glue is not there. Right?
Justin:Like, honey badger doesn't do that right now. But it could.
Brian:And Yep. Yep.
Justin:Yeah. It's it's both exciting and overwhelming.
Brian:It starts to it it is very overwhelming. I know, like, literally all people are are overwhelmed with it right now. I feel like we're as a as a society, especially as an industry, but even larger than that, like nontechnical people too, there's so so much to grapple with around, like, this is gonna change so much, and and everyone feels behind. Mhmm. And like one one thing, just like to take it back to zero back into ground level for us here, like, I'm constantly thinking right now about how is my workflow changing.
Brian:It's changing so much. Literally, less than twelve months ago, it looked completely different. How I create anything looked completely different than it does today. Mhmm. And I've been doing this twenty years, you know?
Brian:But it's Yeah. It's changed in the last twelve months. The the big thing for me right now is that, like, I'm noticing the way that I write content, the way that I create written content, and that includes, like, scripts that I might write for videos, I've really started to lean in and and adopt, like, AI as my assistant in that in that creative process. And and I do think that a lot of people mis represent or or misunderstand the value of using AI in in that sort of creative process. Like, it's absolutely true if you ask AI to create a killer blog post or create a killer video script for you from nothing, it's gonna give you crap.
Brian:It's not gonna be good. It's not gonna be usable. But if you give it your idea Mhmm. Like your like, the thing that you see in the world, have it create a draft off of that, have it trained on your voice, then then go through it six, seven times with the AI and change things and edit it down, then you get to something great. And I think that's even better than if you had written it word by word yourself.
Brian:Right? And so like, now, like, that is how I write. I almost can't write, you know, like, word for word. Like, at least I don't I just don't want to. I I wanna get my idea out concisely and clearly, and I use Claude for that.
Brian:But I haven't fully even embraced that same mindset in my development process. Yes, I'm using Claude code and cursor pretty heavily. I would say probably even heavier than most. Yeah. But I still think I should be ripping it apart even more.
Brian:Like I'm still holding on to the idea of like, but I need a wireframe, or I need a spec, or I need to spend days hashing out all of the nuances of what the data model for this new feature should be. In reality, I should be thinking more like, let's just have Cloud Let's fire up three instances of Cloud to build the same feature all at the same time, see which one builds it the best, and then let's run with it from there.
Justin:Know? And you just see what happens.
Brian:Yeah. Like like like, we really need to be thinking a little bit outside the box. Like, challenge your assumptions. It is changing. Our what it means to be creative is changing.
Brian:Our value in the in the chain as humans is is changing a lot. You know, it's it's all changing.
Justin:I mean, here's the here's the crazy part. And this goes into your next topic. Is we got an email to the panel email address. By the way, you can email us. It's the email is in the show notes.
Brian:I'm still surprised when, like, people tell me that they listen to the podcast. Like, I know that our crew here who's in who's in the live chat, like, I that's our like, I could literally see you here. You're live. You're you're you're awesome.
Justin:Yeah. Right? Yeah. Know they're there.
Brian:But they're, like, other people, like other friends and stuff. Yeah. You're like, yeah. I know what you're working on. I I listened to the pod.
Brian:I'm like, oh, you do? Alright.
Justin:Cool. I mean, that's the beauty of it. But here here's the questions. Corey Stone is saying, I've been wondering if I want to pay for only one AI service and at least minimize the number I'm paying for, which should I choose? For example, I like Claude and Perplexity, but ChatTPT seems to include more functionality.
Justin:And Nano Banana is the best image gen I've seen. So maybe I buy Gemini? Ugh. So here's the here's the other thing. It's like things are changing.
Justin:Everyone's trying to stay on top of things. We're trying to be open minded. And you had this thing about brand loyalty, and we used the example of Claude versus ChatGPT, etcetera.
Brian:Yeah. I I do wanna get your take on it. Like, how do you think about all the big brands that you've been loyal to for a And long and how how we define that now? But I guess to answer that question directly, I have a $20 a month subscription to I think I'm I'm on, like, the $20 a month plan on OpenAI and ChatGPT.
Justin:Mhmm.
Brian:I I do pay $20 a month for Cursor, and I pay a $100 a month for Claude. I don't think I pay for anything else.
Justin:I mean, this is it is getting it it's getting expensive, quote unquote. But for me, I've always felt like this is worth what I'm getting out of it. If I'm if I feel like I'm not getting value out of something, then I'll cancel it.
Brian:I am like and on that quest, I am also think I'll probably stick with my level on Claude, but, like, I might start paying a lot more for for Cursor pretty soon.
Justin:And I mean, for for me, I pay for ChatGPT. I think if you're gonna choose one, ChatGPT is just the in terms of doing a bunch of things well, It's great at image generation. It's good at writing. It's good at research. It's pretty good at code.
Justin:If you want just one general tool and you only want to pay for one, I personally think ChatGPT is incredible value. So you get Sora and video generation and all that stuff too.
Brian:So I don't know how you think about it, but I I'm starting to and I think that the brands themselves the companies themselves are leaning into this positioning. For me, it's like Claude is for work. Mhmm. Oh, ChatGPT is for personal. And I I do I do use but I do some work stuff blends into my ChatGPT usage.
Brian:Mhmm. But anything creative, including coding and writing for me happens in Claude. And, I mean, that's like work related creative stuff. Right? I in ChatGPT, it's all the personal chats.
Brian:It's all the fun, like, let's just ask AI what it thinks. I ask ChatGPT that. Yeah. I ask it, like, health questions when I have a once or, you know, or how to do something. But I also use ChatGPT.
Brian:I have a project in there called, like, business strategy.
Justin:Mhmm.
Brian:And that's like my essentially, it's my business coach, honestly. I I I actually use it in that capacity.
Justin:Yeah. Yeah. I mean, the I I think you're not your setup is probably similar to a lot of people. I I'm actually similar to ChatGPT. I use it for everything, but it's more personal.
Justin:That's like more personal stuff on that side.
Brian:Like, we've been driving on on road trips, and I'll I'll I'll throw up ChatGPT, and I'll I'll have it ask like, do a trivia game for for my kids and us. And, like, just just keep asking us questions, and that's what we do on road trips with it. You know?
Justin:Yeah. I I think I'm I basically always keep Chad GPT and Claude subscriptions going. I I use them all day, every day, multiple times. Sometimes I'll put a prompt in one and in the other just to compare the two.
Brian:One of the things I'm I'm curious to hear your take on this because, like, I noticed as I was listening to Dario Amade on on Cheeky Pint, hour long conversation, really good one. And I've I've heard him on a lot of interviews, but this was a really good one. I I love this podcast, by the way.
Justin:It's so good.
Brian:He was really talking about how, like, like, one of the interesting metrics was that I I I might I might have this wrong, but I think that Anthropix actual revenue might be higher than OpenAI's or close to
Justin:it. That's that would be very surprising to me.
Brian:Very surprising, but their revenue per customer is so much higher. They have far less customers and less traffic.
Justin:Oh, interesting.
Brian:Less users than OpenAI and ChatGPT. Yeah. But ChatGPT, but Anthropic is positioned for business. He spent a lot of time talking about They have some other products that we don't even see. Have Claude for enterprise.
Brian:And that's, I think, where they're really making all their money.
Justin:Yeah.
Brian:And and he spends a lot of time talking about, like, we are selling to enterprise. We are selling to companies.
Justin:Mhmm.
Brian:And our whole product strategy is around making Claude the best platform for the types of things that you would do at work. I mean, they they are getting into, like, health and biomedical. I think all the companies are. But, like, they're really focused on, like, making it a work and business focused tool. And and that's that's why they're easily getting away with, like, charging developers $100, $200 Yeah.
Brian:A month. And and I'm they're they're charging companies way more than that, You know? While at the same time, you look at OpenAI and ChatGPT. And this is where I get a little bit this is where ChatGPT and maybe Sam Altman start to turn me off a little bit when when when I think about brand loyalty. It's like I really gained a lot of respect for for Anthropix ability to focus on that on that goal and that strategy of, like, we're and I and I, as a customer of theirs, feel like they're building for me.
Brian:Like, they really care about what I care about and what I want out of a Claude. Whereas when I look at OpenAI right now, like literally right now in 2025, they've got ChatGPT, they've got Sora, they've got this new Atlas browser. They do seem to try to be getting developers with their GPT-five stuff and codecs. And, you know, Altman is is playing like, doing some trickery with the company structure and the and the nonprofit and all that. And, like and they're raising billions and billions of dollars, which, of course, like, all these companies are.
Brian:But, like, I it seems like they're, like, grasping at straws. Like, how are we gonna make money? Let's let's throw this at the wall. Let's let's throw that in the wall. Whereas Claude seems a little bit more focused.
Brian:Like, we're for business. We're gonna get a lot of money
Justin:for business. It feels a little bit more desperate even. Like like, they've said in interviews that Sora too, like, the reason they worked on that is they they need new revenue models. And and also just I think they need new reasons for investors to be excited about it. And so they're just like, alright.
Justin:This is gonna be popular. This is gonna be popular. Yeah. That's interesting because they they're definitely more of the b two c brands.
Brian:Yeah. They're definitely consumer.
Justin:Yeah. And Claude is the one that like, most of my work stuff is in Claude.
Brian:And most people, most like, quote unquote, normies, like, they don't even really they've never even really heard of Claude.
Justin:Yeah. Few of
Brian:them have on the news. But, like, you know, like like my like my dad maybe just started keying into, like, what Claude is, probably because I talk about it a lot. You know? But he's he's known ChatGPT for the last two, three years. You know?
Brian:Like, it that's the consumer, like, worldwide brand.
Justin:I mean, you know, I think a lot of people are loyal to the version that one thing about switching between different LLMs that sucks is that if one has all of your context, like in Claude, I have a here's Justin's writing style
Brian:Mhmm.
Justin:Project, and it has all of my blog posts and tweets and other things that I care about. And I don't want to set all that stuff up again, you know? Plus, it has all the additional context. And I noticed that Claude just announced
Brian:Memory.
Justin:More memory and stuff. So yeah.
Brian:That's the big thing, I think, for all of them. And this is also the thing where I get a little bit scared of OpenAI is how much personal information they have about me.
Justin:Oh, dude.
Brian:That's Like, I share my health stuff. I share my personal questions and stuff. But then they literally are integrating their memory of you into things like Sora. Mhmm. You know, in into, like, all all the other places.
Brian:And then now they have a browser, and now they have all these other, like, third party apps that are that like, they that's another thing. They're trying to make money with, they're trying to become an app store. Right? Like, they're trying to bring in DoorDash and Airbnb Mhmm. And and all these different things to have, like, built in integrations.
Brian:But then does that mean, like, if if I'm gonna call on, like like, OpenTable through ChatGPT, is all my personal memory data being shared with
Justin:Yeah.
Brian:Some some third party partner there?
Justin:I mean, there's a huge risk there. And I mean, it's even if you try not to put personal stuff in there or, you know, like your company's information in there, it's it's difficult because the utility is so high. Right. Yeah. That's gonna be a massive, massive thing.
Justin:I I do think it it's interesting. Like, right now, I think there's a lot of companies especially. Like, if you're a company, you're willing to experiment with tons of these different LLMs because the productivity promise is just so appealing. My business partner, John, is starting to get a little bit worried we're spending too much money on different AI, because every employee now has different tools. And I totally get it.
Justin:It can get out of hand quick. But on the other hand, go, man, the increased utility of these things, how fast I can move, I'm getting incredible value even though I am paying for ChattyPT and Claude, and I'm paying for Grock right now because I'm experimenting with the video stuff. Oh, interesting. Yeah. Paying for v zero.
Justin:And, you know, and Suno. And, like, it just keeps going and going and going. Sure. But yesterday, we released 12 new our our most popular page on Transistor is a free music download thing. And with Suno, you can generate instrumental music that you have the rights to.
Brian:That's a great thing for you to do.
Justin:Yeah. And so we generated 12 new tracks that we are allowed to distribute that are royalty free. Tons of utility for podcasters. And previously, I was only able to add like three a year partly because we were hiring not hiring. We were paying a human that was making these songs.
Justin:And it was just like, I would pay $300 a song to get, you know, exclusive rights to that little bump or whatever.
Brian:That is a whole industry that's gonna be killed, if if not already. It's like the there's a whole industry of musicians and songwriters who just create instrumental music for film and television.
Justin:I mean, you can see why Adam Watham's getting worried. That the whole template kind of templated Envato you know, like Envato's whole marketplace is here's graphics you can use, here's sounds you can use, here's sound effects you can use, here's templates. That's going away because it's just so much
Brian:Well, I think the templates will be there, just they won't be created by humans.
Justin:Yeah. Yeah, exactly. Exactly. Yeah. So it's interesting to see
Brian:Well, how do you think about it historically with other brands? Think about Google or Apple.
Justin:Yeah. I mean, so much of it is what people get used to. It's even interesting how many people our age. So let's say Gen X millennial, older millennial. I talked to tons of people our age that we over index for using Apple Music as opposed to Spotify.
Justin:Well, why is that? Well, when you talk to my daughter and my son, my sons who are all Gen Z, they they remember Spotify releasing the free version and how that spread like wildfire through their high school. And so for that generation, Spotify is established as the brand. They made a big splash for Gen Z high school students, and everybody loves Spotify in that group.
Brian:I'd say even for us to like like, I mean, I I do use Apple Music, but I feel like I'm in the minority. Most people I know
Justin:use Spotify. Many people our age use Apple Music. It's just consistently when I talk to people our age, it's like, Spotify, Apple Music, they're like, Apple Music, you know? The funny thing is I also pay for a Spotify account for work because I have to test stuff out. But I feel like we're an Apple family.
Justin:And so I can I will I will even though I could stream music in Spotify, I almost always like, if somebody sends me a recommendation in Spotify, I'll still go over to Apple Music and manually type in the song and then get it in my Apple Music thing?
Brian:I do hate Apple Music's interface and the fact that they still use, like, iTunes. And also the fact they still call it iTunes is kinda ridiculous to me.
Justin:But Yeah.
Brian:The I I actually have two different Apple Music accounts, one for work and one for, like, when I'm driving in in the car.
Justin:Oh, interesting.
Brian:Because I have completely different music tastes. Like, at work, I only want instrumental. I never want lyrics.
Justin:I'm surprised more companies have not, like, have not had more, like, profile stuff. Like, this is my work profile. This is my driving in the car with kids profile. This is Yeah. My kids are now requesting songs profile.
Brian:Like, this is the one that's gonna have, like, K pop demon hunters blasting through it.
Justin:Yeah. Yeah. Like, give me in the same way that Chrome gives you different profiles, like, I use that all the time. It's like, oh, this is my work Chrome profile setup, and here's my personal one. You know?
Brian:It is interesting. Like so, like, I just think about this, like, brand loyalty question. So, like, Apple, obviously, have been sort of hooked into their ecosystem for a really long time. Yeah. I think lately in the last couple of years, and this is industry wide, but I definitely feel it is like, you know, they I feel like they dropped the ball with with AI.
Brian:I feel like the the latest version of iOS isn't all that great. But at the same time, like, I'm not gonna go run out and get an Android phone.
Justin:Yeah. You're not going to Omari.
Brian:I'm so tied into the Apple ecosystem.
Justin:Come on, dude. Omari, that's the new thing, But
Brian:then with Google, we were talking about our data and I sort of got over the fact a really long time ago. Was like, look, these companies, they just have so much on us. No point in even trying to hold on to any semblance of privacy anymore. I do all the best practices and stuff. Like, I don't share, you know, whatever.
Brian:But like, I do know that Google knows so much about me. And I'm not even like a big like, I've had a Gmail account for forever. So that alone probably gives them so much data on me. Yeah. But like, I, you know, I don't have an Android phone or every anything.
Brian:But like, I guess, what am I asking here? Just about, like, like, this idea that, like, I like, at a certain point, like, I'm a Google person. Yep. Or I'm an like, I've had an Amazon Prime account for many years. So it's like, I'm just an Amazon person.
Justin:Then, This is one of the most powerful things in marketing, which is brand affinity. And it's like, why do, like, 90% of the people in the Laravel ecosystem use Transistor for their technical podcast? There's this brand affinity thing. And what determines that so much of it is this Seth Godin principle of people like us do things like this. And I've just seen that play out over and over and over again.
Justin:It's like it's like and you can sometimes see these cultural shifts. Like, it was like, suddenly creators are all upset about Mailchimp. And you could see online, like, they're all like it's almost like everyone's waiting to see where the tribe moves to. And then it was like, oh, Kit. Everyone's moving to Kit.
Justin:Okay. We're all on Kit now. Like, if you're a creator, people like us use Kit. And that's how we choose it. And there is other dynamics in there.
Justin:Like, I've been a Kit customer for a long time. And there's other reasons I won't switch because I'm friends with Nathan. Because there's I have a little bit of pride of being the longest running kit paid customer. So I don't like Is that right? Yeah.
Justin:I don't wanna give up my my
Brian:You were number one.
Justin:Crown. Well, I'm the longest running kit customer now that's consistently paid. Interesting. So there's other emotional reasons that people like certain brands that go beyond this. But, yeah, I think there's a lot of dynamics here.
Justin:All emotional, by the way, that people do not consider strongly enough.
Brian:Yeah. If we could pull this down to like the our level, like you're talking about, like with SaaS and Yeah. Other products and stuff like this it I brand, man. Like, brand does matter a lot more than people think, you know? And it's and you can't measure it.
Brian:But I I do think that it does make things harder when your company doesn't have some version of, like, people like us do things like this.
Justin:Yeah.
Brian:Because that's where you see the customers who come in and trial your product for one hour and then they cancel because some button didn't work. And then they just go try the next one. The customer who gonna look at your not even try your product, they're just gonna look at your pricing page and see if it has the list of check marks that they want to see compared to the other five competitors. And they're gonna decide based on that. They don't care if they actually like your product.
Brian:It just doesn't have these check marks. And like, it's so hard. It like, it's so hard to make a business on that. You know?
Justin:I mean There's pros and cons to this, which is once people have an affinity for your brand, that can last a long time. Like, I I think men in particular as consumers are an interesting version of this. Like, I'm a 45 year old man looking for what jeans I should wear. And I saw Nathan Barry recommended Dewar, and then Andrew Wilkinson recommended Dewar. And then Darryl Chimko here in the office also recommended Dewar.
Justin:So I'm like, okay. Like, Dewar's the jean brand. And so then I now if I buy jeans, it's like almost always gonna be doer jeans. Once that
Brian:Word of mouth, man. That is still there.
Justin:Yeah. It's like, if I needed a new pair of jeans, I don't wanna have to think like there's a trillion denim brands. And there's even a trillion more shitty denim brands on advertising on Instagram. Like, just give me the simplicity of knowing I know what I know these fit. I know my sizing.
Justin:I I can just order online, get it shipped. I'm done. You know? But it has the opposite effect too. It's it's difficult to get that that affinity and that commitment, from
Brian:the customer. Just to get one person to even recommend your product to someone else. You got to hit a high bar just to do that.
Justin:Can engineer some of that, though. I think we don't ask our existing customers and fans to evangelize enough for us. Like just saying, hey, just so you know, when you talk about our product, like, Transistor is a small company. We're six people. And we're fighting the biggest Goliaths in the world, YouTube and Spotify, and you know, some giant companies.
Justin:And it helps us so much when you recommend us to a friend. Thank you for doing that. Like even just prompting people and saying that that alone can help. And I don't think small brands do that enough.
Brian:Maybe this gets into the AI housekeeper a little bit. But this still sort of ties into the brands or at least getting back into the OpenAI Anthropic. Like, if you think about the big players in AI right now, I don't know if you think about it this way, but, like, I think that they are so big. And especially OpenAI, they are trying to be the everything app. Yep.
Brian:And they're trying to cover so much ground, and they have so much funding behind them, and they're clearly the leader. You can maybe make an argument for Google because they're so well established. Yeah. And and they have google.com. So like like a lot of the AI that, like like, Amy uses is like, well, that's what showed up when I go to google.com.
Brian:So I guess I'm using I guess I'm using Gemini. Is that what they call it? Okay. You know? So but then then you see these upstarts still in all these different categories.
Brian:And when I see something that is like this like, they're they're trying to take a huge swing with a new with with, like, a category that's, like, for everyone. Like and so, like, we saw this is a very ridiculous one, but, like, the what do they call it? Like, the housekeeper thing that that went around.
Justin:Yeah. You want me to show that that this robot housekeeper was the big the big conversation.
Brian:Which is, like, you know, it it it's it's sort of like vaporware, the whole thing. But the the what I'm saying though is that, like, I have no faith that, like, any new upstart is gonna be the next ChatGPT at this point. I think I think the core players are set. Yeah. And so, like, if you're trying to do something so huge with a with a worldwide market of, like, every person who needs their dishes done
Justin:Mhmm.
Brian:Like, I have no faith that you're gonna that, like, at some point, like, the big companies, whether that's gonna be, like, I don't know, like GE or or Apple is gonna have some device like that at some point or Tesla's gonna make a robot or, like or or or OpenAI will. Like
Justin:It is open in that sense. Like, there's lots of very well funded companies right now that
Brian:Like, we we saw the AI pin one, which was which was ridiculous and and died out really quickly. You know?
Justin:I mean, I think to start a true AI company now as an upstart, you would need funding. But Suno, you know, is an interesting AI company that has its own take on things. And, you know, this was they are now the established AI for making music. I think you could have an upstart in that sense.
Brian:I guess so. Yeah. But even that's a little bit more niche than what I'm talking about.
Justin:Like Yeah.
Brian:See, it's like like there was OpenAI just acquired this other one, I think, called Sky. And I think Sky was supposed to be some sort of software app that you install on your Mac and it just watches everything you do on your whole computer at all times.
Justin:Okay.
Brian:And it's sort of just like an AI that's available to you at all times. OpenAI was like, Yeah, that's something that we're trying to do, so we're just going to buy you. Things like that that are just so generic and so big that it's like, I don't know. Yeah. For me as a consumer, I don't have a lot of faith in, Well, I should get on board with this because, like, at some point, it's just gonna be acquired.
Justin:You know? I mean, the the way that it could there could be other competitive brands that come out is things like, you know, these these lower cost models that were coming out of China. And there may be other ways. I mean, this is what's so crazy, is we don't know how this is going to develop. Like, my 20 year old son is still convinced that this whole AI bubble is gonna burst.
Justin:Like, he thinks it's gonna continue to be a thing. But there's all this money pouring into it. We don't know. Like, we know the revenue numbers are high for these companies, but we don't know how much money they're losing. And things can happen.
Justin:Like, markets can shift and all of a sudden things can happen, you know?
Brian:I mean, I think he's probably right. I think we are probably are in a bubble, and I think it is gonna burst. That doesn't mean that AI is gonna go away, though, because there there was an Internet bubble. Right? And all the companies went away except for a few.
Brian:Yeah. And I think that's probably what's going to happen here is OpenAI and Anthropic are not going anywhere. But kind of everyone else, it's like, we'll see. You know?
Justin:In terms of like how like, bringing it back to independence and brands, I do think this is going to be an interesting time for small, like bootstrapped companies. I think there's still tons of opportunity in green space for small little indies to create brand affinity, to become a brand that people choose. And we do have some of these levers at our disposal that are still very effective, like building in public and sharing your journey, sharing your story, giving people
Brian:some I do think that's more important now than it was even last year. Like, people want like, that's what not only what works, but, like, people are hungry for human connection.
Justin:Human connection. And then there's always the opportunity, and this is that last thing I wanted to talk about, to play the counter melody or sing the counter melody. This Derek Sivers blog post, it just should be required reading. The idea is that, you know, everybody, you know, everybody might be over here doing something. And you see this all the time.
Justin:Like, everybody's rushing to create in YouTube land, everybody's rushing to create Mr. Beast like content. He is the prototype. Everybody is just trying to be Mr. Beast.
Justin:But then there's folks that are playing the counter melody like this World by Charlie. I don't know if you've seen these clips, but it's like a very kind of Blues Clues, Mr. Rogers, like narrating, here's how they make buns at the bakery. And he is totally playing the counter melody to Mr. Beast.
Justin:He's the anti Mr. Beast. And these on TikTok and Instagram, you know, these things have blown up. Like he's getting tons and tons of traction being the calm alternative to kind of the manic energy we're seeing everywhere else. And I think this is the opportunity for small little independence like you and I and everybody listening.
Justin:Play the counter melody. Now, doesn't always work. It's not always gonna be the counter melody doesn't always get traction. But when you notice, for example, like people are exhausted by mister Beast. It's like you see a mister Beast thing.
Justin:It's like, oh my god. Like, it's gonna be over the top. It's gonna be all like, I just I just need some calm in my life.
Brian:I mean, it's it I think it really just comes down to, like, have an opinion. Like, what do you actually think? Yeah. You know? That's the thing.
Brian:Because there's so many copycats. For forever, there's been human copycats. Like, let's just make another video that just like MrBeast does. And now you see AI copycats. Like, that that's what happens when you try to ask AI to, like, make me a YouTube script that it that matches the top performing YouTube videos.
Brian:Like, it's gonna be crap. Like, it's not gonna be interesting. And so was listening to Today, I was listening to This is a really good podcast. Plain English with What's the guy's name? Derek Thompson.
Brian:Yes, part of the Ringer network. He he covers a wide range of of topics and and does these sort of deep dives on them. Like, today, he had he had Michael Lewis on as a guest. Yeah. And he asked him, he was like, do you what do you think of AI in general?
Brian:Like, do you see it as a as a threat? What's going on? Like, are you writing a book about AI? And and he said, no. I don't really see anything interesting right now.
Brian:Like that like, I because because he was talking about, like, I look for things that are that nobody's talking about. Mhmm. That nobody can even notice. Yeah. You know?
Brian:And he was like, You know, I'm not really that worried about AI. Like Michael Lewis is saying, like, he's not threatened by AI taking his position away as an author because he was like, AI can't do what I do. He was like, it it cannot do what I do. Like, yes, it could write words. It could even write words in the style of Michael Lewis.
Brian:He but he was like, essentially, what I do is I go find the most interesting characters and the most interesting stories that nobody has ever noticed in the world.
Justin:Yes.
Brian:You know, like, when he's talking about how he wrote, like, The Big Short, like, he was like, there's there's no possible way that AI could have found the exact characters that he found and their exact worldview and and and the book that came out of that.
Justin:Yeah. I think that, by the way, that is a way of articulating what business people need to do is we need to notice things that are happening in the world. But nobody's really paying attention yet. Like that's, that's kind of what you do. Mean, there was a version of this with transistor is that I was, I had just noticed a shift myself even amongst my friends, whereas like way more people were starting podcasts.
Justin:Way more people were listening to podcasts. Like before, I never heard friends talking about listening to podcasts, and all a sudden, lot of people were listening to Serial. And then they were discovering other shows. And I'm like, man, something is happening here. And I spoke at MicroConf about it, like right at the beginning of Transistor where I was like, listen, the reason I personally wanted to pursue this was I just felt all of this stuff happening that was still kind of under the surface that it hadn't really erupted into the public consciousness yet.
Brian:And
Justin:it that turned out that the timing was good. I think this is what we do. We paying attention to this simmering latent sentiment and things that are happening and going, that's interesting. The one I posted in the builder methods Discord the other day was Arvid Kall saying, man, like, as a developer now, I have all these five to ten minute breaks when I'm getting AI to run stuff. What do I do with that time?
Brian:What do we do
Justin:with that?
Brian:Mean, that's a great question.
Justin:That is a tectonic shift in the way that people are working. And we don't have a solution for that. So what is gonna fill that time? Is it gonna be people browsing TikTok? Are people gonna be like, are managers gonna start saying, like, no, we need to use that time this way?
Justin:Or, you know, what's gonna happen?
Brian:That I mean, that that is and I I replied to him, that is such an interesting question too. It's like because, like, literally, like, every day now, I have Claude go off and build something, and then I start to work on the next project, or I break for dinner, or I go to bed, you know, and then I wake up in the morning and it's done.
Justin:Yeah. As entrepreneurs and bootstrappers or whatever, our our job is to, number one, get better at noticing these things. And then number two, get curious about them. Like try to unearth a little bit more around it and go, I wonder if there's something there. Like, let's play out a few scenarios in my head.
Justin:You know, does this I mean, that's my game
Brian:right now with with YouTube is that, like, I I can't just, like, generate ideas. I can't use AI to generate ideas. I can't even plan ideas much further out than, like, next week. Yeah. Because and also when I have my next idea, it takes me a week or two to just think about it in the shower or mull it over for two weeks before I'm ready to make a video on it.
Brian:Yeah. Because I need to figure out, what do I think about this? What's my opinion on this? I know it's a thing. I know it's new.
Brian:I know it's interesting in general. I know there's a lot of traffic to it. So I just posted a video today. My video that dropped today is on the new Claude skills feature.
Justin:Okay.
Brian:Yeah. And this was, like, brand new maybe a week ago, two weeks ago now. What always happens in the in my space, Claude drops a big new feature. There's gonna be 10 videos that drop on YouTube that day or the next day. Yeah.
Brian:And they're all gonna have the vibe of like, oh my god, Claude just dropped this new Yeah. Feature called skills. This changes everything. Yeah.
Justin:This is
Brian:like so great. Like, our this is so much better than whatever it was before. And, like, I have sort of felt that way about some features. Like, I thought that, like, sub agents were a pretty big deal when they dropped. Mhmm.
Brian:I started playing around with skills, and I started to think about it and look at look at the ins and outs of it. And I'm like, I don't think this is very useful for I could see how it's useful for consumers, and maybe it'll catch on there. But for developers and the workflow, I don't really see it. It's like, that's what I ended up making the video about. Was like, here's how they work, but I'm still struggling to see where they fit in the workflow.
Brian:And maybe they're not really a thing.
Justin:I don't know. Is the way that you build a brand. And that's the countermelody is if everybody else is like this kind of hectic, like, you won't believe what just dropped. And then it's like, oh, no. Brian's brand is the calm, thoughtful, not in a rush.
Justin:Like, of course, he's gonna get it out in a timely manner. But it's okay if he waits a week and thinks about it. And then he releases a video going, hey, friends. And they're basically looking at you like, okay, I'm gonna wait to see what Brian has to say about this. Like, I know he's gonna think about it.
Justin:I know he's gonna release it. And then you get to be in the place I mean, Basecamp
Brian:It's a struggle too because, like, I get it. Like, because, like, it's it's literally like sometimes it is like leaving viewers and subscribers on the table. But I'm really trying to not be like I I don't want my job to be dependent on, like, running to to the to the video camera as soon as a news drops. Like, I can't live that way. You know?
Brian:I have to I have to just be be calm with it.
Justin:Yeah. And yeah. So part of this is gonna be your own values and things. But I think this is what's great is, you know, Thirty Seven Signals has made a whole career out of playing the counter melody.
Brian:Yeah.
Justin:And they have been able to again, countermelody on its own. Just because you're opposing something doesn't mean it's going to work or it's going make money or it's going to be successful. But there is some advantage in it from both a product and marketing point of view and a process point of view. Like, part of their countermelody is they have a bunch of small product teams of two people, a designer and developer. And that's how they work.
Justin:And that has afforded them all sorts of pros, like all sorts of advantages that other companies don't have. They've been able to play that into an advantage. That's their countermelody. It works for them. We need to keep finding these things that matter.
Justin:And again, it always has to resonate with customers in a way. But there's lots of countermelodies that will resonate with people that you can build a great business on if you're, you know, exploring them.
Brian:For sure, man. Is that is that is that the ending note right there?
Justin:I think we did it. Yep. Thanks, everybody. Good to see everybody in chat. We had Emmett and Pascal came by.
Justin:Pascal has this recurring joke where he comes in and he goes
Brian:Like five minutes before we wrap.
Justin:Whatever we've been talking about, he goes, is this the discussion group for fans of blank? So it was like, is this the discussion group for the fans of Derek Sivers? Every single time.
Brian:You're in the right place today.
Justin:It's a good bet. Alright, everyone. We'll see you next week.
Brian:Later, folks.