How to Build AI Products in 2026, The Most Underrated Marketing Channel & The Big Lie About Competition

Welcome to the NAPKIMMAP podcast where we talk about the latest stories, tactics and business breakdowns from the world of tech, SaaS and much more on the back of a NAPKIM live from London. My name is Sabah and I'm Charlie and this is NAPKIMMAP. Welcome back to the NAPKIMMAP podcast we got a very special episode for you today. It's not special, it's a killer episode. And because we're going to be talking about how to build AI products in 2026. Also the most underrated marketing channel and that's actually an understatement. We are going to start marketing on this channel the second we start recording. We're also talking about the big lie about competition and we're going to talk about London maxing. No creativity maxing. This is I was creativity maxing and always creativity maxing. Who knows? Find out. See you later. Enjoy the pub. Everything's gone. Well, awful. Actually, we got our first ever hate mail on a YouTube video. I know I saw this. It was a big milestone for us, I think, to get this because it's all been loved so far but finally, ignoring us, loving us, but finally someone's taken the time to hate us. And that means that we are pretty much famous now. I know. It means we're on the map. Yeah. So to be fair, the hate mail wasn't that bad. It was also what is Veed? Is it even successful for you guys to even be giving lectures? This was the third comment, hateful comment in the row. Just like, who are you guys to be talking on this stuff? What do you think is the best way to reply? Do you kind of, the best thing is probably to ignore stuff like this, right? Or do you think it's good to reply to people? I've learnt the hard way that like, if you reply, you're just baiting them for another. You just leave it. What you're meant to do is they want you to reply. You know, talk about it publicly is what they want. I see. Okay. But listen, people are not coming to our podcast to hear about hate mail. They're coming here for business inside and scoops. Yes. Can't get anywhere else. Got any scoops today, Sabah? I think you do actually have a scoop today. All right. So I can either I can I can talk about two things. One is a serious thing or one is a silly thing. Let's talk about the serious thing. I first. All right. So I got the Q2 road map of V presented to me and I just I just went I went crazy. I just went crazy. Good or bad crazy crazy bad bad bad. Oh shit. Okay. And I ripped it up. I completely ripped it up. Okay. And I decided to rewrite everything really. And I don't know it was it was a big deal. It was a big deal. Talk us through it. So I think because V's been building for a while, there's a bunch of legacy that we have. The way we do things, how we think about things are approached to products like this culture and DNA is actually quite ingrained in us at V. And I think that so much has changed in the world of tech and AI in the last couple of years. And it's always been quite slow and gradual this change. We say it's changing really quickly. It's changing really fast. But how many people have actually just stopped and looked at what the hell they're doing and actually asked themselves a question like, are we doing the right thing? Are we building great products like are we? Yeah. I had a really high level existential level sort of thing. Like is this generally the right approach now? Yeah, exactly. And I just felt like we were going to do what we did last quarter this quarter a bit more. And what you end up the result of that is a bit more of what happened last quarter, which means your growth is not going to accelerate. It's not going to decrease. You're just going to be you're going to keep bumbling along. So rather than like show how do we keep iscerating on the thing like is this even the right thing anymore? Correct. Correct. And I think every single company that's so every company that started before AI, this is going to be very important to you. And I also think anyone who's actually had a career in software, this is going to be really important for you. And I've made a framework to think about how to build products in the age of AI. Okay. So before we go into it, the kind of feeling is that like this is almost like a pre AI roadmap for VIDE and the future video editing kind of thing. Do you feel it wasn't leaning in enough to the new kind of context we're in? Yeah, I just think like I think we have so many biases at VIDE to build editing software. And the way that I think about it and I don't know this framework is actually meant to be a bit more general and not just VIDE specific. Because if I make it very VIDE specific, I'll be giving away our Q2 roadmap, which I don't want to do. Sure. Talk about high level because I imagine over CapCup CEO's watching this. What else Final Cut? You'd be surprised. Actually, quick side note. One thing that I do that like I don't think enough people do is like successfully stalk their competitors. Like I listen to every single podcast that Melanie from Canva was in when I was starting to be every single one. I found the head of growth at Canva and I put his name into every podcast player I can find. We never had access to this stuff before, did we? I think people just think it's a given but like you're right, it's actually kind of mad that we actually just able to listen to this level of detail from them. That's not it's not press release. It goes in like real debt. So like this is mad. So I at the weekend, I got obsessed with notebook LM. That's a great product again. Yeah. And actually a lot of the learnings in this framework is what I learned from notebook LM. I googled notebook LM product manager. I found the product manager's name. I can't remember her name. And then I found she did it interview on then he's podcast. I was like, great. I then so I actually spoke for an hour and detail about how they did the product how they built it the origins. Then I asked Gemini. I was like, and what's the name of the designer. I found the designers named. I find his website and he's done like a real long form writing on how he designed the product. So I basically was able to effectively sit in the room of notebook LM and understand all the decisions they made and why they made them. Wow. So this was a very small side quest, but like you need to be better at stalking people. If you want to get anywhere, you need to talk better. There is some absolute gold out there. Yeah. Yeah. Absolutely. You just are curious. Curious enough sometimes. I've said before that curiosity is like one of the most important traits for any founder. Anyway, I agree. Nice. My magic framework. So. So this is a framework for building highly leveraged AI products. Okay. And the magic framework M a G I C. So M for magic stands for model native growth. What do I mean by that? When so a tool like gamma. Basically you put in a prompt. It makes a script for a presentation and then it generates the presentation images. Effectively, they've got open AI or in frock writing the script and now banana creating the slides. No banana 2 came out a couple of weeks ago. And their slides got a lot better overnight. And frock's new model comes out the script for the slides got a lot better overnight. So you need to be the products needs to be built in a way that when the models get better, you get better as well. Lovable when the new and frock model comes out, love ball gets better. Right. So I think this is the first thing. And this is a very big and this is a very big difference to say, oh yeah, but we've got these like AI features. No, no, no. The core products needs to get better. Yeah. Not the side feature that does a little summary for you. Right. Grenola, the way that granola works is when Texas speech gets better, it gets better. The way that your summary is written in granola gets better when I'm from it gets better. That is what I class as model native growth. Yeah. And he's really only is really. Yeah, whatever the for generous of AI, whatever it generates, whether it's like a rapper for image, video code text kind of thing. Yeah. And you're in a good place. If you have some good distribution. So here's where it's here's where you're doing it wrong. Right. Imagine your feed. And after you finish creating a video, it says, oh, do you need to generate a description for your YouTube video. That is doing it wrong. To be completely honest with you, that is doing it wrong because that is not the core product, simple as that. Number two, the A in magic is agenteic and API first. So we spoke a little bit about what it means to build an API today. And I think that software has got so, so easy to create that actually if you can create decent APIs that can enable vibe coders to integrate your products or your APIs into. That is a whole new growth channel. Yeah, but the time the address will market of people working with APIs is exploding right now. Correct. Because of tools like Florida. Correct. And if your product can also integrate into these amazing, let's take gamma for example again. Right. If gamma integrates into clay, I can finish a sales call, transcribe the sales call and then create a presentation for the next time I speak to the prospect that is based on the call that we just had. Right. So there's two levels here. One is making your core products pluggable. But then the other thing is making your core features API ball as well. Right. So this is really building this API first, the genetic world. We've seen something really interesting happen recently is that a lot of tools that are like web interfaces. The API usually comes in one of the like the more expensive pricing plant. And H refs the SEO analytics tool we talk about every episode basically. They just made their API for the next time. the first time available in every plan, no matter how cheap it is. And I think that is just a sign to where things are heading as we talk off to on this pod. But I think that they have seen pressure from API only products like a lot of people started using like the data for SEO API for SEO data. And I think the big people are noticing this and are like, well, it has to be like table stakes to include this now, you know, 100%, 100%. I mean, like, I think the other thing is like, we have this sort of some level general intelligence that has been built. Right. Yeah. The language that this thing talks is API. It's that simple. Yeah. Why would you not make the smartest piece of technology in the world talks to your products? It's like, it's mad. Anyway, yeah. Carrying on. So the G in magic is a generative focus. And what I really mean by this is building a lazy UI to let people generate. And if you look at chat, you see, look at Gemini, look at Amphropic. I think I could probably be the front end developer on those teams because they've got a black color screen. And then they've got this prompt box. And then they've got this button. Yeah. The design work is quite minimal. The engineering work is quite minimal. And the reason why is because most of the leverage is happening on the model layer. Yeah. So don't go and build heavy UIs below to the UI and don't try and build layers and layers of products. Well, everyone has to learn every new UI, right, for every new product. But nobody needs to learn when it's a text box, you type it, and you don't need to learn that. Yeah. Exactly. So like, let's just keep it really simple for your developers. And like every I was reviewing some designs of the team. And I was like, what's that? And they're like, oh, this is like a way to navigate like other templates. I'm like, that's two weeks' worth of work. Get rid of it, right? And I was joking. But like the point is just like, keep it simple, guys. Like keep it really simple. And I think the best fast like look at loveable. Like the UI is a chat on the left hand side. Like, is it a joke? It's a joke. You know, anthropic. Didn't they say they don't like, you know, 14 billion run rate? Is that true? I believe they said this. Yes. How much crazy? How much do they have? Yeah. Not too much, I don't think. I see more thinking caps than UI. I do. Anyway, really, really simple. Really, really simple UI. And now the next thing in this lazy UI as well is like one shot, right? If you go to a product and you one shot something, you get good guys. Like, you know, if I go to a lovable and I say, hey, make me a chess chess game, deliver me the chess game. Yeah. Don't, don't half-ass it. If I go to Nanobinana, I want to one shot something. Just, just wow me. The great thing about one-shotting as well is it's really easy to tell the value. If you record your screen and share you one-shotting something, people just get it immediately. And as you've discussed, it's, and we've discussed in this pod, like, that's how a lot of these kind of new AI products grow by video going viral on social. And so if you can show something being effective really quickly, that's going to work so you can work well. And like, I think we forget, like, we love our companies and what we do in the work we do that the users don't care. They're just like, just, honestly, just give me the video. Like, you know what I mean? Yeah. I'm so tired. I've got two kids at home. I've got a cook dinner tonight. I've got to do a laundry. Just give me the video, right? So like, your one-shot experience needs me amazing, right? So that's the G. Yeah, just like, as you said before, people don't necessarily want to edit videos. They want a good video quickly. Yeah. Simple. Really simple. Next is the eye in magic, which is instant time to value, which is related to the one-shot. But within this instant time to value, I've actually created a subframe work, which is what I call my great product framework. And I think great products basically work every single time incredibly simple to use and get you to the aha moment really, really quickly. And if you have to spend time figures configuring, setting up things, you've failed, right? So like, instant time to value is so, so important. And, you know, on the topic of being obsessed and stalking people, I've listened to every single podcast that didn't feel from figmo's ever done. You serious? Yeah, it's in your handful. Yeah, interesting. Like, I'm sorry, but this is a smart guy. I mean, these guys literally built like a currently $15 billion company in every single podcast. He keeps talking, he kept talking about time to value. And when someone says something quite a lot of times and they're really smart and really successful, you should kind of listen. Yeah, of course. And he kept talking about time to value. And I was like, and like, the lightning round is like, what do you think about the most is like time to value? Like this guy is like, it's almost like the company has like one metric for growth. And that is time to value. And I was like, V, we don't think about time to value. Like we do time to value. You can use the product without signing up. We get you in there. But like, what does time to value mean when you take it seriously? There's some products. It's harder to do than others, but it's important. No, no, no, no, I'm not saying there's some products you shouldn't do it for. We should try and do this with everything. But there's probably some way it's a bit more set up and learning curve required, right? Then say like a, I don't know, a cow AI or something. The job of any great product designer is to make products more accessible, easy to use. Totally great. What's a good way? What's a good resource is to learn like how other people are doing this? Like, a good time to value. Like I've had mobins pretty good for this, for example. It's like a user flow repository of like all the big and best products and things like onboarding flows, off-boding flows, that sort of thing. I think again, it's like down that obsession rabbit hole. Yeah. Being a, also, I think just using good products, being a student of what's working, what is it? I think that's a big thing. So like, this is a, this is a related, but I, I wanted to look at the jewel and goes pricing page. As you do. So I downloaded the app and I was like, cool, I'm sure I can sign up for jewel and go and click, you know, pay and basically get the paywall. It's green shot to take it into my research. And so I don't know, jewel and go, I sign in and then it's like, what language do you want? I was like, Spanish, thinking that I'll now get the paywall. It was like, oh, less. And I was like, oh, it was like, say cat. And I was like, where's the pricing page? And then, and then, and then 30 minutes later, was it timing to value two halves and you can find the pricing? Well, like, well, like, 30 minutes later, I'm like, deep in a Spanish lesson. It's like I was just looking for a pricing. And I've got this like, jewelline go streak going that I don't want to like give off. And I've done very, very well. And I never got to the pricing page. And what that was was great time to value in great activation. Yeah. And they don't want to charge you any money before they know that you're activated. They take that so seriously. Right. And I think like a company needs to obsess over something. Right. Yeah. And jewel and go, it's the activation experience. I think, um, figmas time to value. Anyway, um, guys, fixie time to value Dylan said so. Nice. So that's I. We're on the final one. Okay. See, um, compelling leverage. Uh, so I think the input to output ratio you get from a product should be hugely disproportionate, disproportionate. Sorry. Let me explain. I can go to notebook L. M. I saw I write what's the history of REM. So I put five words in. And then it goes and searches 20 links on the web. Pools of the content, puts it into a check experience. Summarizes it all. So it's basically like looked at 20 web pages for me. It's now summarizing all the content with a large language model. And then it says, do you want to make a long form podcast, a video, a slide deck, a mind map? What do you want? And I click video. It's like, all right. Cool. Didn't that be to the podcast as well? Yeah. Fine. So I've now got a 40 minute podcast, a 10 minute video and a slide deck that's been researched by 20 links, summarized by a large language model from writing in five words. Crazy. In less than a few minutes. I love if you showed this to someone like in 2010, they would literally think you're like a wizard from 100 years in the future. Yeah. Yeah. Basically. Like it's kind of, we're just used to it. Like, oh, you know, oh, yeah, that's what you can do now. Then like most people don't look at it again. But to me, it's still wild that the stuff even works at all, let alone that it's available for like 20, quite a month or something. No, it's, it's, it's, you can generate, you can do free generations a day every day on notebook. Oh, you can use it for free. Great. Thanks, Google. Yeah. Thanks, Google. And like, I think that that's incredible. I'm going to put it out there. I think that's incredible. Me too. Yeah. And that is dish proportionate input to value created. And it's always been the case. But in the age of AI, it's even more disproportionate. Another example of disproportionate input to output ratio is it's all like lovable or a tool like granola. Like I actually, granola is like, I just click a button, turn it on and it gives me a proper great, great down of my meeting. Like that's disproportionate. I did nothing. Yeah. I just turned up on the call. Lovable. It's disproportionate. I say, give me a chess game and it gives it to me. it knows that you want the end result not just like to do the work. Don't half-ass it. Like we said earlier, like a great one-shot experience guys. I like the framework. Model and A/S of growth, A/Gentic and API first, Genitive Focus, Instant Time to Value, Compelling Leverage, Any NAP, KMAF listeners building API tools. Would love to know your feedback on this. What are you building as well, anything new? Has this, as thinking through this, like is that changing anything about how you're working on VID? It sounds like it has for the roadmap completely. Completely changed. I think we were working on all the right components at VID, but I don't think it was architected in a way that was, quite, on quite modern. I see. To be honest. And it's like, you know, it's like, it's like the year that Facebook said, "No, no, no, we're doing mobile." You know what I mean? It's like, you had all the right things. You had the user base. You had the right tools. You were building the right things, but you were just doing it on the wrong platform. And like we were doing all the right things at VID, but I just don't think we were putting it together in the way that made sense. To me anyway, and the vision that I have for the company. It definitely feels like we're going for you one of those, you know, the technology shifts that happen every five or 10 years, where a bunch of companies grow 10X and a bunch of them basically go away because they don't take advantage of it, right? Yeah, yeah, yeah. But like, I think it's still so early. So like, there was a study done by the guys at A16Z. And I think it was like 30% of people in the US used chat to be pretty once a week. I mean, like, come on. It was pretty early. You know, it was like-- I figured December-- That's like, that's basically like saying only one in three people use the internet. You know? In December-- Yeah, it's about-- Yeah, it's about four months ago. The best estimate I could find a few months ago of Claude uses was like 100,000 users. I think it has exploded since. But like, we are definitely-- If you're listening to this, you're in a bubble. You might not realize this, but you are almost certainly in a massive bubble. Hey, enjoy the bubble, guys. It's cozy in here. Well, actually a small cozy bubble, sorry, of people who talk about HREFs and Notebook LM every day. Yeah. By the way, summarize what Notebook LM does to those that are-- before we move on to the next one. So first of all, if you're living under a rock, Notebook LM is actually bigger than complexity. Did you get that? Interesting, didn't know that. Notebook LM is asking-- it came out of Google Labs, actually, which is Google's effort to kind of doing more experimental things. And they asked themselves a really interesting question, and a very simple question, which is like, hey, instead of starting with a prompt, can you start with a resource? And the reason why that's really interesting is because when you ask a large language model a question, the answers of Phil can feel quite generic. When you start with a source, it's more of you, right? It's taking my marketing deck. And then when I ask it a question, it's taking the answer from my marketing deck. I think if I look at what I actually-- every morning, this is what I do. I open whisper flow, I open my Notes app, and I just talk for about 20 minutes. What's on my head? What I care about in as much detail as possible. In the morning, I get to the office, I then ask a large language model to summarize it. That's an example of using AI to synthesize large-fossil information. I love that. It's so good of being able to do this in so many formats, like it created for graphics, videos, now, podcasts. It's just crazy to me. Yeah. What do you think of my magic framework, Charlie? I like it. I think it's-- anyone building-- on the verge of building my first AI feature for products I have, it's basically helping you use AI to build better communities and understand them better. But I like it. Especially some of these are quite AI-specific. So model-native growth, as the model gets better, you want your product to get better. Some of it, I think, is just good product principles, instant time to value and stuff like that. Which I think is, in some ways, might be easier with certain AI products, especially if it can one-shot stuff. So yeah, I think it's a good way to think about it. So a few interesting case studies here. So like, FreePick is an incredible company. They are killing it with video generation, image generation. They are one of the leaders in aggregating many models in one place. Really proud to be working with them as well. They've got our models in their platform too. And they went from being a stock image website to being the leaders of like, generative AI for like, you know, lows and lows, millions and millions of people. They took the business and turned it on its head. They started a refresh piece of paper. They could have said, oh, we've got all of this stock. And people love it. Let's have this way to edit the stock of AI. That is adding AI as a feature, not reinventing the product. It's really uncomfortable to do this. It's really uncomfortable to admit that I've got a company that is not actually riding the tailwinds of AI that might not be relevant. That's really tough. Really tough. But you've got to have to deal with it. You are. You're literally going to have to deal with it. But yeah, you know, ultimately, people can have nice, cushy jobs. You're in the arena. And this is what that means. Is that you have to go through uncomfortable situations like this sometimes? You know, daily. I want to switch gears slightly to some of the topic I'm calling creativity maxing. Do you want to hear? God, go on. So there's a guy, Stephen Erby, comes to ramen space. Great guy. And he's basically been working on something, which is-- it basically doesn't really track with any of the advice we talk about of how to find-- who would say, you know, look for, like, search demand, look for, like, competition, that sort of thing. And I love great counter examples like this. So he started doing 25 products, ships, and 25 weeks, and like, sharing them online, and that sort of thing. Lots of really interesting variety of stuff. But he ships something recently, which just absolutely blew up. It's called channel server. It's basically like, you know, like the old TV guide, where there's all the different channels, and the different times, and you flick through. It's very like sort of, you know, 2000s, that kind of thing. And it's basically a way to browse YouTube, and a really nostalgic kind of TV guide kind of way. And it's really fun. It's really nice. It's got quite well curated channels and videos and stuff on there. And this was like, I don't know. Maybe like the 10th thing. He built or something. And he would share it on things like Twitter or-- Can you get it, guys? Yeah. Let's just share with the camera. I hope the camera can see this. And Charlie needs to clean his screen. Oh, we can maybe try and screen share this, like so. But like, it's just that kind of like, when you click TV guide on your remote in the 2000s, you'd get this. Yeah. Yeah. It's really nice. And this absolutely blew up. People just love that they-- you instantly get it. Because everyone recognizes TV guide, and they know what YouTube is. And you just get it immediately. And there was number one on how can you use, covered by tech front crunch. 10s of thousands of visits. ASAP Rocky, like his team reached out, because they're like, oh, we're trying to make something like this. We love to know how you did it. I mean, of course, ASAP Rocky team would reach out. Because-- and I'm sorry. Like, what are they finding hard about building this? Who knows? But they reached out. Oh my god, your answer is some really good IP here, guys. And it's just like, I think sometimes there is something to be said for creativity maxing, which is just create lots of stuff that you want. Do you think is really cool and sharing online? And sometimes the internet just-- it just takes off completely. He started to make money on this as well. We started getting sponsors and things like that. So yeah, I just want to shout out Channel Server. And one of the other things I'm interested in about how this grew is-- this is like the third time in, like, last week I've heard of someone blowing up via threads. Not by Twitter, not by SEO, not by ColdArage, by threads. I forgot threads existed. Have you been hearing this as well? I am unfortunately-- if you hear the same thing-- I'll say I love threads. You know when you hear the same thing in three weeks? If the same company product or thing has been mentioned three times in one week, you need to drop everything you can get to it. I keep hearing about threads. So I then look this up. And apparently-- So I might be losing the AI war, but he's actually doing really well with threads. So it's supposedly threads now has more daily mobile users. The Twitter has got 141 million versus 125. Is that a joke? 141? What was the million of it? What do you mean? 40 caraters? Anyway, so-- Oh, yeah. It's joking there somewhere. I've worked-- I've worked-- I'll put it in. But yeah, apparently threads is like-- so when Elon Musk took over Twitter, there was a bit of an Exodus. And lots of-- some people went to Blue Sky, which is full of kind of quite crazy people at my opinion. All the proofs are cool. Lots of like-- Lots of like kind of really just quite normal well-adjusted people went to threads instead, which is where they get their Twitter like hit without it being-- Blue Sky's got a very specific culture, right? Or we're going to go into it right now. But loads of normal people are just using threads. And it's really apparently not a bad place to share your content, because it's got a lot of people who are very active on it now. OK, so on the top of the time to value, as you literally just spoke about it, I downloaded it. And because I've already got an Instagram account, I'm in, and it's pulled my social graph. We-- That is times value and an already smart magic. My boy Pat Walsh is there with 53 likes on a starter story post. You know Pat Walsh is in all of it. Of course he's beat on this for like two years from. Let's just say. toasting our. Let's just say, if Pat Walsh is marketing in a certain place, you should be there. You should also be there. Pat, I hope you're watching this. We're watching you. We're looking at you. Pat Walsh was like one of the first people to market on Reddit, like really successfully, until he like. Did he get banned from like. He didn't get banned. No, no, no. He got banned from like a small. Not all the whole of Reddit, but there was a specific channel. No, this is a good thing. He did it so hard for so long. Got so many views. They had to get rid of him. They were like, you take it over our entire subreddit part. You need to go. Yeah. That's the sign of success, when you're posting too hard. Yeah. So yeah, I don't know. You need to maybe need to read the thread strategy. You know what? We're getting on threads. Sorry, we're getting on threads. Yeah. And there's a startup idea there as well for someone. Well, okay, here we go. Also, basically any plug-in for running Twitter, like. You just do it for threads if the API allows it. Yeah. Do you reckon the API allows it? Yeah, probably. You think so. And Twitter's 50k a month for the API. Yeah. Look at every plug-in for Twitter that's making money and just build it for threads. So. I just want to quickly touch on it. And it's clear that it's putting my Instagram graph based on who it's suggesting I follow. But I don't think the tech bros, the finance nerds, the Bitcoin Maxis are going to leave Twitter anytime soon. No, I don't think so either. However, I think your Instagram user is on threads. So if you want to market to like develop hers and founders and product people, I would go to Twitter, X. If I wanted to market to a more normal mass market consumer base, I'd go to threads. And I reckon there is some really under-malatized attention on this app. Yeah. Like I reckon this app can be hacked really nicely. And I'm sure Pat Walsh is doing it. Anyway, I'll shut up. Yeah. Quick comment from Seven. This was in the group chat he posted this. I'm telling you all, don't sleep on threads. It's become bigger than X. I'm dead serious. Buffer and other apps will agree. Must kill the X brand for normies. Plus X seems to be optimized for accounts. They are already big or therefore kind of inflammatory to ask rage break posts. I think there's something to what you're saying. All right. This time next week, I want to see your threads marketing strategy, Charlie. Yes, let's do it. I don't want to be cross-posting on both platforms. So I also want to talk about competition, which is kind of a running theme in this. Like it's good or bad, depending on how you think of it. We often talk about, oh, run towards demand and run towards competition as a good way to find good markets, which I still think is true versus what Peter Seal says about. Avoiding competition, but you know, it's kind of, depends if you're looking more of a bootstrap versus VC, sort of way of thinking of it. Anyway, that's not what this is about. Sean Vp. So Sean, who, what's Sean best known for? Is this Sean, my first believe in Sean? My first believe in Sean. My first believe in, is probably what he's best known for. He's known for just being incredible storyteller. Yeah. He's brilliant. But anyway, he shared a post, not on threads on Twitter. Maybe it's on threads as well, about the big lie about competition. People tell themselves, "Well, kind of lies to say that ego saying, "Oh, it's so hard to succeed, it's so competitive." He shared this chart, says, "For a thousand people, 500 of them are not going to try. About 400 of them are going to give up. The last bunch of them, they're either moving too slowly, they get distracted, they're not focused." And he said, "There's only like a handful of people who are actually really trying for every kind of idea and are focused and are good at what they do." And that, there is, it's not, if you're going to fail, it's not because of the competition most of the time. So, I want to ask you, because you've been through a bunch of phases of V-D, right? So you've gone from when, obviously, it didn't make every business start this way. You didn't make any money. To like mid-stage, making like a millionaire, now you're making a lot more than that. I don't know how public you are, if you're revenue. But how has, how you think about competition changed over time, like running a, what's now a big start-up? When you're like a startup, you have no competition. Like, as in, like when you begin. Yeah, like no one's a competitor. Yeah. Everything is blue ocean. There's so much room to grow. Oh, I got it. Yeah. Yeah. Whatever. Like, you shouldn't think about competition at all. Like, you should be happy that you have it. But if you, yeah, if you have it, usually it's a good sign. You're in a decent market, right? Yeah, I mean, I'd be very concerned if I didn't have no any competition. Yeah, it means you're either, no one wants it, or you're just very early to something. Correct. Which is not that common. Correct. Correct. And then, I think when you start, like, it's really good to be informed about what is happening in the market. But like, you need a vision, and you need to go and attack that vision. And if you copy everyone else, what you end up doing is being like, three months consistently behind the market. Like, that sucks. Imagine being three months behind the market, like, you're a loser. You want to be free months ahead of the market, which means you need to stop yourself looking at everyone else and say, hey, this is what we're building. And let's go and get it. And be led by your users and intuition. Yeah. And the thing that I also find that happens again and again and again is like, the other people you use some Pete with, for me, and who it be, like, are competitors. They're all quite smart. And we, I think we all kind of end up on the same ideas anyway. Like, you don't mean like, you can, I mean, you're probably getting similar user feedback. Yeah. You have access to the same technologies. Yeah. Yeah. Yeah. So we actually kind of arrive in similar places, which is super interesting. Yeah. You know, eventually, eventually you kind of end up similar places. Yeah. from the early days you're like, yeah, whatever, that's not the worry, the worry is getting customers. What stage did you start to be like, you're like, oh, we kind of have to think about our competition. What stage was it and like, why did you start to have to start thinking about that? - I've, like, so I know people that have like a competitors channel in slack and they post about the competitors and what they're doing. We literally only started one like a month ago. - Maybe that's a bad thing. Maybe I need to remember that. - That's quite, that's a later than I thought. - Yeah. - But I think that just shows you how we're kind of doing our own thing, you know? We're just biving ourselves over. - Oh my God, Jesus. - I've been. - Okay. So, but you know, you just started that channel, but like, was it, was, was there not a time that slightly earlier where this wasn't more of an issue? I didn't like, I don't know, worrying about like sealing customers or staff or something like that. Or has it not really been a massive thing for you? - No. - It hasn't. - No. - I think it's going to be more interesting if there was. - All right. So, when it comes to stealing talent, yeah. - Funny enough, the company that is stealing our talent at the moment, what, like, what happens in every company is I'll say to my recruiting team, these are the companies that want you to hire from. Right? Like, I like these companies. I like the products that they build. I like the talent that they have. I like the culture. This is the list. There's a company in London that has us on the list. Right? And they've taken a lot of our team. Not a lot, like, three, four. But like, for a 200 person company, that's quite a lot. Right? - That's 2%. So, no, I don't care. And like, you can't keep people forever. Like, it's good. Okay. Now, on competitors, I think that when I look at where I'm sitting, when I see, but like, you know, when I see that there's like another YC company trying to do like vibe coding video editing, I'm like, good like guys. I know so much about this market that like, you are literally just going to waste a year and a half of your life. So that's the other thing. And then I just think like, know what you're good at. Like, I look at my data. I can see what's good. We just double down on the things that are good. Do you know what I mean? Like Jimmy Carve wrote this book called Before Laughter. And it's actually really good book. And it was about what happens actually before stand-up comedy. Because it, you think it's, he just stands up there, but he doesn't. It's very well rehearsed. He basically says, like, if you try and like, smoothen all your rough edges, you're not going to be that special. Leaning to your, your edge and you'll be very special. Right. And I think like, so doubling down on your strengths kind of thing, double down on your strengths. I think that's it. And like, you know, like, the lead in it's in the editing world that we're in, we're really not like, we do editing, but you can get better editor's elsewhere. We're amazing at subtitles. We love the subtitling tool. It's that simple, but they just love it. Like, and we've really lent in. And our subtitles working every single language in the world. We do write to left. We do Japanese. We do Chinese. We'll animate them for you. We'll change it into any other language straight away. If you want, we can just do it like that. The animation capabilities are unmatched. It works with every single font in the world. We can, we can subtitle a feature film for you faster than you could ever imagine. We can also batch subtitle every single, you know, video you have on a platform. It's insane. It's just great. So I just, I just focused on doing that. I just, like, as long as we do that really, really well, we're fine. Yeah. I hear that. I hear that. So when someone says, "Oh, I'm like, yeah, we get to it. Like, I don't care. One final thing I want to talk about today is, unless you have something? - No, I was just, no, I think there's something interesting in that conversation around like, I think no, I think the reason why I seem unboggled about competition is maybe because I know what we do and what we don't do. - Yeah, and that gives you the confidence not to have to look around you too much. - Yeah, yeah, yeah. I hear that. So it's just one final topic I want to talk about today. You ready? Hit me, Charlie. So I think it's very interesting how AI is giving content creators a completely new angle and a lease of life to extend themselves. So an example is Lenny's newsletter, famous products podcast and newsletter and everything. He just dropped something called Lenny's data.com. It's his whole newsletter and pod archive, 300 of each plus as Mark Downfiles, MCP server, GitHub repo, and he's gone out and say, well, his massive following said, build on top of this, basically. And in me and I will give a free, like, you know, one year subscription to the most interesting thing built. Which I thought actually was a little bit, Stinge. I thought, you know, like Lenny could do a bit better than that. But anyway, that's all ramp sponsorships not that bad. - Yeah, like, you know, we know that you could do better than that. But, you, by the way, love you Lenny. Lenny's amazing. - Yeah, Lenny, don't, like, don't start anything Lenny. - Don't be, we don't mean, we don't try to be free. - Yeah, Lenny will, I love you Lenny. - And even Bartlett though, - Same Bartlett is different. - Yeah. - Coming for you. - Yeah. - Bartlett, watch out. - Yeah, he's probably watching this right now. - Yeah. - With a lot of people watching this right now. - Yeah. - Anyway, - Lenny's. (laughing) - So, yeah, people immediately built like chat bots on top of this, like a pro, a Lenny product council of like, of different people on the different episodes giving you advice for like product decisions. And another part that's all to do something like this is like Corey Haynes is like a well renowned marketer, Aegean, a course creator for marketing, released marketing skills, big repo of like Claude Park plugins, it's given away for free. But both of them have just like, you know, some people say, oh, can't tune you just like charge for this stuff, but they have just generated like masses of top of funnel attention for people who are like using these tools, building stuff with them, building on top of them. And this wasn't possible before. We have a mutual friend who's almost finished a kind of educational course. And they're wondering whether to include the sort of prompts that they used and mark down files, they're not sure. And I kind of think like this is just like an important part of educational content now is like extend yourself with AI and plugins. And you know, let me make it more interactive in that way. So I think this is a growing trend. Like if you are a content creator, think about like how other ways people could interact with your content beyond just listening to it or read it. You come across-- How can they use it? We come across delfi.ai. You know what actually this is a good example of extending yourself, right? So it creates a digital twin, is that right? Lenny is hanging out on delfi too. So like, he's probably put a few hundred grand into that, isn't he? But I heard it's a good product. It's cool. It's cool. And by the way, it follows the magic framework. It gets better as AI gets better. Yes, it does. It does. It's saying. So look, I think the reality is, anthropic and open AI, they're just going to train on your data anyway, mate. Right. So like, you know, I'm-- you're going to have to give it up. Always going to take it anyway. So actually, if you go and make-- if you go and make yourself easier for someone to do something interesting with, it's any good for you. Yeah, I think so. It's a good way to extend yourself. I think the discourse of creativity, ownership, IPs, all currently changing right now. And I don't mean that in a-- in like a bad way or like-- I mean it. I don't know. I think we're trying to work out what does the business model look like if we give our stuff away. Because the business model is, this is my IP and I keep it. And there is it. And the business model is that you pay for it. That's it. And I think what these people are very early signs of trying to understand is, what if I give it away? Is there a way to monetize it in a slightly different way? And that might actually be the best thing for the market, in general, like for IP owners. For copyright owners, like this, these guys might chart a new path. A couple of interesting threads to talk about here as well is one is that-- I think one reason this works is because they are already like quite well-known, like characters. They have very strong personal brands. And that is kind of-- I think that gives them a bit more of a moat to be confident in that. Or they have great distribution. People like them and their content. And another thing is like, it always has been traditionally like, oh, the content is free. But then you pay for the software. But for this, the content costs money. But the software is free, you know? Because they've made the calculation that that is better. I think it's quite interesting anyway. And I think we're going to see some more creative ways that people are going to leverage their content. And if you have a catalog of content, there's no reach. And there's no reason you can't do that with that now. So anyone-- it's not just new people. Anyone with a good catalog is able to bring this old stuff to life in a bunch of new creative ways. And I think we're going to see some really unusual ways people do this as well, you know? Yeah, it's really funny. I was trying to watch something on BBCI Payet. And it was like, sorry, that one's in the archive. I'm like, I don't understand. It's in a different room. Like, do you not have this on like-- I know. It's just like, go and get it. Yeah. Like, go and get it. Yeah. Well, we might need to wrap up now. Yeah, let's wrap it up. Thanks, everyone. Thanks a lot. See you next time.