(163) 3 Ways to Build a $1M Startup with AI ft Greg Isenberg - YouTube https://www.youtube.com/watch?v=_F8734qTQOs

Transcript: (00:00) Hey everyone, we have an exciting show today. The one and only Greg Eisenberger is here and we're going to talk about three ways to build a milliondoll business with AI. And you can go and do any of these right away. Greg has shown you all the tools, all the prompts, everything you need to go build these businesses. Let's get to today's show. (00:21) Uh Greg Eisenberg back on the show. Greg, entrepreneur, uh, YouTube creator, documentary maker. Not sure how you like to be thought of these days, Greg, but uh, yeah, like you're up to a lot of different things. I think on this show, one of the things you're phenomenal at is showing people the opportunity around AI. (00:44) You have an incredible documentary series called, I think, the AI gold rush. And so we would love for you to show our audience if you had to start today, how would you use AI to make a million bucks? Oh, easy. I I can do that. No big deal, right? Yeah. So, by the here here's what I can promise to people. By the end of this episode, they will get their creative juices flowing around uh a framework for how you could use some of these AI tools. I'll share the tools. I'll share how I think about optimizing the prompts. (01:16) I'll share how I'll think about building distribution. And you know, you you might make 10 million, you might make a h 100,000, but it is the framework for how the new line of businesses are being created. And I promise I won't hold back. We want it all. Let's do it. The first one I'm going to show you is a tool I built for myself, but then I ended up making it public. (01:36) So, I co-founded this. It's It's called ideabser.com. Have you seen this? Yeah, I have. Very cool. So my thesis is that the right idea, you know, people say ideas mean nothing, but it's just not true. The right not true. It's just not true. (01:54) You know, now more than ever, if you find the right niche idea based on a trend in a world where you can vibe code anything, I believe that there's, you know, life trajectory, you know, there's there's a huge uh opportunity there. So I built this because I run a holding company. What it does is it scrapes Facebook groups, it scrapes Reddit, uh platforms like that and uh it creates an idea of the day based on the trend. (02:17) And one of the more interesting features I find is there's this build this idea uh section. So what we can do today is like we can look at this idea and pretend like we're building it and like how we would build it and what what tools would we use? Does that sound like a fun? Let's do it. I would love that. Okay, cool. So, let's actually look at the idea. I think it's a grim one. (02:41) I know. I'm actually looking at it here. No, I actually I was at a funeral last week and I actually had an idea similar to this. I think it's perfect. Okay, so should we do this one? Yes, let's do it. Okay, so I'll read it for people. Cemetery management software that digitizes historical record for small uh cemeteries. (02:59) So, small cemeteries operate with paper records. Um, the plot maps remain handdrawn. Wow, that's crazy. I didn't realize that. So, basically, pricing ranging from $30 to 200 a month based on cemetery size. MVP features a web dashboard where managers could upload photos of paper records that convert to searchable text through OCR. You know, market seems big, $320 million. (03:24) So, and it gives you like, you know, actually your your value ladder, like your stack, right? So like what lead magnet you want to do, maybe a digital cemetery transformation guide, the cemetery helper introduction package, the core offer which is a cemetery helper pro. (03:43) So this is like using AI to basically, you know, study people like you can see Russell Brunson who's like an OG, you know, kind of funnel guy, right? So we're going to do this idea and it's like, okay, how do we actually like build this idea using AI and how do we actually market it? So you can see on the left hand side there the problem that people make or the problem is that people don't know how to prompt the the tools very well and they also don't know what tools to prompt. (04:07) So what this does is it shows it gives you prompts for ad creative on all the different platforms. It gives you uh prompts for the brand package, the landing page, the content calendar, the email funnel sequence or funnel system, the email sequence, the lead magnet, the product. (04:25) All right, we've got our prompt of the week for you. This one's designed to help you generate million-dollar ideas using AI. If you're feeling inspired after hearing about these tools and strategies, this is your next step. Scan the QR code or click the link in the description. Now, let's get back to the show really quickly. (04:43) Is it um how are you doing this? Is this like a bunch of different prompts? Is it one prompt that's kind of like pulling data from like Plexi? Maybe just like a little bit about you don't have to give away your kind of you know your magic but what is what has AI allowed you to do that kind of helped you collapse all of this into a single place. (05:01) I would love to give away all the magic cuz to me like you know you asked me like what do I do like I I I I consider myself in the light bulb business you know if I can help people's light bulbs go off that to me is a successful day. So I'll tell you exactly how we do this. We for each idea we have all that data that data gets downloaded. We we funnel that data and context into these different sections. (05:26) And the issue that people have when they're, you know, a lot of people vibe code or they'll use chat GBT or or or code or whatever and they don't get the result they want. So they're upset at they're like, "Oh, this is useless." But what's great about this is it gives the right amount of context. Yeah. I say context. Yeah. (05:46) So, so it's all about the context as you know you guys know this, right? Yeah. Yeah. Um, so this just is like the way I see it this is like training wheels for anyone to prompt. So, you know, how I would start with this is I would I like starting with a PRD. So, if I was there there would be two options. (06:08) Either I would start with a landing page, put up a landing page and then start, you know, generating traffic to that, or I'd start with the product requirements doc, the PRD, and I would use like a V0 to actually build the product. Oh, interesting. Okay. So, there's a bunch of different like vibe quote unquote vibe coded tools. (06:27) What's your preferred tool and why, you know, for the next kind of step you're going to to kind of build that landing page? So, uh, for landing page, I mean, they're all kind of similar. Some people like Lovable more, some people like Vzero more, some people like Replet more. Replet agent has like a I don't know if you've seen this. They basically have you can like put in the PRD and the agent you can you can have it do like a fast version in like a minute or you say like use replet agent and it'll build out everything, but it's going to take 20 minutes. Yeah. Um, so the new agent is insane. It's like one (06:59) of the more autonomous agents in the market. Yeah, actually this is interesting. Reddit, how how complex of an app can replet agent make? Oh, this is a year ago, so that's like 40 years in AI land. It's true, but even Yeah, but even, you know, this person says like it can build apps that are fairly complex. (07:19) I just finished building an RSVP app and that was six months ago. So, if you're just trying to build a landing page, I like pick whatever whatever you know brand feels good to you. If you're trying to build an app, that's where tools like Vzero, Lovable, and Bolts, you know, they're better at the front end, you know, they're getting better at building full apps, but I would say like if you're, you know, a non-expert non-technical person, I would use Bolt Lovable Vzero to get something out there. But if you're technical, I would use something like claude code or cursor. (07:53) I have an app for the most part minimal viable version built. And when I got the 30,000 lines of code, I needed developers. I have two developers and uh it started just get what what actually it just they start to kind of lose I don't know track of the code and start to like integrate weird things into the codebase. (08:15) And so the mix I had to use was real developers and cloud code plus one of those apps. So they definitely still have like some scalability issues. I I agree. And I think like it's okay. You don't need to be a you know Sam Alman says a one person billion dollar company. Like we're trying to make a million dollars here. (08:33) So you know it's okay if you need to hire, you know, on Replet. Yeah. A dev, right? Replet has something called Replet bounties. if you don't know what to do, uh, you can literally just have a task. Um, I'm not like associated with Replet or anything. I'm just, you know, I think this is interesting. Yeah. (08:50) Um, I think it's interesting like because I've I think we've all been there where we're vibe coded something and we're like, I'm tired, you know, and like I just don't know what to do here. Well, it's the the interesting thing of like Vibe building anything. I I I include uh graphics like you're going to go to build a landing page. Actually, these tools are great for landing pages, but the creating part is magic. (09:11) The editing part literally makes you want to set yourself on fire. Like, uh, when you start to really have to do editing in the codebase, that's where you start to lose all that magic and you just start to say, I'm I usually have I've had to start from scratch. (09:27) Like, it's easier just to start all over again than actually try to use these tools for editing. I think that's still the problem with AI today is like huge the consistency and reliability of the editing is nowhere near good enough to ship a lot of production ready things. I I totally agree and I I also think you know going back to the topic of this like how do we make a million dollars with AI? I probably wouldn't build a full application to start realistically like I would probably start with a landing page. (09:54) I would honestly probably start by like grabbing some of these prompts around like go to market like be you know there's the saying measure twice cut once and I think that we've all been so trigger happy to like vibe code but like you need to understand like what you're building before you're building or else you're not going to make a million dollars you even if like replet agent builds a thing right so you know I would do I would use something like this right so let's see you know go to market strategy. (10:22) So, it's going to give us a prompt for a go to market strategy and then I I would use probably chat GPT. Yeah, that's what I would pick too. Yeah. So, I would probably do that. And it has this longass prompt that says, "Create a comprehensive go to market strategy that provides a clear road map for launching and scaling this business. (10:44) Focus on the most effective channels and tactics based on the target audience and competitive landscape. And then you have to be like the editor and like you you can't just because you got the idea browser prompt doesn't mean like don't don't accept it as like 100% perfect right like you need to read this and you have to agree with it. (11:11) So by the end of like some of these prompting these business prompting that you're going to do make sure that you believe in what you're building. You know what I mean? Yeah. This is where like you have to you have to have some beliefs. You have to inject them into the strategy and the prompt and you got to iterate and go back and forth with this for 20 30 minutes to get to a point where you're like, "Oh, this is actually good and I could see this working. (11:30) " So, what do you guys uh you see my screen, right? So, what what do you you know, you guys are are worldclass marketers. So, what do you think of what what what chat GBT spit out? Like, do you buy it? Do you think it's you know, right? I think the market analysis is pretty good, right? I think the core question where you would uh where marketing is really going to come in is in the next couple sessions, right? Like how do you position it and what are the actual channels that you're going to go after to build a business (12:00) like this, right? So, let's see what they say. See what Sam is. I don't know if it's trying to be tongue and cheek but that I love the positioning statement which is for small cemetery manage still buried chat is a creative marker that's I love that line buried in paperworks a and what what I what you need to know here like this is where like doing some basic understanding of the audience right like this is a very non-technical audience right and so you need to make sure that the positioning and the language is as basic as possible (12:33) and it's literally just like we're going to turn your plot drawings into digital records to to help you continue to build and manage your cemetery business, right? Like it's got to be very basic. It's pretty like if you think about it, right? You've spent probably 15 minutes going from business idea to full analysis to go to market strategy. Go to market strategy is pretty good. (12:59) uh like even from what you're shown on screen right now again you have to be able to I always think with AI AI is going to do 80% and that you do the last 20% but you have so much time to make that last 20% much more impactful um and so you can really refine this but I think it generally it looks pretty good the one thing you just mentioned I think is important for our listeners Greg and you are you know a seasoned entrepreneur one of the things that I think the average person the mistake they make is they kind of build too much of the product prior to marketing or getting feedback on that product. You did mention that. So is there any (13:34) anything in here around like a phased approach where you would try to build some initial feedback like your waiting list try to get some initial user traction just to get feedback on the minimal viable version or is this really just your fullced marketing plan? So this marketing plan looks like it's the full scale thing and if I was like what I might say this is great but I don't want to build the MVP um all at once. (14:05) What I want to do is validate that this is a good idea so I don't waste years of my life building building digitized symmetry records. Yeah, people don't want. So I need you to come up with a, you know, marketing plan for a more, you know, more actually like lean startup approach to building this landing page, weight list, etc. Dude, no one says the word, no one even says lean startup anymore. (14:37) Most startups are just lean startups by default now, right? Like do you remember when that book came out? Yes. It was a transformational book for entrepreneurship, but it seems like it might as well be like a hundred years ago. Wow. Kind of. It's insane actually. It's kind of wild. I hadn't thought about it. (14:54) I had even thought about that book in years. What What do you think replaces that if that was the lean startup given the way you can build businesses today? What is like the version of that for entrepreneurs today? I mean, this is the new lean startup which is like you find an idea based on data. (15:14) Like you find an idea based on data that AI basically scraped based on real conversations on the internet. You then build that idea, but before you build it, you validate it. You have to be like a decent business marketing person to basically edit that. So you have to get smart on on that. But you go and do that and you research it. Like you spend a bunch of time researching it. (15:43) And when I say a bunch of time, I mean like three hours in AI time. That's a lot of research. Yeah. You could Yeah. Yeah. It's a whole building the entire app. Yeah. And you use, you know, Chat GBT and Claude and Manis and companies like that to go and do that. And you and and by the way, use multiple of them. Like I just use I use chatbt. Great. Now I've done this with chat. Now I'm just going to go and copy and paste it into claude. (16:07) Okay, now you know Claude said one or two interesting things and then I'm gonna go to Gemini and then I'm gonna go to Manis and I'm just gonna do that. That's the three hours. And then by the end of it, I basically but by the end of that process, what I do literally is I have a notebook and I write down with a pen and paper what the idea is, what the marketing strategy is going to be, how am I going to grow it, what is the MVP look like before actually going into Vzero, doing a PRD, all these things. Now, I don't know if (16:44) that's a boring way to do it. Like, I know people want are listening to this and they're like, "Oh, I just want to like my my fingers are burning to vibe code." It's like, "Slow slow down, my friend. Slow down. Spend time over here." I think what you're really telling people, Greg, is like, you can go run and vibe code stuff, but the likelihood that you waste a bunch of time building something that nobody wants and make little to no money is pretty high. Right. (17:14) And you're but we're here we're doing a whole show about like how you would make a million dollars with AI. Part of that is like you have to validate the actual company and idea in a way that makes sense. You need to have a strategy that you can then go get some feedback from potential customers on before you go and build everything. (17:32) because it might take three, four, five, six ideas until you get to the one that's really the million-dollar idea that you have the unique skills to go build out and do, right? Like the you can't just go instantly into like making something that's going to be a million-dollar business. Still, the truth is your first idea is never the million-dollar business. Never. (17:56) You know, there's a good chance if you're lucky, it's your second idea. It's true. You're lucky. You know, I'll give you, you know, a famous example of of uh of a lucky uh a lucky second idea was the story of Instagram. Before Instagram became famous, it was an app called Bourbon. (18:16) And Bourbon was a Foursquare competitor when you you remember you used to like check in to different Yeah. Tell people where you were. It was so weird. So So weird. You used to check. So strange. And one of the, you know, popular features on Bourbon was photo, you know, posting a photo to a location. So, I think it was actually Kevin Cyram, the founder of Instagram's girlfriend, who was like, "You guys should just focus on photos. That's what people care about. (18:37) " They built it in a short amount of time, new app, Instagram, and it took off on the first day, got 25,000 users. So, they needed to go through the the messy middle of building Bourbon to get to the place where, okay, the real insight is around photo sharing. (19:00) And I think if we were gonna actually build this cemetery management software business, which by the way is the most grim idea. So I don't even know if I can personally do that. But but but that's also why it has a higher likelihood of making a lot of money because there a lot of you're going to have less competition, less people who'd want to do it. Right. That's right. That is true. That is a really good point. (19:20) So and we are trying to make a million dollars. So I I I take it back. So if we you know I I I think that through this process around like here's the customer interview guide that you can get through AI the competitive analysis through the go to you know actually putting in these prompts and on those four or five different platforms and then using a notebook and why I I use a notebook is it it just like I don't know why it like clicks in my brain more when I write things down but writing down the insights there spending (19:49) like a day doing that and then the next day building the landing page, building the brand package, building the ad creative, but we might realize that, you know, this idea actually was close was the right niche, but actually the MVP is actually quite different. Yes. What what these what these cemetery owners need is actually very different than what this idea is. (20:14) Yes. You know that the it found the right problem, but the idea needs to evolve to better fit the problem. Right. Yes. I want to show one more thing around I don't know why more people don't do this but you there's like this feature here which goes through different niches scrape and just scrapes Reddit like uses AI agents to scrape Reddit and Facebook groups that's and just tells you literally like what are the pain points in these different subreddits. So for example like backyard chickens and urban (20:48) homesteading. So this is like a fun like I mean we we are covering the best topics. Cemeteries, chickens, homesteading. Yeah, this is crazy actually. Um but it's it's that's like kind of the point is like you probably would haven't thought about backyard chickens a lot. Um but apparently it's a high pain level and the most effective people are these new hobbyists. (21:20) So you can you can use AI to look at what are the underserved segments, what are what are the pain points that people are facing and it literally has like you know someone on Reddit like lost three hens last night. This is getting expensive and heartbreaking. Are there any products that actually keep raccoons out or am I wasting money? Wow. That's the cool part about AI is you get data. Yes. And especi Right. (21:44) And then you can use that data to give you an insight around what to be building. And also, by the way, for the people listening to this, I would just like copy these user quotes, post it into chat GBT or Claude or whatever you're using and be like, make a landing page that speaks to landing page copy that speaks to these user quotes. I don't know why more people don't do that. (22:07) Well, and by the way, what's a there's a couple other things that I think are really interesting on this particular like Reddit scraping example, Greg, which is one, if the channel's big enough, it's your whole marketing strategy, right? Like you're just taking the the core problems from that channel, engaging with the channel, and that community is your target market. (22:23) And maybe that's not a million-dollar business, but it might be a $100,000 business. Yeah. Right. Which I which I think is interesting. The other thing for things that might be I know we talked a lot about vibe coding, but you can make money a lot of different ways. Like for something that for example that stops raccoons from getting in chicken coops, it might just be like there's a bad solution in America and somebody in like Sweden has figured this out. You just need to like import it and sell it here, right? Like there's very basic things (22:47) you can do to make money once you understand like the problem and demand in the market. Some of my one two of my favorite entrepreneurs, I can't recall their names now. There's a great uh write up on them and one of the publications are the German brothers who have made billions by just taking US businesses and replicating them in Europe and uh that's that's what they had did right they they didn't have to validate anything because they were validated in the US and they just replicated them really fast. the the other um the other business that you're you just kind of (23:18) mentioned something there that I think is actually really important, Greg, which is like why why do more people not plug in their customer language to craft their content? The other business I think is a really good business um and we kind of showed a way to kind of manually do this but someone can automate it is creating a digital twin of your customer. (23:41) And so the way you would do that is you would create an app and that app you would plug in things like your gong calls and all of the internal information you collected about your customer and then it would collect all of the external information you collect by your customer. Then anytime you do something to market to that customer or sell to that customer, you could have your digital twin edit it for you and review it for you and then actually replicate it in your in your customer's language. (24:04) Uh and we kind of showed how to do that in cloud projects, but that that is a really good use case. I think mo most uh companies have marketeteers and people that don't do that. They just like guess. That's why everything sounds like jargon, right? Totally. And that's why the average conversion rate on the internet is 1 to 2%. Like no one talks about that, but 99 or 98% of people are coming to websites and are literally just leaving. Not like not finding anything irrelevant to them at all. And we're like okay with it. (24:36) We're like, "Oh, yeah, that's the normal. It's one to two percent." And like, but I think that opportunity is disappearing. So, what's I think here's what's going to happen to the web. I think the website is actually and are going to we're going to look at the website and say, "Wow, that was like a really antiquated thing to do because a website is a static library of content that you have to like read through and research. Chat is a much much better version of that. (25:00) " And and so when you look at the data, you know, on average someone coming from chat converts like anywhere between four to 12 times better. Now people look at that and they think, "Wow, that's a great opportunity." But what's the reason that is is because all of the research is happening away from your website. (25:17) And so the the one one of the business opportunities is what is the modern website in a world where most of the research does not happen on your own website. And I actually think your website will probably morph into just a a kind of some sort of multimodal agentic seller because you come there fully researched. And so you're not looking to navigate and consume content. (25:41) You're looking to have a real interaction with someone who can actually demo the product to you, you know, get you the last 20% of things you can't get from chatbt. And Greg Brockman had a the co-founder of Open AI had a fascinating quote during the week where he said he feels like the website like Chach has shown a light on the website and just like how redundant it is. (25:59) And so there's there's another business opportunity on like what is the AI version of a website and I think that's a fascinating thing for someone to like build around. Well, I yeah, I think uh Open AAI has made it clear that they want to make Chad GBT the operating system. Yeah, (26:23) they want Correct. Yeah. Yeah. So, the the announcements from Devday, you know, if you're looking at my screen around these apps that exist within Chat GBT, like you know, you might not never go to ideabrowser.com. You might just at idea browser and just be like, "Hey, I'm thinking about an idea in Cemetery Space. (26:43) like hook me, you know, what do you think? You know what I mean? Go get the frame the problem an opportunity for me. Yeah, exactly. So, I'm just one of those people that I'm like, this is where the world is going. I'm going to make this beneficial for me. You know what I mean? Like, I'm just going to go like I'm just going to go build a bunch of apps and and see what happens, you know? Like, so but like, okay, I'll answer your question directly. (27:12) Is it a good thing or a bad thing? Obviously, it's not a good thing because we don't, you know, the old way of the web was you control the experience and we're losing control to Open AI. Open AI is going to be able to control the experience. And you can see that when I added here, there's one, two, three, four, five apps that came up. Booking.com, Canva, Corsera, Expedia, Figma. Someone at OpenAI decided that booking. (27:39) com for whatever reason. I guess it's alphabetical but still like I don't see idea browser up here. You know what I mean? So the you know the same way that the app store you know there was a lot of early winners and if you know you knew the editor of the app store you got back in the day you would get thousands and thousands of downloads. I think the same is going to be true with the app store on chat GBT. (28:06) They're also going to have an editing approval process. But what do you got to do? Anyways, going back to the topic, you want to make a million. Another way to make a million dollars is build apps for chat GBT apps. I don't know why more people aren't like, why are you even listening to this? You know, you know what I mean? Like my my hot my hot take though is that the cemetery thing is better than the chat GPT app thing. Yeah, that's my hot that would be my that my my my burning hot take is yes, the chat GPT opportunity is (28:40) there for sure. But there are still so many problems that are completely not served and solved in this world that like if you just go and find one like the cemetery problem that the technology is so good and easy now to address it, you can make a million dollars. I hate to agree with you. You mean you could please disagree with me, but that's like that is probably I love AI. I love chat. (29:05) I love everything, but I I I mean, if you were just like, "Go make a million dollars as fast as possible." Like, I'm going to go start the cemetery business and uh going to grind it out. How would you figure out what app to build for OpenAI? Like, what's a example that someone should like how do you build differently, I guess, for OpenAI versus the Cemetery app? Like, is there a difference in how people should think about that? So, I came up with this little framework for coming up with chat app ideas. (29:34) Um, so and and what the opportunity is and we can actually talk about a couple of these startup ideas if you're interested. But, uh, you know, the goal is to build SDK native apps that chat GBT automatically surfaces when people ask for help like help me get a mortgage, plan a wedding, or launch my newsletter. That is the opportunity. (30:01) If you want to build a CH chat GBT app, you want to basically you want to be the first thing where it says, "Help me get a mortgage." Just like how in the SEO days, if you were the f, you know, Rocket Mortgage or whatever it's called, help me get a mortgage, you're number one. Like, you made literally billions of dollars. So, there's going to be an opportunity. (30:20) Keep in mind, there's 800 million people who are weekly active users to this, right, to Chat GBT. So as at you know if this works if chat GPT app works and they don't like it it doesn't go the way of custom GPTs which we can talk about but uh was a massive failure. If this works, then if you want to create a, you know, not a $1 million idea, but like a $10 million plus idea that has like a bit more risk of not working, then I would say try to own a high value keyword, solve a specific high intent request that chat GB chat GBT can't handle natively, (31:02) and have a custom UI that's visual and and and like I'm sure you saw uh you know they really want like visual like this is Zillow, right? It's like a visual UI. So I mean maybe if we want to build a cemetery app idea it's like you know my grandmother died and like need help with funeral planning and then it like pulls up a UI like this. Uh and the you buy a plot. (31:34) Yeah. Okay. So you have to monetize in some way. So I guess you would you would build that app in a way that allowed them to look at things, but as soon as they needed to do an action, you would monetize the action by them having to click through your app or you can monetize in the app. (31:50) I actually didn't understand that part. Yeah. Yeah. This is like this is a website. Like what is the difference between this and a website as far as I'm concerned? You know, it just instead of it it kind of you know what sucks about it is h don't openai don't hate me on this but what sucks about it is the web the web was an open protocol or is an open protocol right if you want to create a website you don't need permission from anyone to create a website in this environment you do need permission from openai to approve your app so basically they they are the curators whereas in the open web there was no I (32:30) guess the algorithm was the curator but like you're not having to have a human have a the human doesn't have oversight. This comes back to kind of my point of why the cemetery business is a better business than building like an app on chat GPT is because you have full autonomy over that cemetery business. (32:49) When it comes to chat GPT, just like you said, Greg, Open Eye has got to like let your app exist on their ecosystem. But the other thing is you either have to provide functionality that's limited that's available in their technical infrastructure or you have to send people off to your app on on its own URL, right? And that then becomes two things to manage. (33:13) It then becomes kind of a user experience challenge where your top customers are going to want to get into a level of detail in most of these use cases where they're going to need a standalone experience because at least in the short term what's capable within chat GBT isn't going to be enough of what they what to do. (33:31) Uh but isn't the business opportunity for chatbt that there's just going to be a new way that people interact with stuff and so like your old apps don't make sense in a world where their expectation like the consumer is going to want something different. So like the the kind of symmetry business has to be customuilt to the way chat GBT users will use chat GBT. (33:50) And so maybe the way this works is all of the stuff you can there's like a monetization part where you can do most of the things in chatbt, but you have to like go to the app to like complete some sort of tasks. But anywhere that has 800 million weekly active users, probably soon to be like double that at the rate that they're growing, it feels like that's it. (34:15) You have to it's going to be a forcing function to build into that ecosystem. Like there is a huge opportunity if you can figure out how to build apps in a way that those consumers want to you know consume. Same way that the app store is very sim like the app the app store top list are all curated. They're not I guess I guess my point is I don't know how this is that much different than the app store where you had the standalone mobile app but you still have a website and a more and in many cases a more detailed app beyond the mobile app experience. You know that I think that's the that that's the opportunity. Um, but Greg's point is I think your point here (34:46) is on the topic of building a million-dollar business, it's very early on chat GPT and there's not a ton of apps and so there's a lot less competition and that makes it worth playing by their rules. Yeah. Right. I mean, I'll give another You want to make a million dollars. Uh you another way to make a million dollars with chat chat GBT apps is you build chat GBT apps for companies like uh we we run a design firm that works on AI products and we announced that we're building GPT apps for a lot of the you know for Fortune 500s and (35:25) like there's there's infinite demand. So like another if you're a developer I would say like build these apps for companies too right maybe you don't have relationships with Fortune 500s but you know the I think that you can easily sell some of these just like how there's opportunity you know there's web designers for websites and small businesses and people make you know millions of dollars a year doing it you can also start building these little apps and charge a couple thousand dollars or you know even better charge them like $50 a month or $100 a month and build a portfolio of it. it is what (36:05) is what you're if you're if you're going to like watch this episode and say I want to be early in the the app the chat app store chatb app store I want to build a product in here is how you decide what to build like the idea different if you choose to build it into chat GBT versus how you would just build it anyway like do you have to have a different is like does the idea differ like you have to have a different way of validating your idea it's it's the way I see you know apps within chat GBT is very similar to how like mobile versus web, you know. So if (36:39) if you think of like there's there's mobile first companies and there there was desktop first companies like Facebook for example couldn't really figure out getting onto mobile and that hurt them. So they had to buy Instagram which was basically the mobile first version of Facebook. So I think that like at the at the core of it, you look at Facebook, it's it's a photosharing app and then you look at Instagram, it's a photo sharing app, but it's it's a nuance difference. So, I think the apps that are going to crush it on chat GPT (37:11) are going to be chat first, very like focused on this experience and how do I natively build this UI so it is the best possible experience which might just be like a little bit different than how their web experience works. Um, but I think it would be hard for someone to like if we were working on that cemetery idea, I think it would be hard for us to build the web version and build the chat version at the same time. (37:44) I would recommend someone just focus on either one or the other. Could we just can I get your thoughts on is there a million-dollar idea on around the use of sort of 2 or V3 like these kind of new creative video video clip tools because some it's like I love to create things and they're the tools that I freaking love. (38:11) Uh now I don't have access to because I'm a sad European and we're not allowed nice things but I'm curious to get your thoughts. That's untrue. You know, you guys, you have the quality of life. So, come live. Come live where I live. All see, all I heard is that less AI is higher quality of life. Is that that was my takeaway? I have a friend, his name is PJ Ace. Have you seen some of his commercials? I've seen them. His commercials are really good. (38:37) So, this is a guy who's gotten 300 million views on his AI ads. To me, he's like the top AI ad man. Um, and he comes from like the traditional like I wouldn't say Hollywood but like traditional video making ad making and he just applied his skills to using VO3 and and now Sora and his Okay, so like I'm just I'm I'm not there's no audio here. (39:08) I'm just like I want the audience to see like how look look at this how this looks like a Super Bowl commercial. You know what I mean? So look at this. This is insane. He's done this all all these are all via AI video and um so where's the opportunity to build a million-dollar business with AI video like Sora and V3 versus by building audiences with like if you can consistently push out content like this then then you go to something like idea browser or come up with your own idea or find a trend and then go and sell them something because you've done the hard part around finding an audience, right? I have this um framework I build all my businesses (39:54) on which is like I call it the ACP funnel and I think I've talked about it on here before. Start by building an audience then you know convert that audience to community and then build that product. So I think I would start you know by building audiences, creating faceless accounts, creating brand accounts, but create super high quality con videos and I can tell you how how he creates this because I know how he creates this. (40:24) Um I'm happy to to share that like what the Yeah, tell us. Yeah, Kieran's like please. Yeah. So what he does is he writes a script with chat GBT and Claude. So he fully writes a script. Um but it's all with AI. He has a framework for, you know, coming up with scroll stopping scripts. Largely that has to do with juxipositions, being funny, like highly visual, and then using characters that like people recognize. Now, I know someone's going to say like, well, that's like illegal. (41:07) You can't use, you know, Sam Alman's likeness. Well, he licensed it. He He allowed people to use it on store, too. Oh, okay. I didn't even know that about him. By the way, this could be a great ad for our cemetery business. Yeah, exactly. I was just thinking it's pretty sweet. (41:24) But what's hilarious is you would kill Sora 2. You would kill it on Sora 2 with like cemetery AI content, right? How much Sam regrets the decision to having his likeness be open on Sora 2. I mean the beauty is I mean probably a lot um because he's like a meme but you know you don't have to use Sam Alman you can use pe people in the public domain like Plato and Socrates or things you know things like that and so he writes the script and then he uses a tool like free pick. (42:09) Have you seen this? This is one of the tools where you It's a kind of like an all-in-one image generator. So like um Okay, we can do it live actually for fun. Oh, then he creates the images here to put into the video to get consistency. Yeah, exactly. Ah, so let's just use your your sweet hair. I my you know my hair. I'll take that compliment. I mean it is good hair. You should be proud. (42:40) Actually, should we use should we use one of you? Let's use one of you. I'm going to use Kieran because I have a doodle for profile. You still got your um Yeah, man. You know. All right. So, we're gonna upload this photo and use image. And then you could like let's say we wanted to make a video around this like let's say have him eating a banana or something. You can pick the model here. So, oh wow, I have not seen this tool. (43:15) Yeah. So you start by the model and you pick the right model and then what he does is he would pick let's say Nana Banana which is probably the best right now and he um he puts it all on like a Figma of like all the different frames. So he creates the frames first uh and then takes the images for each frame and makes them into clips and then Okay, look at Yeah, I look like Daniel Mr. the photographer. (43:49) So he'll take this and then he'll basically put his script plus storyboard in one like Figma and then once he has he he has the frames he'll then use a video generator to go and like we can take this for example turn into a video and on a tool like this you could again like pick the different model, the right model for it. (44:24) So you can see here all the different models like one two right now, which is Alibaba's um open source model is like really good right now. So you might want that. But you know, if you're trying to do more of an animation, maybe you want something like cling, right? So you pick the model that works best for the type and style uh that you're trying to trying to figure out. I love that free pick. That is a cool tool. (44:49) I'm I'm actually gonna spend my weekend using that tool. Oh yeah, Kieran's now gonna be on free pick all weekend. Thanks, Greg. You just you just killed his family time. So your family, you know, your $1 million idea here is like get really good at scroll, stop, and videos. Build that audience and then you can kind of build a product to market to that audience. 100%. (45:10) Yeah, you could probably like back in the day people just built handles with millions of followers that were memes back when like memes were newer and they just sold the handles. So I think you actually did say that somewhere where at some point people may just buy over like sort of two accounts or like handles that have tons and tons of viewers for their market which are just these kind of videos. (45:33) I saw so you might not even need you might not even need to build a product. Yeah, I mean, you know, right now is a race. It's an AI gold rush, but it's also a gold rush. It's also an attention gold rush, right? So, if you can build Yeah. If you can build that attention, then you have an unfair advantage. (45:54) I saw this tweet the other day from Brett, who runs a uh he's a designer. He said, "I just made a Tik Tok account, posted a Sora video of squirrels jumping on a trampoline. 7 million views, 377,000 likes. will hit 20,000 followers today. Now, this is like just funny content, so it's easier to get likes and easier to get followers. (46:20) Like there this is there's no niche, right? So, I don't I mean, it's awesome that he was able to do this. That being said, like if I'm thinking about building a million-dollar company and I'm trying to build an audience, I want to think about what kind of videos I can create that build an awareness in the niche that I care about. This episode is awesome because I've been looking to get much better at the video tools. (46:40) I've been using Flow and V3, but uh Freepick looks like the tool of choice for me. Yeah, there's Freepick and there's also Enhancor.ai. Those are the two ones that I've used. Well, if you can we g we gave multiple ways to make a million dollars. We talked about how to use idea browser to build a million-doll business. How to build a million-doll business by building chat GPT apps and then how to leverage AI video tools to basically build an account or a community and either sell that or use that to sell a product to make a million dollars. (47:12) Like those are three great avenues to a million dollars. And as normal, Greg, you're killing it. And you are, I think, the idea guy of all idea guys. I love that you're on ideas matter as like a core corner because I do think they matter a ton and I think we gave people some things that are hopefully gonna get their ideas going as they watch the show. I think I hope so. (47:36) I hope so and I think uh I don't ask for much but the only thing I'll ask for is if you like and comment this video. So uh I mean we gave you million dollars of ideas, right? Hit like, hit comment. Go check out Greg's amazing YouTube channel. We'll link that up below. So, go subscribe to that and we'll see you really soon on Marketing Against the Grain. (47:56) This data is wrong every freaking time. Have you heard of HubSpot? HubSpot is a CRM platform where everything is fully integrated. Whoa, I can see the client's whole history. Calls, support tickets, emails, and here's a task from 3 days ago I totally missed. HubSpot grow better.