He Built a $25,000 App for $250... From His Couch

Content

00:00:00 -> 00:00:15 : what you do at Google during the day and just kind of like your foray into AI. >> For sure. Yeah. I am the UX manager of the data analytics team on Google Cloud. >> I don't know who said it but someone
00:00:12 -> 00:00:26 : said the future of the new coding language is English. You create Home Genie. What is it? How you came up with it? And what did you use to build it? >> Yep. As you mentioned earlier, I'm in
00:00:24 -> 00:00:35 : the process of like moving to Arkansas from Texas and that involves >> Did you get a lovable subscription for this or did you first start? >> I started with the free version, built
00:00:34 -> 00:00:42 : out a landing page, you know, I was like I didn't know any I had my first time user I started using it like 2 weeks ago literally. >> So what would something like this cost
00:00:41 -> 00:00:52 : for someone to build? >> So in total I've spent about $250. >> It's not bad at all. >> So I think the future's bright. I'm looking forward to continuing playing
00:00:50 -> 00:01:03 : with these tools and seeing, you know, how my little project goes. And I'll probably spin up more cuz I have lots of ideas and before it was all about time. >> What's up y'all? Welcome to the channel.
00:01:00 -> 00:01:18 : This is your boy Amit and today I'm talking to Isaac who's going to show me how he built this mega site on Lovable. The same site would cost you upwards of $25,000 to build. If you were to hire
00:01:14 -> 00:01:27 : someone, he is able to do it for $250. So, with that, let's get started. >> I have Isaac with me. Isaac works at Google. Just moved to Arkansas from Texas. Isaac, welcome to the show. How
00:01:26 -> 00:01:38 : are you doing, man? >> Thanks, EMTT. Yeah, doing great. Appreciate you inviting me on. Excited to chat with you today. >> Yeah, I stumbled upon your reply and
00:01:35 -> 00:01:46 : your app on this uh long Twitter post that everyone was commenting on. Uh someone said, "Hey, what cool tools have you built?" And then you jumped in. and you're like, I built Home Genie. So,
00:01:44 -> 00:01:57 : we're going to get into Home Genie. You are a Wii coder, but uh let's talk a little bit about what you do during the day. All of us Wbeers, we are hustling at night, weekends, just building cool
00:01:55 -> 00:02:10 : things, right? But let's start about what you do at Google during the day and just kind of like your foray into AI. >> For sure. Yeah, I am the UX manager of the data analytics team on Google Cloud.
00:02:08 -> 00:02:18 : So, we are responsible for big products like BigQuery, if you've ever heard of that. the >> yes >> dataf flow pub sub data proc there's a
00:02:16 -> 00:02:28 : whole bunch that we kind of manage in our portfolio um I have a team of designers and we're kind of responsible for all the user experience we've been diving deep into AI over the past year I
00:02:26 -> 00:02:39 : would say and pretty much everything we're doing revolves around it now and so as part of that I wanted to just kind of explore outside of Google and see you know what is going on because I hear all
00:02:36 -> 00:02:50 : about vibe coding and I come from more of a traditional web development design background ground. Something could take weeks or months to like get a prototype out there unless you're really
00:02:48 -> 00:03:01 : ambitious. And now it just seems like the all those barriers have been removed completely and like anybody can just take an idea to a working app in a matter of hours or days. And I just find
00:02:59 -> 00:03:11 : it fascinating. >> It's crazy. Like I would be sitting just anywhere, right? And a thought comes to my mind and I'm like, "Oh, let me just put this in lovable." And then 10
00:03:09 -> 00:03:22 : minutes later, boom. It's like my vision comes to life. It's a wild time to be alive. I don't know who said it, but someone said the future of the new coding language is English because
00:03:20 -> 00:03:33 : that's how we are coding now. All right, let's get into your app. I'm going to share my screen and let's talk about Home Genie. How you came up with it. I think I know a little bit, but for our
00:03:32 -> 00:03:43 : viewers, >> talk to me about this. You create Home Genie. What is it? How you came up with it? And what did you use to build it? Yep. As you mentioned earlier, I'm in
00:03:41 -> 00:03:56 : the process of like moving to Arkansas from Texas. And that involves selling my house in Texas and buying a new house in Arkansas. And so, as part of that uh whole journey, I've learned a lot about,
00:03:53 -> 00:04:06 : you know, the pain points in real estate and like buying and selling houses. And being, you know, product designer, developer in my career, I look for these kind of opportunities. Like that's where
00:04:04 -> 00:04:15 : the gold is. If you can find major pain points and problems that lots of people face, now you've got, you know, a market to cater to and then you just got to figure out how to like build something
00:04:14 -> 00:04:26 : that can help solve their problem. So, I kind of started building some small tools in here to help solve my own problems. Like, what's my new mortgage payment going to look like? And what
00:04:24 -> 00:04:35 : should I even sell my house for? I don't even know what it's worth. And where's this new place I'm moving to? Like, I need to learn neighborhood. And all these questions, right? And so, you
00:04:33 -> 00:04:46 : know, Redfin and Zillow and there's actually a lot of tools out there that can do bespoke things, but I was like, they're all scattered all over the place. And also, like I don't see a lot
00:04:43 -> 00:04:57 : of AI stuff like built into these things and working with my realtor. Like, she's great, but she had to do, you know, lots of manual like work creating listings and typing this stuff and we had to edit
00:04:55 -> 00:05:07 : it. And I'm like, better ways to do this stuff. Like, AI is amazing at this stuff. And so, I was like, I'm just going to go build something. And then as I mentioned earlier like at work we're
00:05:04 -> 00:05:18 : actually doing a lot of work with AI you know infusing it into all the tools and products we build. And so I'm like let's use this as an opportunity to kind of explore the tools that are available and
00:05:16 -> 00:05:28 : look into this whole vibe coding thing. Like I've been a skeptic I'll admit like I was like this is never going to work. Like you know it's just you can't possibly sit on your couch and build an
00:05:26 -> 00:05:38 : app and expect it to work and everything. And surprisingly, I've been proven wrong now after doing this. Uh I think, you know, Lovable has done a great job with their interface and also,
00:05:37 -> 00:05:45 : you know, other tools. I was like, you know, I got four kids. I full-time job. I don't have a lot of time. Like, if I had to do this the traditional way, I probably would have
00:05:45 -> 00:05:56 : >> Oh, yeah. >> gave up and been like, I don't have time to go do this. But I was like, you know what? I'm just gonna take it on as a challenge and see what vibe coding is
00:05:53 -> 00:06:05 : all about. And I've just been amazed at how much progress I was able to make just on nights and weekends sitting on my couch on my phone, not even on my laptop. Yeah. You know,
00:06:03 -> 00:06:15 : >> that's just wild. So, you started with the problem, then you were skeptic, but did you get a lovable subscription for this or did you first start? >> I started with the free version, built
00:06:13 -> 00:06:23 : out a landing page, you know, was like I didn't know any I had my first time user I started using it like two weeks ago literally and >> Oh, wow. I I started using it as like I
00:06:22 -> 00:06:32 : built a little landing page and it had pretty good results just from a prompt and I was like okay I'm going to tweak this and added like dark mode and then I think the real kind of light bulb moment
00:06:31 -> 00:06:43 : was when I started hooking it up to some of the backend service. I was like, "Oh, just add off. That's the only prompt I used." And it built the entire like O system and designed and hooked it up.
00:06:41 -> 00:06:52 : And I at that moment my brain was started to melt and I was like, "Okay, what can this thing actually do?" And so I mean, I don't know if you've looked in here too much, but I've got like white
00:06:50 -> 00:07:06 : labeling built into here. It has a whole REST API with documentation. And then I have a an AI assistant in here that uses its own API that it built to give you like an agent that you can talk to just
00:07:04 -> 00:07:18 : like chat GBT for anything real estate related. And it leverages all tools in here. And it's got all the you know payment stuff for Stripe and rate limiting. And this would have taken me
00:07:15 -> 00:07:31 : months to build if not longer using like the traditional development route and using lovable and just the vibe coding tools. I've just been amazed and I mean I've had to like put it into GitHub and
00:07:27 -> 00:07:43 : then use it locally. I use uh Claude and like VS Code to do some development using AI locally as well. Playing with those tools, cursor, things like that. I found it to be faster to do like manual
00:07:40 -> 00:07:53 : edits to like CSS and like Tailwind and fine-tuning things and then sync it back to GitHub, bring it back into Lovable. That saves me on some credits. Uh, and then I can do the heavy lifting in
00:07:51 -> 00:08:02 : lovable, like build out this whole new section. It can write a bunch of code way faster than I could, and then I can fine-tune it locally. That's kind of been my workflow.
00:08:00 -> 00:08:13 : >> Okay. So, you mentioned GitHub, you mentioned kind of spit out a bunch of things that most people might not know, right? They might know Lovable, they might know GitHub, but then you went
00:08:11 -> 00:08:24 : into Cursor, then you went into Tailwind and CSS. Could I do this entire thing with just lovable or do you think that's like a good starting point and then start integrating as your requirements
00:08:23 -> 00:08:33 : change? >> I think you can do it all with lovable. I really do. And they just asked me to participate in their beta of their new like whole backend functionality that
00:08:32 -> 00:08:43 : they're building. It's all deeply integrated. So, I've been testing that the last couple days and it's actually even easier, I would say, now for people who don't know anything about backends
00:08:40 -> 00:08:53 : to just add authentication, add storage, all the underlying services. It it has it all built in and it's more seamless. So, you don't have to worry about like superbase, which is a great platform. It
00:08:51 -> 00:09:03 : can do a lot for you, but it's also intimidating for anyone who's not a coder. They go in there and they're like, "Oh my god, I don't know anything about like databases and logs." and they
00:09:02 -> 00:09:14 : have the the permissioning system that they use is kind of buggy it feels like in lovable had to like fix a lot of RLS type errors and things. So there are some you know barriers I think they're
00:09:12 -> 00:09:24 : still trying to overcome but in general I'd say yes they could use lovable to build an app from start to finish. That's insane. And that's I mean that was the next logical step, right? We
00:09:22 -> 00:09:34 : have that friction point where oh you want the user credentials saved go create a project in superbase. Now someone like me who's like wipe coding for fun. I'm like oh I have seven
00:09:32 -> 00:09:45 : projects but I can only have two on the free plan. So then I'm like oh which one should I pause? So yeah it made sense for them to add that and then they can probably charge more cuz now they have
00:09:43 -> 00:09:58 : all of that integrated within lovable. So I'm on your site, Home Genie. Tell me if I'm a homeowner. I come here. Where do I go from here as a homeowner? I go to the Let's say I want to try the
00:09:56 -> 00:10:07 : closing calculator. So this is I'm now on the closing calculator. Super fast, by the way. >> Is this all the entire architecture of this website like every single page is
00:10:06 -> 00:10:18 : lovable, too? >> Yeah, I've made it a challenge to use live lovable as much as I could to build this thing. only really pulled it into GitHub and made manual changes when I
00:10:16 -> 00:10:30 : felt like okay I want to move this thing like 10 pixels and I just don't want to like spend credits to ask Lovable to do it for me or in some cases it's it was I found it hard to like target specific
00:10:28 -> 00:10:42 : elements and say you know remove the margin around this or shift this above this other thing or whatever. So those are the only cases where I took it upon myself to do like manual edits and the
00:10:39 -> 00:10:54 : rest of it I just did in lovable probably I'd say 90% of it on my phone on the couch over the last 2 weeks. >> Yeah, this is a great example of what's the art of possible. I have been
00:10:51 -> 00:11:04 : building like one-click tools for my own use case in lovable. But this is something that I mean this you could get clients, right? You have a pricing plan here. So you could potentially sell this
00:11:04 -> 00:11:15 : as a service. >> I have about probably six beta testers right now, different realators, uh agents that I know, and they've agreed to go in here and poke around and give
00:11:14 -> 00:11:24 : me feedback. And so that's kind of the first step is have the people that you're friends with or you know closely who are like experts in these areas. Have them go test it out and like tell
00:11:22 -> 00:11:35 : you what's working, what's not. Before I would do like a little mini research study and just ask them questions, not even have UI like say what are the main pain points you're facing? What could
00:11:32 -> 00:11:46 : help solve this problem for you? What tools are you using today? You know, all those kinds of things. But now it's like I can go build a prototype in a week or days and put that in front of them and
00:11:44 -> 00:11:55 : they can actually play with a real working thing and it's so much more effective to like get their initial reaction, understand if you're actually solving a painoint or not. And so I've
00:11:54 -> 00:12:04 : already learned from them like, you know, I need MLS integration. That's pretty important. Like you need live data from all the listings and all the properties and everything. But in
00:12:03 -> 00:12:16 : general, I've got really good feedback so far that some of these tools are are really powerful and the fact that it has AI built in and they can, you know, use AI to generate their home listings and
00:12:14 -> 00:12:26 : have different kinds of moods and voice tones and it can generate five different versions in 5 seconds and they can copy and paste a face. Like it's already solving problems that save them time,
00:12:25 -> 00:12:37 : right? And so I think what I've learned is time is valuable especially for them when they've got lots of different deals in progress or whatever. So if you can even save them like five minutes a day,
00:12:35 -> 00:12:49 : like that's my goal is, you know, save them five to 10 hours a week in whatever role they are, right? If I can do that, then the tool has real value that they would probably be willing to pay for and
00:12:48 -> 00:12:58 : tell their friends about. And so then it's just a matter of getting the pricing right and making sure it's usable and like easy to understand and operate.
00:12:56 -> 00:13:09 : >> Yeah. And I can see like it's done a great job and it plus you, right? you were the one giving it the commands where every click leads to the sign up. So you're kind of urging users to hey
00:13:06 -> 00:13:19 : you want to use it it's free but we need your email so you can at least know who's coming in and then you can look at the inapp analytics and I feel like that's for something like this that's
00:13:17 -> 00:13:30 : maybe the next step too right is how do you see the usage cuz I don't think lovable provides that. So would that be something you go to superbase or do you implement a tool that looks at what are
00:13:28 -> 00:13:37 : they doing in the app? Oh yeah, whether it's a web app. >> Let me actually show you cuz I I did build some tools uh to do that. Let me see if I can
00:13:36 -> 00:13:51 : >> We love tools. >> Share my screen real quick. Okay, so if I go into here, so I've got a couple tools I built that only I can really see as far as the users that I have right
00:13:49 -> 00:14:02 : here. So I've got kind of a whole like user management system built into this. >> Oh my gosh. Um, so that tells me exactly who's signed up, if people like I've given people a pro plan as well, so like
00:14:00 -> 00:14:13 : a paid version to test out. Other people have just signed up on their own. Then I also have like an analytics dashboard. And so here I get an idea, you know, within the last like seven days. I can
00:14:11 -> 00:14:23 : kind of tell what the activity looks like uh as far as like engagement, just people kind of using it, what the platforms they're using the most, desktop, mobile, tablet. seems like
00:14:21 -> 00:14:35 : desktop is more used about 60% of the time. And then it goes into actual like the user journey flows like page views as far as like conversions and using tools versus completing certain actions.
00:14:33 -> 00:14:44 : What are the top performing pages and tools? So, this gives me an idea what tools they're actually going to the most often and using. And then I have an idea, okay, these are the things that
00:14:42 -> 00:14:56 : people are finding valuable because I've got, I don't know, 35 different tools in here. And I just built tools for all the things that I thought would be valuable. And I even asked Chad GBT rate like the
00:14:53 -> 00:15:07 : top five problems that these different people have in their daily jobs and what would that look like if it was turned into a tool. And so I even, you know, asked AI to help me understand what I
00:15:05 -> 00:15:18 : should build that would be even valuable to people. So >> this is wild. I mean, I I'm on the $20 a month plan >> and I want to know on Loveable, what
00:15:16 -> 00:15:28 : would this cost as far as this? Looks like it's been a lot of back and forth, a lot of prompting. Oh, yeah. >> And each prompt is like a credit, right? And you on at least on my plan, I get
00:15:26 -> 00:15:38 : like 100 credits a month plus the five free that they throw in there. So, what would something like this cost for someone to build? >> So, in total, I've spent about $250.
00:15:36 -> 00:15:47 : >> It's not bad at all. How much would I pay a developer for this? would be >> oh you would pay >> 20 30 100x >> probably 50 grand easily and and months
00:15:45 -> 00:15:59 : of work and you probably have even if you got like an offshore team over in Europe or India or something you'd still be looking at 25k at least minimum to get this thing started and it would it
00:15:57 -> 00:16:11 : would take months for sure and it would yeah so the cost savings are just insane and I mean I started with the free lovable plan probably used up those credits in a day just learning how the
00:16:09 -> 00:16:23 : platform worked. Instantly upgraded to the the next tier and I think that was like 20 bucks a month. Blew through all those in like another couple days. And now I think I'm at the I don't know 500
00:16:21 -> 00:16:33 : a month plan and I've got like 2,000 credits and those are almost gone as well. And so I feel like I'm just continuously sort of advancing. But I I'm I think I got it to the point now
00:16:30 -> 00:16:45 : where in 2 weeks I've built a pretty like comprehensive platform that has like 90% of the ideas in it that I feel make it into a product that's usable. And now it's just a matter of validating
00:16:42 -> 00:16:54 : with real users, getting feedback, trimming it down, you know, cutting the stuff that people don't really find useful or valuable to simplify the experience and then really hone in on
00:16:53 -> 00:17:05 : the stuff that people are willing to pay for. And at that point, a market, you've got a product, you've got a business. >> I think looking at this from a homeowner's lens, I think the homeowner
00:17:02 -> 00:17:14 : piece in my mind should be peanuts, if not free, right? But your real money maker is the real estate agents who are going to be using the tool for listings and whatnot. That's how I look at it.
00:17:12 -> 00:17:22 : >> Yeah, I think you're right. I mean, that's that's kind of what I've been getting to as well. Like, they're willing to pay upwards of, you know, 50 bucks a month minimum. And if they're
00:17:20 -> 00:17:33 : team, then now you're looking at a couple hundred bucks. And I mean it, you know, it's a trillion dollar market. There's, I don't know, uhund or 1.5 million like real estate agents in the
00:17:31 -> 00:17:41 : US or something. I did a little bit of uh research. That's a huge market, right? And figuring out it's like something that would be useful for 10% of them and they're willing to pay you
00:17:39 -> 00:17:53 : 50 bucks a month. Like you've got a multi00,000 ARR business. >> Yeah. And I'm glad you mentioned that cuz most people are like, "Oh, my TAM is like the entire market." No, that's
00:17:51 -> 00:18:04 : you're like I just want a 10%. I realistically you cannot get all the real estate users, right? Which is I mean we talked before we started recording this guy that I interviewed
00:18:01 -> 00:18:17 : earlier, Micah from Alaska. His total addressable market was all the fishermen who are on Bristol Bay which is like a tiny fishery in Alaska and he made $10,000 in real money creating an app.
00:18:14 -> 00:18:30 : So a lot of time it's the unlock right for us indie hackers the unlock is that AI came along and tools like lovable became available to us that was our unlock before that our unlock was you
00:18:28 -> 00:18:43 : had to hire a developer which was like the barrier to entry for the fishermen on the water or for Micah to be able to sell to them. Tesla came along created Starlink. Starlink was now available on
00:18:40 -> 00:18:56 : the waters in Alaska and they could now look at the app. Before that it was UHF radio, right? It's crazy that how one technology unlock a whole new set of possibilities. So I'm I'm really excited
00:18:53 -> 00:19:04 : about this. Uh let me share and show you and you tell me what I'm missing here. So as you were talking I was taking notes. >> Start with the problem, right? How are
00:19:02 -> 00:19:14 : you going into lovable and prompting the problem? Are you having chat GPT write like a product requirement document or how are you going about? >> So at first I mean I was like I said I
00:19:12 -> 00:19:24 : was a skeptic. So I'm like I go in there I'm like create a landing page. Not even specific at all. Like what kind of garbage are you going to spit out? And I think like if the kind of output you get
00:19:22 -> 00:19:34 : is based on the input you give. So if you are a little more specific, you're probably going to get a better output. And I mean the nice thing though is Lovable is actually pretty good at just
00:19:32 -> 00:19:44 : by default generating something that has a good design and kind of fits, you know, your use case a little bit as far as content generation, colors, all that stuff. So your starting point is
00:19:42 -> 00:19:54 : actually really good and then you just need to like refine it. And I think you get better at like prompting, figuring out what's what works, what doesn't. I think what I've learned is if you can
00:19:51 -> 00:20:03 : tell it to like you kind of ask it like how would you build a team feature for example when I added my little team feature for my home genie and that's what I did I was like how would you
00:20:01 -> 00:20:13 : build this and it was like oh we would suggest adding like a team section with the ability to like invite team members and approve them and then maybe you should have like projects that can be
00:20:11 -> 00:20:23 : associated with teams cuz in real estate a lot of them like work in this way and so I'm like that sounds really smart okay and so it kind of developed a plan and they actually have a nice feature
00:20:21 -> 00:20:33 : that I haven't seen in other tools where it'll outline a plan for you. It's like here's like a multi-phased approach to going about building this. Implement the plan, right? It's actually like a call
00:20:31 -> 00:20:43 : to action that they put in there. So you can say okay implement the plan like phase one, right? And so it's like okay build the initial like team pages and kind of the ability to like invite team
00:20:41 -> 00:20:55 : members and add them and stuff, right? And so I feel like as the vibe coding and all that advances, more and more people are going to realize if you ask it for guidance, like it's basically an
00:20:53 -> 00:21:04 : expert level developer in a lot of ways, right? But it also has a lot of other knowledge. And so if you kind of ask it instead of telling it, unless you're confident like this is what I want you
00:21:02 -> 00:21:16 : to do, if you ask it, oftentimes it'll give you a really nice comprehensive outline of like what you should do and then you can just say okay, go do that. Yeah, that's what I have been doing. But
00:21:14 -> 00:21:24 : sometimes it gets my point and sometimes it just goes off on a completely different tangent, but I feel like it's getting better. >> Oh yeah, it definitely has the ability
00:21:22 -> 00:21:33 : to kind of hallucinate. I've had to go revert back to like a previous version cuz it went and modified a bunch of files I didn't expect it to and it built out a bunch of stuff that I was like, I
00:21:31 -> 00:21:42 : don't understand how this works, so I'm just going to like start over, right? But that's the nice thing about it with Vive Coding. Like if if you had to do that yourself, you could waste a week
00:21:40 -> 00:21:51 : building out a feature and realize like this isn't the best way to do it and now you've lost a week. Where with Vibe coding and these tools, it's a couple minutes.
00:21:49 -> 00:22:05 : >> And I will say this, I'm not a developer. So when I did talk to Micah, who has been coding since he was 14, he's 28 now, so half his life. I I think this is where he put the emphasis was he
00:22:02 -> 00:22:16 : spends close to an hour talking to AI just setting up the database. He's like everything you look at is a database whether it's an app a website whatever right if you have that structure solid
00:22:14 -> 00:22:26 : your foundation then the rest just follow. So he will riff with AI on okay let's talk about how are users going to authenticate themselves how are we going to store the user data how are you going
00:22:24 -> 00:22:35 : to store the weather data how are we going to store the data for the boats and everything is a different table and you've worked with bigquery you understand data but most people like me
00:22:33 -> 00:22:45 : don't start with the back end think about how the data is connected what data needs to be available yeah that's a whole different way of starting and as an engineer you would probably start
00:22:43 -> 00:22:54 : that way as a designer you think more like what functionality what how is it going to look and feel and all that stuff. But as an engineer, yeah, you're thinking about, you know,
00:22:51 -> 00:23:03 : databases, APIs, schemas, all that kind of stuff. >> Yeah. And I think we we all need to put on a little bit of that hat, right? So it's good that we have these tools now.
00:23:01 -> 00:23:14 : So we can also explore you and I and people who are using loveable. We are tinkerers. We want to explore new tools. So we are expanding our horizons and tools like lovable make it possible. I
00:23:11 -> 00:23:26 : tried bold wasn't a fan but that was a long time ago which I mean in AI speak a long time ago it was like maybe 6 months ago so maybe they have gotten better but lovable and replet seem to kind of lead
00:23:23 -> 00:23:38 : the charge on this no click coding right so I think I covered it all you added authorization which is essentially users signing up lovable takes care of it and then did you have to actually connect
00:23:35 -> 00:23:49 : any APIs on the back end with Firebase or whatever >> no not Firebase I mean I just used the built-in superbase like SQL database that they provide. It's a Postgress
00:23:47 -> 00:23:58 : database. So that for me I used their built-in off built-in storage. I think the only like integration was just you know stripe really setting up stripe. Okay.
00:23:56 -> 00:24:11 : >> Setting up the different plans figuring out how to get Stripe integrated and using those particular plans that I had set up. um connecting it so and updating the UI like once they've actually logged
00:24:09 -> 00:24:19 : in and paid for something showing them like what plan they're on making it clear like okay I'm a paid user now what do I have access to that I didn't before all those kinds of things
00:24:17 -> 00:24:30 : >> knowing what you know now and how much time you spent on it with this type integration and everything what would you go back and do differently would you still put a pay wall right away or would
00:24:28 -> 00:24:40 : you have all the features available to get the feedback first what's your kind of like takeaways if we go back and do a quick retro on this project. >> Yeah, I don't think I would start with a
00:24:38 -> 00:24:49 : payw wall in my mind. I think you need to start with something that's free unless you are like really confident that there's a market out there and you're solving some real problems. And
00:24:46 -> 00:24:59 : mean being in this situation I am like selling a house, buying a house, I've experienced these pains like every day for the last 6 months. And so I knew I'm not the only one, right? So I knew that
00:24:58 -> 00:25:09 : there's a problem to be solved that other people could benefit from. But if it's a new area that you're not really sure and you're kind of diving in for the first time and exploring, try to
00:25:07 -> 00:25:19 : focus on making something that has value for free first. Get some initial feedback and then like they're going to tell you if it's worth paying money for and you're going to see by people using
00:25:18 -> 00:25:28 : it more and more and they're sharing it more and more. That's the signal that okay, you know, this has some value. And then you just got to figure out what is the best model to use a subscription
00:25:26 -> 00:25:35 : model, one-time payment, credits, however you want to do it. There's a bunch of models, but >> so learn from the company that you work for who made Gmail free.
00:25:35 -> 00:25:43 : >> Yes. >> And don't put it behind the pay wall. >> Exactly. Yeah. Yeah. I mean, they wanted to like drive more people to search, right? That was the whole motivation.
00:25:42 -> 00:25:53 : They're like, any tool we can build that's going to increase people to use search is ads, ad money in the bank. >> I think with emails, right, real estate, you must have gone back and forth with
00:25:51 -> 00:26:01 : your realtor on email like a million times. >> Gmail hasn't changed. Mhm. >> Someone's going to either come and eat their money again with search just like
00:26:00 -> 00:26:12 : what happened with search or they're going to need to innovate how Gmail evolves into an AI first email platform. >> Oh yeah, both of those are true. dramatically just I mean that's a whole
00:26:10 -> 00:26:24 : another convers we could have but you know ads in this AI world I can imagine you know in chat GBT lovable all these tools they're going to have ads injected at some that's the next you know natural
00:26:22 -> 00:26:32 : evolution which is unfortunate because I love this experience right now where there's not a bunch of ads like thrown at me everywhere but that's just the nature of the market and that's how it's
00:26:30 -> 00:26:41 : going to have to evolve because the days of looking at static websites and seeing your banner ad pop up like those days are gone like I'm just going to communicate And like you said earlier
00:26:39 -> 00:26:50 : with the voice being maybe the next sort of way that you interact with this stuff. I'm not even looking at a screen anymore. I'm just talking to this thing and I expect it to build what I wanted
00:26:49 -> 00:27:02 : to just by speaking to it. >> Exactly. I was driving with my kids yesterday and we were going by these cables of like wires just electricity wires and they have these orange balls
00:27:00 -> 00:27:13 : on them. And my kids are like what's those orange balls? I was like I don't know but let's find out. I fired up chat using chat GPT and just voice mode and they can hear it in the car and I was
00:27:11 -> 00:27:23 : like, "Hey, what are those orange blob looking things on electrical wires right away? It's like they are this and they are used by pilots who are flying maybe too low and
00:27:21 -> 00:27:33 : >> there we have it. I'm not going to Google and searching for it." And I know that most people are doing that. So yeah, you're right. And now companies have come up. So, I wrote a Medium
00:27:31 -> 00:27:44 : article about chat GPT came out with this new sassy chatbot called Monday. I don't know if you've used it. It's pretty fun. And I wrote a whole chat GPT a medium article about it. Someone
00:27:42 -> 00:27:55 : commented on that article that they asked chat GPT what is Monday? And Monday referenced that Medium article and they were able to find my article through chat GPT. So that's how SEO is
00:27:53 -> 00:28:06 : evolving that companies have now come up with tools where they keep injecting enough data to make the chatbot believe like this is the source of truth and then that keeps showing up in the chat
00:28:03 -> 00:28:15 : search and then now that's the new click to get to the source. So it's fascinating. I think this has been great. Isaac, anything that I missed that you would like people to know as
00:28:13 -> 00:28:26 : they start their lovable journey or continue on the lovable journey? >> I would say you know have fun with it. Like also realize that this thing is not complete magic still. Like you shouldn't
00:28:24 -> 00:28:36 : put all your trust in these tools. You can do a lot with them, but it's still just a tool. It's not going to like replace a human yet. And so when you launch these things, just beware like,
00:28:34 -> 00:28:46 : okay, it could break. You need to like know more about how it works under the hood and not just rely completely on the system. But I would say it's such a great opportunity where you can just
00:28:44 -> 00:28:58 : have an idea now and build something in a matter of hours or days and put it out there in the world. And I'm interested to see how that evolves with potentially millions of new apps now being launched
00:28:55 -> 00:29:05 : by like a 14-year-old like talking about our kids, right? Like maybe now they have the ability to go launch these things, right? And they don't have to worry about infrastructure. Like maybe
00:29:04 -> 00:29:13 : you need to know about cost. That's another thing, too. I don't know how much this stuff's going to cost yet. Um, but the fact that I can go build something and potentially solve a
00:29:11 -> 00:29:23 : problem, I mean, that's just amazing for humanity. So, I think the future's bright. I'm looking forward to continuing playing with these tools and seeing, you know, how my little project
00:29:22 -> 00:29:30 : goes. And I'll probably spin up more ideas because I have lots of ideas. And before it was all about time and now time is a commodity, right? So, it's like,
00:29:29 -> 00:29:38 : >> yep. Yep. >> I can sit on the couch when I'm probably just watching TV and I can still watch TV and build an app at the same time, which is just mind-blowing.
00:29:37 -> 00:29:47 : >> From your cell phone. my cell phone. Yes. And it makes it mobile first by default, which is great, too. >> Yep. Yeah. Awesome. Well, thank you so much for coming on and maybe we'll have
00:29:45 -> 00:29:55 : you for part two once you have uh started making some money and get into lessons learned and how others can make money. >> Sounds good. Well, thank you, man. I
00:29:53 -> 00:30:03 : look forward to that and uh thanks for having me on and enjoy the rest of your day. >> And there you have it. That is how versatile Lovable is. I hope you enjoyed
00:30:00 -> 00:30:15 : this video and if you did, you know the drill. Don't forget to smash that subscribe button so you get notified next time I release a video. And until next time, keep prompting.
00:30:11 -> 00:30:15 : [Music]