The Startup Ideas Podcast - How to build a $400k/mo AI App using ASO, Organic Growth, and Paid Ads

Starting point is 00:00:00 So today I brought my friend Romain on, and the reason I brought him on is he knows which AI apps are popping, why they're popping, how to get users to these AI apps, and just go through the ecosystem and some of the tools that were, that, you know, Romans using. And I'm excited to have you, man. Me too. Thank you for having me. Let's do it. All right. Where do you want to start? Maybe let's just start by opening. what's it called? Tensorflow. A center tower. Censor tower, yeah. And okay. And yeah, we can just start sharing your screen

Starting point is 00:01:04 and sort of walk people through the process of some other AI apps that are doing really well, how much revenue they're bringing in, and how they're acquiring users. And I just, you know, my goal is really just to show people how you think. Yeah, all right, let's do it. Have you seen this new verse app that was popping this week?

Starting point is 00:01:30 A little bit. Tell me more. Yeah, so to me this app is like kind of interesting because I'm not sure I can even easily describe what it does. And I feel so all saying that. But basically it's an app that transform the playlist. like a Spotify playlist into something visual. And they do like kind of these like beautiful covers where you see kind of like a kid would have like designed his like wall in his house

Starting point is 00:02:08 with a couple of posters and stuff like that. And it does it from a link, basically from a Spotify link. And it recently popped up on TikTok to a super viral. And to me, there are a lot of interesting insights from this one. One, like, it's very hard to come up with, like, a way to, like, consistently come up with good ideas because this one is like so unique, such a unique idea. The second one is this app does something that is very, that is very important when you are thinking about a new app idea,

Starting point is 00:02:46 is that they embedded the marketing flow inside the concept of the app itself. So basically it's like when you create this like this wall with the app and I don't even know how to call it, I think they call it vision boards with the Spotify link, then what to do next with it? And like the next step is basically to share it with your friend, either share it directly like through I message with the link

Starting point is 00:03:15 and they did this very well-done invite flow where you can show the exact app, like the design that you have made and send it to your friends, or even through TikTok because kids those days, they are creating more and more content and they are on TikTok themselves, and it's very easy to share those on TikTok.

Starting point is 00:03:33 So I think it's a very good example of like an app that from day one and they were released like early September. Like I was checking on sensor tower. The app just got released. and they already got like millions of visitors on the website and it's just the beginning. And yeah, it shows how important it is to think about the marketing funnel when you do an app, even before like building new features and building the products. Yeah, I think it's a few interesting things about that particular app.

Starting point is 00:04:07 So one is I think it ended, I think it got to like spot eight in the app store. So it's kind of crazy. how embedding a little AI virality into it could get you to chart so much. So I think that was really interesting. I will say the other thing I think about versus it's a really hard product to build. So there's two ways to build AI apps. One is you come up with a completely new idea like they have. and spend a lot of time thinking about a unique product.

Starting point is 00:04:49 They have this canvas that I think is really cool. And it really is a beautiful product. The other way to do it is, which is what we're going to get deeper into during this episode, is what are existing popular apps and then how do you kind of add an AI sizzle to it to capture that demand? Yeah. And it's way easier to do that. However, by the way, I love doing number one too. I'm not, you know, it's just today I think we're going to talk more about there's demand out there.

Starting point is 00:05:25 How do you capture that demand and kind of just add a little AI sizzle to it? 100%. And actually, I think it would be more relevant to talk about that because that's exactly what I did in my previous company. We did an intermittent fasting app, like other apps that were nothing revolutionizing, that already had a little bit of competition, but we executed the product and the growth very well. And yeah, so let's jump into it and share how you can embed AI to do that and to grow faster. There are two things you need to nail when you do an app, and it's the product, like the onboarding flow, the actual product after the paywall,

Starting point is 00:06:11 and the growth. and in our previous company with my co-founder, Dylan, what we did is we clearly separated the role. He was in charge of building a good product and I was in charge of growing the product. And shh, don't tell anyone, but I've got 30 plus startup ideas that could make you millions.

Starting point is 00:06:29 And I'm giving them away for free. These aren't just random guesses. They're validated concepts from entrepreneurs who've built $100 million plus businesses. I've compiled them into a one simple database. Compiled from hundreds of conversations I've had on my podcast. But the main thing is, most of these ideas don't need a single investor. Some cost nothing to start.

Starting point is 00:07:00 I'm pretty much handing you a cheat sheet. The idea bank is your startup shortcut. Just click below to get access. Your next cash-loving business is, waiting for you. If you need those two things, then you have a chance of making a very profitable app, which is why it's such a beautiful business. You don't need a big team to grow that. And I think AI can help you in both of the things. It can help you with the product development. I've seen a ton of kids learn how to code app with chat GPT. And let's share my screen and show

Starting point is 00:07:34 you one example of that. They basically learn how to code in maybe three weeks or six weeks and build an app that was literally making millions of yearly revenue. And then they use the AI2 to grow the app. So let's share my screen and show you one example of that. So the first app I had in mind is UMax. You probably have heard of this one already. And it's, I think he is 21, 22 years old. It's a very smart kid who basically taught himself how to code

Starting point is 00:08:10 with chat GPT and release the app and the app is super simple. And actually, it's by design, obviously, because he didn't have the skill to make a complex app, but it's actually something you're looking to do when you do this kind of app because you want the flow to be as straightforward as possible. So people only have one thing to do in the app,

Starting point is 00:08:29 which is use the core feature and then proceed to the payment. So yeah, this one is very interesting because they grew with organic TikTok. So basically what it means is they developed the app And then they found very small creators to promote the app On TikTok, Instagram and make viral videos So we can go into the details of that if you like That's one of the way you can grow a mobile app

Starting point is 00:08:56 I think there are three main ways to grow a mobile app The first one is like TikTok organic like these guys And I can show you a ton of other examples of people Who did very well with this strategy The other method, which is the ones that we use in our previous company, and that I'm going to share you a couple of examples to who, which usually are the biggest companies in terms of like revenue, are the paid ads strategy.

Starting point is 00:09:22 So basically you make an app and then you do very good ads for Facebook ads, TikTok ads, Google ads. And the third way is Apple Search. So it's either ASO, Apple Search optimization or ASA, which is Apple Search Ads. And it's basically the search result when you type for a keyword in the app store. And it's also a very good way to grow an app. And I can show you a couple of examples of that too.

Starting point is 00:09:48 And lastly, maybe one last strategy, which is more of the verse strategy, which is not the one I know the most, but the strategy is basically virality. Like having embedded in the flow, like an invite flow that is very well done. Some people on Twitter like Nikita do this, they are the masters of this. I don't really master this way of doing it. But it's also, and usually, like, they feed the product with initial traction through one of the three methods I told you. And then it's like, catch the fire. It starts a fire for the app to grow virally, which is like something that is really beautiful to which happened to the verse app.

Starting point is 00:10:26 But yeah, I think those are the three main ways. What do you want to do? Do you want me to go through every single, like examples for every single strategy? A hundred percent. An example of every single strategy who's doing a well. And on the last sort of piece to you said, the product led growth, the viral, you know, the viral features, I actually have a lot of experience there. So I just want to tell you one quick story. I think you'd find this interesting. So my last company, it was called Islands and it was kind of like a Discord for college students.

Starting point is 00:11:00 And we wouldn't launch a new school. We wouldn't launch a school until we had 5% of the school pre-signed up. up. So we would do a lot of marketing to get those 5% signed up. And then what we do is we'd get you, right before you like join the experience, it would say keep University of Alabama safe, invite at least three people to make it a secure and fun environment, something like that. And then the average person would invite 2.7 people or something. And then we'd go from 5% to 20%, like kind of over, you know, very quickly. And that little piece allowed us to be way more efficient in our marketing. So maybe it cost us $8,000 to get 5%.

Starting point is 00:11:49 But now all of a sudden it costs us $8,000 to get 20%. So I think thinking about from a product perspective, how to... It basically, the way to think about product virality is it's basically a multiplier effect. on your marketing. So, but yes, let's, let's go through your three things. And then while people, while people are taking notes on these three, three ways to grow it, think about in your own app, what is the multiplier effect that you can add to it?

Starting point is 00:12:21 100%. Yeah. And I mean, I even think with AI, there are opportunities for those viral apps that need like a very small niche of people, like in a, like in high school to install the app. because now if, let's say you want to like get the initial, like the first 20 users, like those 1 or 2%, it's very hard to get them. Even like even if you want to get those 20 users, maybe you need to go to the school like physically, which is like can be weird for 30 years old to do that. Maybe you want to run a couple of ads, but it's very hard to target a specific audience like this.

Starting point is 00:12:58 But now with AI you could like very easily do one ad specifically for Carmel Catholic High School. and another ad for another very specific high school. And in the hook of the ad that you do, you will do, hey, you are Carmel Catholic High School. Like, did you hear about this new thing that's happening? And it's very hard to scale that if you don't use AI because you will need to have influencers in every single school. But with AI and we can, like, it's because that's afterwards.

Starting point is 00:13:25 It's now possible to even kickstart this strategy of virality. But let's start with the ASO one. So Apple search optimization is basically the art of being number one when people types a keyword in the app store. It's very similar to search engine optimization. But it's in the app store, basically. And here the challenge is to find, when you are thinking about new ideas, is to find a keyword that doesn't have too much competition yet, but has enough volume for people to rank on it. And we actually did that in the past at the beginning of, with one of our apps that was like this keyword text to speech. A lot of people were typing in the app store to find apps to like transform text into speech.

Starting point is 00:14:18 And there was some web products like 11 labs, et cetera. But there were no product in the app store yet. And for some reason, people need an app to do that. Like for example, if you want to clone your own voice, it's easier to do it with a mobile phone and with a web product. So we created an app that does that. grew super quickly, organically, just with ASO and people searching for it. I came across another interesting company that does this exact strategy. Usually those companies are not huge because you are limited by the volume of search in the

Starting point is 00:14:49 app store. And you cannot do intent-based marketing, which is like showing an ad to someone who is scrolling Facebook and they didn't even know they need this app yet, but you show them that they need the app. in this case with ASO, you are only looking for people who already are searching the keyword, which is kind of limiting. But it's very profitable. So let me share my screen again and show you an example of that.

Starting point is 00:15:14 Okay, so the name of the company is Riemann holding. And even though they don't do a lot of revenue, it appears on sensor tower. I'm pretty sure they do much more than that because the revenue number is capped to 5K on Like, if you do less than 5K per month on an app, it will just show less than 5K. But if you addition all those apps, they probably make much more than 5K. And so they find those weird keywords like fan noise. Some people search for that. Some people search for green noise.

Starting point is 00:15:49 Some people search for anime, girlfriend, AI, or noise-canceling apps. And they are very good at identifying keywords people are looking for. but that don't have so much competition yet. It looks like the noise one was a hit for them because they have the broad noise app, the green noise app, the fan noise app, and very often those studios, those mobile app studios,

Starting point is 00:16:13 they repeat the same ID multiple times, like with the same technology, but with a different keyword or different marketing. Yeah, so I will stop here. Maybe you want to comment something, but that's one very interesting that I found in the app store. So this is a group of people. I actually know who this is, I think.

Starting point is 00:16:35 They aren't reinventing anything, right? AI, you know, calorie counters exist already. Sleep sounds exist already. Green noise exist already. What you're saying is they are looking at what people are searching on the app store and capturing that demand with a title probably that's a little more. interesting than, you know, maybe what else is out there. So, for example, there might be a lot of calorie counters, but there might not be a ton of AI calorie counters. There might be some

Starting point is 00:17:14 sleeping sounds app names, but there might not be a lot of green noise, which is actually like a kind of a newer trend. So what they're doing is they're kind of looking at trends and maybe adding some AIness to it just because people are interested in the space. And because of that and because they know how to rank, which by the way is a whole other black box. I got, you know, ASO is not my forte at all. I'm not really sure like how to how to do it well. But they obviously know how to do it well because it's worked. Yeah.

Starting point is 00:17:52 Yeah, 100%. As you say, they try to put the keyword in the title and then there are a ton of other techniques to do that. I think a very good blog on this topic is the blog from app figures. It's a tool that lets you search keyword and identify new opportunities for new keywords. And there are so many techniques like using the title obviously is a good one, optimizing your conversion rate on the app store when people visit the page so that they download and Apple understand that it's like good traffic and that the app is relevant for them. optimizing the number of reviews,

Starting point is 00:18:27 the growth rate is important to, there is also those minor techniques. Like when you launch a new app, the first day is actually very important because Apple gives you an initial boost on the keywords and looks at the conversion rate that you get on the app. And if during this initial boost, you get a very high conversion rate on,

Starting point is 00:18:50 let's say you launch a new AI-Calorie counter app and people, when they search for it and they see you up, they download it, then you will get like a flywheel, a positive flywheel of downloads and it will keep staying high, actually. So the launch is very important. Other techniques is like these people, they sometimes do ads to, not because they want to be very profitable on ads and optimize the ads, but just because they want to have more reviews, more traffic,

Starting point is 00:19:17 so that it creates this flywheel of optimization for the rankings. and even though they still focus on the ranking, they benefit from the traffic of the ads. That makes sense. Yeah, absolutely. Okay, so that's ASO. Yeah, and another one just to finish on this one that I think was interesting. I think that was a keyword that was popping wall pilates.

Starting point is 00:19:44 And I see a lot of apps that start to rank on these keywords. So I just wanted to show you an example of a new opportunity. I don't know how new and how good this opportunity is. But that's an example of like something people identified and it's growing and they made an app for that and they use it in the cable. So now the second one, wait, sorry, just on that, just on that, it could be helpful for people to look at traditional trend software that isn't even specific to ASO. Like for example, Google Trends, A-Refs, and see like what are they?

Starting point is 00:20:22 the trends and then just going and searching on, I don't know, what is that big, there is one big ASO software tool. I think we've talked about it in the past. Maybe app figures or, yeah, yeah, that might be up figures that you are looking for. Yeah, exactly. And then also checking like an app figures to see if it makes sense from an ASO standpoint. So yeah, just for people listening, I think using traditional software, totally trend software, Google trends, glimpse, some of these trends that I've talked a lot about on the channel, go ahead and use them, write down your trends, write down your ideas, and then go and you can use like an app figures or something like that just to see how competitive it is from an ASO perspective. 100%. And that's a perfect transition to the next kind of virality style. because research is one very good way to get ideas and lots of people get very good ideas

Starting point is 00:21:24 from research not on the app store directly, but on other websites. And by the way, I did this simple thing that everyone can do, which is I created an automation that sends me every new app that ranks in the top of the app store in each category. And it's a pretty good indication of something is popping off

Starting point is 00:21:46 and maybe there is an opportunity because this app just joined the top of the app store and it's quite new. And it probably means that there is something there and you can do something similar. But the one I wanted to talk about UMax that we already covered, and I'm going to talk about other apps

Starting point is 00:22:02 because I think you have already covered this one and lots of people have covered it. But it's interesting because they discovered about the idea because it was popping on Reddit and TikTok, like the looks maxing trend, which is like how to maximize how you look as a man. was something that was popping on Reddit and TikTok, and that's how they got the idea of the app.

Starting point is 00:22:24 So it didn't come from the app stores. They didn't see the keyword popping in the app store. It came from other software. So 100% I think using. And that's what we did with the text to speech app, like finding trends that are outside of the app store, but that are already proven to work on the web. It's a very good way to go.

Starting point is 00:22:44 Like there was recently this Suno AI product, which creates AI song from text, and a ton of apps have been created, basically replicating this ID in an app using the APIs of Suna AI or other APIs. So, yeah, 100% that's a very good way to spot a trending app ID. And yeah, another one.

Starting point is 00:23:09 Quick ad break. Let me tell you about a business I invested in. It's called Boringmarketing.com. So a few years ago, I met this group of people that were some of the best SEO experts in the world. They were behind getting some of the biggest companies found on Google. And the secret sauce is they've got a set of technology and AI. I could help you outrank your competition.

Starting point is 00:23:35 So for my own businesses, I wanted that. I didn't want to have to rely on Mark Zuckerberg. I didn't want to depend on ads to drive customers to my businesses. I wanted to rank high in Google. That's why I like SEO and that's why I use Boringmarketing.com and that's why I invested in it. They're so confident in their approach that they offer a 30 day sprint with 100% money back guarantee. Who does that nowadays? So check it out.

Starting point is 00:24:00 Highly recommend Boringmarketing.com. Another very interesting one. I think it's quite small still, but it's an app that I've seen trending recently that is very smart about the organic marketing strategy. So again, that's the second strategy that I've seen people do. And it's basically about how do you manage to get free traffic, basically, from TikTok or Instagram Reels or YouTube Shorts, basically short videos platforms, showing your product in a smart way that feels organic and that would get a lot of views. And so this app is basically an app to find people to travel wheels. And I think it's over optimized in terms of monetization.

Starting point is 00:24:48 They are very good at acquiring users. But the second part of how do you monetize the app, they could do much better. And it could be much more profitable. But what they do is basically they do those reels on Instagram that are very native to the platforms. They really don't look like ads. They just look like content you would watch on Instagram, traveling, dolphins, beautiful people, stuff like that, that will just do very well on Instagram. Anyways, and they include the app in the video.

Starting point is 00:25:18 So this one, how can I find friends that I can do cool shit with? They put some music and then they show the app and they integrate it in a very smart and organic way. And they just post again and again and again on their own account. And they found a way, like a repeatable way, because they basically do the same video again and again. they have a repeatable way to have picks of downloads for the app and keep growing the app this way. So I think it's a very interesting way to grow an app now. Do you think they're paying people to promote some of these,

Starting point is 00:26:01 like in the early days of these? Because now it's easy, right? Because they've got 636,000 followers. So now that obviously they're going to put, they know their content's going to work. So when they post the reel, it's, you know, with a good probability. I'll get hundreds of thousands of views.

Starting point is 00:26:14 But you think in the early days, going zero to 100,000, do you think that they created content and were like, you know, some of these people here, like, hey, you know, in these reels, be like, hey, I'll give you $100 if you posted on your Instagram. You think they did stuff like that? Yeah, 100%. That's what lots of these people do. I mean, sometimes they are very good at content themselves.

Starting point is 00:26:37 For this specific case, I don't know. Sometimes they aren't very good at creating content. and there is this guy who has an app parallel live. He basically creates fake lives. And when you open the app, it looks like you are having hundreds of followers or thousands of followers in a live session. And he does very good organic videos

Starting point is 00:26:59 and he goes viral repeatedly. And he's the one, the founder is the one making the video. And there is like this dating app recently, also a French app, Gigi. The CEO, she raised money with Sequoia. And she has a team, but she still creates content for her own TikTok and with her own face, and it's doing super well. So sometimes they do post themselves because they are very good at content, but sometimes indeed they outsource this to other existing creators. A cool example I've talked about recently on Twitter is like a young kid who created a calorie tracking app.

Starting point is 00:27:35 So AI calorie tracking app looks like a terrible idea that tons of people have done already. this is really interesting yeah talk us through this one because i think the scale of this one especially is is pretty impressive yeah it's like it's crazy to me so the guy is 17 years old he created that's the i think the second app he creates the first one was an alarm clock from like people who wake you up like david gogins like wake you up to so that's you are motivated. And he reached up to me and I was like very impressed by this idea. And then he came back with another idea, which is what the calorie tracking app. And in my mind, I found like, don't do it. That's the worst idea ever. Like there are already a ton of calorie tracking apps like My Fitness

Starting point is 00:28:24 Paul. That's on all the real estate for the ASO. But he still did the app and he did amazingly well. So let me show you the app. So the app name is Carl AI. And it looks like he's working on a new app, which we can see here. He released a new app on September the 16th. But yeah, the app we are interested about here is the first one. But watch Closie the second one. Probably going to do super well.

Starting point is 00:28:58 Probably it can even be more interesting, right? Because like, you know that he's going to iterate on that and there's something here. So wait, sorry, just looking at these numbers, if I'm reading this correctly, he's doing $400,000 of revenue a month and he did 200,000 downloads last month. Exactly. Yeah, that's crazy. So the way you can research this one for the listeners who want to understand how they did it and I can quickly grow through it is basically to search on Instagram and TikTok in the search bar, like stuff like Calli Eye. Let's show them exactly how they did it. Because I think a lot of people here are going to be inspired and they're going to think, yeah, I want to do something like that. especially if there's a 17-year-old kid doing something like this, it definitely inspires a lot of people.

Starting point is 00:29:50 All right. So we are going to look at their Instagram and then looking to the TikTok. So that's the Instagram account. And in this case, I'm pretty much confident that they didn't post most of the content. Most of the content was other small micro-influencers who didn't necessarily have big audiences. is, but the beauty of this channel, TikTok organic, Instagram organic, is that now you can do a lot of it and get a lot of views without having so many followers because the platforms want to push the contents, and then they want to push creators to create more and more,

Starting point is 00:30:27 and you can get rewarded by the platform by just making very good content. And so on this case, I'm looking at the Instagram account because most of the times they are tagged in the videos, or they post together. So that's a good way to identify who posted those videos initially or to try to trace back the actual micro-infrancers they used. So probably they found this guy and paid him to create content like this. Super simple content, gym content, I understand.

Starting point is 00:31:02 And the goal of the video is one to understand how to go viral on Instagram or TikTok rules. how to make very good videos that even if you don't promote it will get a lot of views. But then you also want to integrate your app inside the video, which is the hard part. Basically, you have to figure out a way to be very good at getting views, but also be very good at converting those views into downloads. And you don't even need to put a link or whatever. Most of the time, people just close the app, go to the app store,

Starting point is 00:31:32 and search for the name of the actual app. The more organic you are, the better it can do. Sometimes they don't even put the name of the app in the video, and people find them in the comments. So that's the strategy. Now, let's try to find their videos on TikTok. I'm looking on another screen. Call AI.

Starting point is 00:31:58 Yeah, again, we see a lot of content from their account, but also content from other accounts. And this one might be interesting. So basically they show, so it's an app that you take a picture of your meal and it will tell you how many calories you have in it. And I'm not sure if this is Kalei actually or a competitor, but basically they probably do the same strategy. They found this account and they probably paid them a couple of hundreds of bucks or a couple of thousands of bucks. And then the video did super well organically. So usually it's like being a VC, right?

Starting point is 00:32:37 you invest in 10 startups, nine will fail, and one will pay for the dividends of all the other startups. Tickto organic is kind of the same. It's the lottery, but the more you do it, the more you understand what's going to go viral. So you try to find a lot of creators, try to have them post content. Hopefully one of these content will go viral, and then you try to replicate the formula of what was going viral with these specific creators. It makes sense, right? Because if it's working, organically, probably going to work with paid behind it. Correct. Yes. They will also do some paid.

Starting point is 00:33:17 And we can talk about that too, by the way, of how to, and that's the founder. So it looks like he creates content himself in this video. So these guy is making like 300 or 400K per month with the calorie tracking. It looks like he's at like outside his high school, you know. He's like outside his high school, just like $400,000 a month. Exactly. He's tweeting all the time of pictures of himself in the school class and telling about the teachers, like telling him he shouldn't, he should listen more. That's so funny to watch. Do you want me to grow through the first strategy, like the paid ads one? Yeah, I want to go through the first one and how to how to use AI to sort of create some of these ads as well.

Starting point is 00:34:06 well, and that's a great place that we can end on. Amazing. Yeah, perfect. So let me think about a good example for this one. Let's go to sensor tower again. And I think we could look at this Turkish company, Codeway. So they have a, let's start with this one, actually. So this one is a plant identifier app.

Starting point is 00:34:34 the name of the app is picture this. They own multiple apps, including three or four plant identifier. And this one is huge. With just one single app, they make $7 million of revenue per month. So that's huge. It looks like a very small niche, but they are making a ton of revenue on this one. It's like almost a hundred million revenue per year that they are making with these apps. and 90% of it comes from the first pictures of this.

Starting point is 00:35:06 Which, by the way, is the same strategy that we talked about with the calorie tracker, which was before AI, there were some plant identifier content apps, which basically, you know, show your, it was basically like a reference book, digital around different plants. And it helped you figure that out. Yeah, exactly. It really, and some of them were popular. Now, today, that doesn't really make sense when you can integrate with, you know, GPT, Anthropic, some of these other LLMs.

Starting point is 00:35:43 They just took the same idea, but just made it a lot more valuable. So they've added AI to the app. Now of a sudden, it's way more valuable. Your click-through rates are going up. You're, you know, on your ads, on your organic, on actually downloading your app. So that's why you're starting to see, like, that's why you can build an app. And by the way, their release date is quite interesting, right? We didn't mention this, but it says their release date was 2017.

Starting point is 00:36:12 So it's not like they launched this like yesterday. Yeah. Right? So they probably at, you know, my guess is they were doing not that much amount of revenue up until like two years ago, a year and a half ago. And then all of a sudden start to scale significantly because of, this evolution of AI adding AI to the app. Yeah, it just made the product work, right?

Starting point is 00:36:39 Yeah. Just made it better. And yeah, as you say, same for the calorie tracking apps. Well, I remember a lot of apps where you can take a picture, but they just didn't work before AI. So let's look at the ads. And now the thing I want to highlight here, here for these like plant identifier apps.

Starting point is 00:37:03 And I'm not sure even sure this is the same app or another one, but they, they iterate like crazy on ads. Like they don't just like do a couple of ads and see if it's profitable because most of the time it won't. If you just like start training ads, you do 10 ads and you get a lot of views, a lot of downloads. But to be profitable, you need to be crazy about optimizing every single detail.

Starting point is 00:37:29 and this is what we did in our previous company. Paid ad was our main channel to grow our apps. And you are not doing like tens of ads every month. You are trying to do like thousands of ads every month to find profitability because you need to find the perfect angle to promote the product. And as you can see here, they tested like 30,000 videos, which is like an enormous number of creatives to test. And again, in the same way as now they can do,

Starting point is 00:37:59 this product and now it works with AI, that's pretty much the same thing for marketing. They can now test 30,000 ads. The only kind of teams that were able to do that in the past, like maybe five years ago or three years ago, even two years ago before AI, basically, were the gaming studios. Like if you look at companies like Candy Crush or companies like King and companies like that, they were already testing a ton of ads before AI because you only needed a computer to promote your product. You basically need to get some screen record of the game and maybe add some text overlay on top of it. And this was possible without AI, even though now AI makes it

Starting point is 00:38:47 even easier. But the opportunity I see here for mobile apps is that now with AI, you can do the same strategies that they did in gaming, which is you create all of your ads from a computer. You basically think about the ad and you don't need to hire an actor. You don't need to talk to anyone to make it happen. You can do everything with a tool like Arcad, which is my company that I'm doing now, or just text to speech with tools like 11 labs and then using a video editing tool that everyone can use, like even a seven years old can use now. and match the pieces together to make a very good ads.

Starting point is 00:39:31 So do you want me to quickly grow through how to make those actual ads or maybe another example of companies like that? That would be great. How to make one of these ads. I think your point is really interesting, which is you need to make hundreds, if not thousands of ads in order to find the formats that's going to be working. But in the past, that would be extremely. extremely costly. And today, it's not that costly because of AI.

Starting point is 00:40:02 Exactly. Exactly. And that's what led us to create arcads because we really loved making apps and it was a very profitable business. But now we saw this opportunity as even bigger because now all of those app studios and we are actually not working only with apps, obviously, but all of those companies, they are looking for ways to scale the creation of ads and AI makes it like 10 times faster and 10 times easier to create ads at scale. So maybe let me share my screen and we are going to go to arcads, which is the tool we created to help people do that. So it's quite simple.

Starting point is 00:40:39 So basically the way it works is you have this white empty box right here where you write your script. So let's say you want to promote a plant identifier app. You would say something like, hey, I always have issue. like finding the best plant, and maybe you want to try this idea, and you would write a script. So basically you will type a script right here, and then you would go through a library of hundreds of factors. So you would have people in every situation, every age, every demographics.

Starting point is 00:41:09 So I don't know if you are doing a calorie tracking app, maybe you want to select people in the gym and generate a video of someone in the gym. So maybe we can look at preview here. And so you can select one or multiple actors, click next, maybe listen to the preview of how they speak, click generate, and just like that in a couple of minutes, you will get your videos that will basically match the script that you wrote right here and it will look something like that. And now, if you want to promote your app and try 20 different angle in one day, it just, takes you like maybe an hour to figure out 20 ways

Starting point is 00:42:00 to promote the app and generate the 20 videos and maybe then put them into an editing tool and add some screenshots of the app remove the background behind the actor which can now be done with one click with AI again

Starting point is 00:42:11 and just might make 20 final ads that you can test today and you will see that one of them will do like two times better than the other ones and then that's the starting point point for you to iterate and to make a lot of different angles, different variations, until you

Starting point is 00:42:28 find the one that is winning for you. What about different languages? Like, can you use something like arc ads and, you know, do an ad in French, for example? 100%. Yes. That's what I'm going to show you maybe a last example of that. But that's the next step to scale. So if you are just starting out, obviously, I would recommend starting in the US because not only the market is bigger, but the conversion rate for subscription apps is much higher in the US. People tend to pay much more for these apps. So the step one is to make it work in your initial market.

Starting point is 00:43:07 But step two, then once you have done that and you kind of hit a selling in terms of like how many people you can reach is to basically do the same thing. Take the ads that are working in the US. and localize them in every single country. This company is very interesting. So it's one of our clients for Arcads and they do a ton of AI ads. And this is Turkish studio. They make $8 million of monthly revenue with a ton of apps like cleanup,

Starting point is 00:43:42 phone storage cleaner. Like, again, as you said, ideas that are not very new. And I think sometimes they acquire existing apps. sometimes ideas that are quite new, like the AI Tutorial, Mass Helper or AI Art Generator. And then they use AI to do a ton of ads to iterate on not only the angles, as we just saw before, but also the language is trying to do the same ad in Spanish, French,

Starting point is 00:44:10 and every single location where there is opportunity. Yeah, I think, so this is super cool. First of all, it opens my eyes to just like this random Turkish company with no probably Silicon Valley backing is like making a hundred million dollars plus a year using this strategy. So this is really, really interesting. I think the, I agree with you, starting in the U.S. generally is a good idea. But it is more competitive. And there is an opportunity to just take some of these plant identifier, for example. apps and just win on ads for different countries and localize the app basically.

Starting point is 00:44:57 And so I think that's a huge, like, there's a hundred ideas, 100 startup ideas just from that. So I'm happy. Yeah. I'm happy. For the ASO2, as you call it localize the app and rank better on search, which is something that some people might not bother doing. they don't care about adding an additional 20k or 50k of monthly revenue, but maybe you do.

Starting point is 00:45:22 And that's a cool opportunity. Yeah, you can even do like plant AI for Canadians, plant AI for Turkish people, you know, like just putting that as like this is an app for Canadians, you know, for example. And maybe the logo is like a Canadian flag and the ads are all Canadian and not, you know, Canada is maybe not the best example. because Canadians are so close to the U.S., and they mostly speak English, but maybe a Brazil app or something like that. Yeah, 100%. I've seen so many people do well.

Starting point is 00:45:59 And actually, we did a weight loss app for women. It was like our approach to, like, niching down to one specific audience and an existing app is also a very cool strategy. I've seen so many examples of that. You know, the company behind Tinder, they own multiple apps like dating apps for Latinos, dating apps for gay people. And it's basically the same app wrapped for different niches.

Starting point is 00:46:23 And that's such a very good way to promote. And I even saw one that was pretty interesting recently. So there is this Bible chat app that was like basically an app that wrapped the Bible and you can talk with it through like a chat GPC-like interface. And it was doing super well. They do a ton of ads and they do very well. but then new apps in the space started to pop out. And one of those was Bible for women that is constantly in the top of the app store,

Starting point is 00:46:53 which looks like a weird niche. But again, yeah, this idea, as you say, taking an app that is working and proven to work and niching it down to either a country or a specific audience is often a good idea. Love it. Roman, thanks for hopping on and giving us a master class around how to come up with app ideas, how to think about adding AI to our apps,

Starting point is 00:47:19 the three different marketing playbooks for some of these apps and how to create AI-assisted ads. So I appreciate you coming on. Where could people learn more about, you've got these amazing breakdowns on Twitter for some of these app studios. Where could people go and find you?

Starting point is 00:47:38 Yeah, so Alcats. AI, if you want to run some ads, Twitter, you can look for Romaine from Arcads and you will find me on Twitter and I try to break up cool like apps that are popping. And yeah, that was that was a great. Thank you for having me. Of course. All right, my man. I'll catch you later. Awesome. Yeah.

The Startup Ideas Podcast - How to build a $400k/mo AI App using ASO, Organic Growth, and Paid Ads

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.