Big Idea

The argument in one line.

iOS App Store farming is viable at solo scale when you cap time-per-app, source ideas from rising-but-unsaturated search trends, and automate the research-to-submission loop with Claude Code.

Who This Is For

Read if. Skip if.

READ IF YOU ARE…

You have an Apple developer account or are willing to pay the $99 fee and want a concrete system for generating app revenue without paid marketing.
You are comfortable with Claude Code or AI-assisted coding and want a practical iOS workflow beyond demo-level vibe coding.
You want an asymmetric experiment with low hours invested and low stakes per app, rather than a single hero product to build a company around.
You are curious how someone automates App Store Connect listing submission using a CDP-controlled browser and Claude Code.

SKIP IF…

You need an audience or existing distribution to drive downloads — this approach relies entirely on organic search and trending demand.
You are building SaaS or subscription products; these are simple one-time or free utility apps.
You want a deep tutorial on Swift, SwiftUI, or Xcode fundamentals — the video assumes you can follow along with AI-generated code.

TL;DR

The full version, fast.

Two apps built in roughly four total hours have earned $789 from 319 units, producing an effective $197-per-hour rate with zero marketing spend. The system works by sourcing app ideas from trending Reddit communities via SubRift and Google Trends before they saturate, then using Claude Code in plan mode to scaffold a complete SwiftUI app from a single prompt, building and testing in Xcode via an automated connection. App Store Connect submission is handled by a CDP-controlled Chrome session where Claude Code fills in all listing fields automatically. The hard discipline the creator enforces: never spend more than a day on any single app, because traction is hit-and-miss and iterating on a winner beats grinding on a loser.

Free for members

Chat with this breakdown — free.

Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.

Create a free account →

Chapters

Where the time goes.

00:00 – 02:11

01 · Results and framing

App Store Connect dashboard showing $789 / 319 units. Calculates $197/hr effective rate. Sets up the farming mental model and sunken-cost rule.

02:11 – 03:33

02 · White rabbit treasure hunt

Explains the pixel-rabbit watermark: a crypto wallet puzzle hidden in the video. Side segment, not core to the tutorial.

03:33 – 08:38

03 · App idea research workflow

Live demo of Google Trends (US, 7-day, by category), SubRift (top growing subreddits by weekly growth), and Google autocomplete. Finds 'nontoxic' as the live example.

08:38 – 10:08

04 · Claude Code research phase

Pastes Reddit sources into Claude Code with a research prompt. Runs agentic sub-agents to compile a research package. Zips output and transfers to MacBook.

10:09 – 12:50

05 · Building the app

Loads research into Claude Code on MacBook. Sets plan mode with one prompt: neo brutalist design, on-device, category-based toxic ingredient guide, no APIs. Claude Code scaffolds and builds via automated Xcode connection.

13:13 – 15:23

06 · Reviewing and fixing

Walks through app in simulator. Spots stacking bug, sends screenshot-based fix, two changes made, rebuilds.

15:24 – 17:19

07 · Automating App Store Connect upload

Triggers custom surf agent (CDP Chrome). Claude Code fills listing fields automatically. Adds build, sets compliance, submits for review. Total elapsed time from idea: about 1 hour.

17:40 – 19:33

08 · Outro and growth chart

90-day growth chart from zero. One month in, $70/day revenue now. Plugs clip rewards program and white rabbit hunt link.

Atomic Insights

Lines worth screenshotting.

Two apps built in four total hours generated $789 — an effective rate of $197 per hour with zero ad spend.
Sunken cost fallacy is the primary failure mode: spending a week on one app when the model is hit-and-miss destroys the economics.
SubRift ranks Reddit communities by weekly growth and surfaces niche demand months before it reaches app store keyword saturation.
Google autocomplete on 'is there an app for [letter]' is a free, two-minute gap-finder that needs no tool subscription.
One Claude Code plan prompt with design constraints, category scope, and a no-backend requirement is enough to produce a shippable SwiftUI first build.
A CDP-controlled Chrome session lets Claude Code fill out App Store Connect listing fields automatically, eliminating the biggest non-coding time sink.
The $99 Apple developer fee is recovered before you need to think about marketing — it is a fixed cost, not a recurring one.
Apps seeded on trending topics get organic installs from search; the creator spends nothing on marketing across both of his paid apps.
Iterating on a published app with real users is higher ROI than building a new one from scratch — ship new only when traction stalls.
Neo brutalist design specified in a single prompt produces a distinctive look without custom assets or a design tool.

Takeaway

The one-hour app cycle that avoids the sunken-cost trap

WHAT TO LEARN

Successful App Store farming requires a hard constraint on time-per-app, not just better tools — find demand before it saturates, build the thinnest useful version, then let the market decide whether to continue.

Treat each app as a low-cost market experiment: the moment you spend more than a day on something with zero downloads, the mental model has broken down.
Trend research does not require paid tools — Google Trends filtered by category plus SubRift weekly growth sort surface emerging niches in under ten minutes.
A single detailed Claude Code plan prompt with design constraints, data source, and no-backend requirements stated upfront produces a shippable first build; reserve human review for the simulator, not the scaffold.
Automating App Store Connect submission via CDP browser removes the biggest non-coding time sink and makes the one-hour target realistic.
Revenue from your first two apps should cover the $99 developer fee many times over before you need to think about marketing spend — the unit economics work at small scale.
Iterating on a published app that has even modest traction is higher ROI than building a new one; the existing users and keyword rankings are already there.

Glossary

Terms worth knowing.

App Store farming: Shipping multiple lightweight utility apps at low per-app cost to generate a portfolio of small, organic-traffic income streams rather than betting on a single hero app.
SubRift: A web tool that ranks Reddit communities by growth rate (daily, weekly, monthly), used here to identify rising-but-unsaturated topics to build apps around.
CDP Chrome (surf agent): Chrome DevTools Protocol — a way to programmatically control a browser session. Used here to let Claude Code navigate and fill out App Store Connect pages automatically.
Plan mode (Claude Code): A Claude Code interaction mode where the model produces a detailed implementation plan before writing any code, allowing the user to review scope before execution begins.
Neo brutalism: A UI design style characterized by bold colors, visible borders, flat shadows, and intentionally raw typography — chosen here for visual distinctiveness without custom illustrations.
Sub-agents: Claude Code's ability to spawn parallel or sequential agent tasks within a single session, used here to parallelize research across multiple Reddit and web sources simultaneously.

Resources

Things they pointed at.

04:05toolGoogle Trends ↗

05:05toolSubRift ↗

06:15toolClaude Code ↗

15:24productApp Store Connect ↗

Quotables

Lines you could clip.

01:26

“I don't wanna get, like, sunken cost fallacy into them. So you keep spending working a week, two, three weeks on one app because it's really hit and miss.”

Names the exact failure mode in plain language — immediately relatable to anyone who has over-invested in an unvalidated idea.→ TikTok hook↗ Tweet quote

13:39

“I am kind of tempted to just post this on the App Store. I remember this was one prompt.”

Lands the core claim — a shippable app from a single AI prompt — at the moment you can see it on screen.→ IG reel cold open↗ Tweet quote

01:04

“That's like $197 per hour. Right? That's pretty good.”

The number does the work. No setup needed.→ TikTok hook↗ Tweet quote

The Script

Word for word.

analogystory

00:00Hello. Hope you are doing well. So in this video, I just wanted to talk a bit about, uh, how I've been kinda automating using AI, uh, the App Store, right, or creating apps that I put up on the App Store.

00:12So you can see my results here. This has been going really well. Right?

00:16We have sold three nineteen app units for $789, and you can see I think we are kind of trending upwards here.

00:27So we've been making, like, $70.60, 50, around $60, uh, each day now on these apps.

00:35So you can see I don't have a lot of apps. I only have, like, two, uh, that is paid and two testing apps I did.

00:42So, basically, we basically have just two apps. And I think both these apps, spent around maybe one to two hours creating.

00:50So let's say I spent, I would say, maybe four hours on these two apps. So if you do like $7.89 divided by four hours, how much is that per hour?

01:04Yeah. That's like $197 per hour.

01:09Right? That's pretty good. That's a really good payoff.

01:12Right? So if we can scale this so this is what I wanted to talk about this video. Let's I just wanted to talk a bit about how I kinda think about these apps because I really want them to be not low effort.

01:27I guess it could be low effort, but I want don't wanna get, like, sunken cost fallacy into them. So you keep spending working a week, two, three weeks on one app because it's really hit and miss.

01:39Right? And I don't even spend any money on marketing on these apps. Right?

01:44So there is possibilities out there. So today, I just wanted to talk a bit about my process, and we might do, like, a simple example. This video might be like some kind of voice over because I'm gonna do that on my MacBook.

01:57But, anyway, that's what I wanted to talk about today. If this sounds interesting interesting to you, how we can kind of farm the App Store to, yeah, make some kind of change here if you want to.

02:08So this is the video for you. So, yeah, let's get going. Just one more thing before we get into it.

02:13So if you see this rabbit icon in the top left here, you might be thinking, what is this? So this means that this video that you are watching now has like a hidden treasure hunt that I talked about in my previous video. So if you see this rabbit after this video is posted, you can go to my Twitch account, follow, and you should see a tweet like this.

02:35So the hunt is live. You get this wallet address, and there's a clue. And in this clue, there is a way to solve this, and you can find like a hidden treasure.

02:47If you find the wallet address and the private key that is hidden somewhere in the video, this could be a link to another page on the video. It's some kind of puzzle I really wanted to do in my videos going forward. So I did the first one last, uh, last night or last day, I guess, and someone sold it after twenty minutes.

03:06So hopefully, this one is a bit more, uh, advanced. So I try to make the puzzle a bit more interesting, but that is why this white rabbit is in the corner.

03:15This means that this is a treasure hunt video. So if you wanna dive more into that, I'm gonna leave a link in the description a bit more about the the the hunt contest or whatever you wanna call it. So, yeah, that that that's why that is up there.

03:29So now let's go back to the to the App Store. Okay. So, uh, what I wanted to show you is kinda how I have been thinking about how to find these trending topics, I think that's the best way actually to do this.

03:43Because if we have, like, a trending topic, we don't need to do that much marketing.

03:49Right? So we kinda wanna find something to create into an app that, uh, is like a trending topic or something that is rising.

03:58So if we could catch that early and there's no market for it, I think that's a good way to think about this. So the sources I have been using to research this is like Google Trends. So I've been just putting it to United States past seven days, and I just sort by categories.

04:17So maybe I think one good category could be like gaming, right?

04:22So let's say there's a new game that is very popular. You could do something with that. Let's say we find Subnautica two or you can look here, yeah, Halo, Wordle hint.

04:35So this is a good way to do this. Let's say we check something like entertainment.

04:41Here, you can find something. So I've been just looking at this and sorting by search volume. So you can see Drake, okay, is on top on entertainment.

04:52And for example, shopping. Uh, here, I have, a popular watch.

04:56This could be maybe something. You can check this out. So swatch some kind of, yeah, new watch here.

05:03You can look into that. And another thing I've been using is something called SubRift. So this is discover top growing communities on Reddit.

05:11So this could be very effective too because you can sort by, let's say, weekly, and then you can do all sizes and yeah.

05:20Let's say weekly. That could be interesting. And we can scroll down here, and you can see weekly, this random video sub rate has been growing really hard.

05:29Let's say if there are anything interesting here, olden era, Veronica Miller.

05:36And you can just look through this. This is kind of things that are kind of trending. Right?

05:40You can see daily growth, weekly growth. And of course, you can visit this subreddit and see what it's about. This is about a game.

05:47Okay? Right. And this is something old and era.

05:52Yeah. That was the game. Right?

05:53What was the the other one? Veronica Miller. Yeah.

05:58This is a a girl, I guess. Okay. That's fine.

06:01But, uh, yeah, you kinda get the point. This is a good place to look at this to find trending things. Uh, let me close those.

06:10Okay. And, of course, uh, another thing I use is Google. So this is good too.

06:17You can do something like, for example, I like to do is is there an app for, and I just do a. I check, like, AirPods.

06:27Let's do g, gas prices, Google Forms, Gemini AI, Y, YouTube, and Yorkshire Water.

06:37This is something you can do too. ICloud, Indeed employees.

06:42But, of course, this is a bit short list, but you could find other ways to get longer lists. Right? So this is the three things I have been using, Google Trends, SubRiff, Google, and, of course, if you are on YouTube, uh, like, I'm not on TikTok or Instagram, but if you are there too, you can probably find some stuff.

07:05So basically looking or just browse normal Reddit. Right? So, basically, I'm just looking for things that are kinda trending up, and I think these are good things to to target.

07:20Right? That is what we are gonna do today. We're just gonna find something simple, turn it into something just to show you the workflow here.

07:27So I think I'm gonna find something now and then try to do something simple with it. Okay. So after been searching for, I guess, five minutes, I found this nontoxic.

07:37So I found this from subref. Right? Nontoxic.

07:40This was, yeah, around here. So this is like a growing subreddit, and I checked explore.

07:47You can see this is like a rising topic, nontoxic. Haven't looked at it myself, but I thought we can just make a simple list or like a simple app that kinda lists, uh, some products that is nontoxic.

08:03Right? So I think that is gonna be pretty interesting, and it's gonna be super easy to make. So I'm just gonna hone in a bit more on exactly how I want it and how what kind of posts are kind of popular here.

08:18So we can kinda go to top, and we can check what posts are kinda top people like here.

08:25I think this one is a good one to start right? And we can look at that and see how we're going to turn this into an app. And from there, uh, we're going to build it out, and I'm going to show you kind of the workflow of this.

08:38Okay. So let's just do this on Cloud Code. So let's create a directory, nontoxic, right?

08:44And let's just do that. Uh, and let's just go into Claude here. Uh, let's just do this.

08:51Okay. And from there, we're just gonna do some research. So I'm just gonna do, uh, I'm gonna take this and paste it in as a source.

09:02Okay. It's fine. Uh, let's do a couple other sources.

09:09Okay. So the instructions I'm gonna give cloud code now. You can see we have all the sources here.

09:13These are sources for our new iOS app called nontoxic. The app should be a very beginner friendly app for people to understand what to look for to find nontoxic products. Your task is to do research for the app and create a research package we can use to build the app.

09:27Good luck. Use sub agents if needed. So now I'm just gonna send out this.

09:31So this is gonna be like an agentic workflow. So hopefully, when I come back now, we have, a good research package I can bring over to my MacBook Pro so where we can actually start to build out the app based on this.

09:43It's not gonna be a complex app. It's not gonna be something that is really difficult to build or really difficult to sign up to. It's gonna be a super easy UI front end focused app, so people can just look at and kinda learn about nontoxic products.

10:01That is the simple idea. Right? Again, this is just an experiment.

10:05So I'm just gonna let this run, and I'll take you back when we have the package. Okay. So that was our research packet was done.

10:11We did a lot of research on this topic. So but I'm gonna make it pretty simple. But, basically, I did zip this.

10:18So I wanna move this over to my MacBook. So I'm just gonna reveal, and I'm just gonna move this over.

10:23So I'm just gonna share this with my MacBook. And from there, we're gonna try to build out the app using some automation I have used before. So let's just head over to the MacBook.

10:34I'm probably gonna do voice over there, but, uh, I guess that's fine. Okay. So now you see I am gonna move this research.

10:40I'm just gonna paste it in here. And if you go here now, you can see it. We have all the research.

10:47So that's pretty good. And this is kinda my automated setup for doing these apps.

10:54So I have done a video on this for if you wanna go back into my category. You can kinda see how I automate using or creating apps. But now we should be ready to actually build out this app.

11:04So I'm just gonna say read, uh, and I'm gonna tag the research document, uh, folder here.

11:12And when we have done that, I'm just gonna give some instructions how I want this app to look, how I want it to feel, and hopefully, Cloud Code can build out this app for us now. Okay. So now I'm just gonna set Cloud Code into plan mode.

11:26I'm gonna say next step is to build the app. Super easy to use. The goal of the app is to help users what to look for to find and avoid toxic products.

11:34Uh, I want to design to have a smooth neo brutalism look, colorful, interesting icons, non complex, only help users to avoid to find and avoid toxic products based on a category with tips and tricks. It should be on device, no APIs, create a detailed plan.

11:49So that is basically what I'm gonna do now for the first iteration. Iteration. This should be fairly automatic when we have the plan, and I can just come back later and have the app.

11:59And if the app kinda is showing some kind of interest, let's say we publish this and it has some kind of interest, I think it's better to build upon that when we have that.

12:11So let's just let this, uh, plan cook up. I'm gonna start it. We're gonna build out the first iteration, and we're gonna have a look at it in the simulator.

12:20So, also, I wanted to mention that now I have kinda Xcode here kinda hooked up to my automated system. There is a video on my channel. You can find out more about this.

12:32But, basically, we have created some kinda automated system to plot code can control Xcode. So you can write and connect to Xcode here.

12:42So, yeah, let's just get this going. Okay. So this looks pretty good.

12:45Right? You can see we have the neo brutalist styling here. We can scroll down, and you can see we have a where to start.

12:52I don't know how good you can see this. We have some popular categories, kitchen, household cleaning, personal care, know your ingredients.

12:59Can we click that? No. That's maybe a small bug.

13:02But we can go to categories, and here we have a lot of categories. Let's say we click into hair care.

13:08Right? And here we can see what to avoid. That's pretty good.

13:12What to look for, fragrance free formulas. We have some better swaps. I saw we have an issue here, so this is like stacked on top of each other, we gotta fix that.

13:21We have something called glow. Sorry. How do you pronounce that?

13:25But here we can see all the kind of the chemicals. Parabens, we can read about that, what they do, why it's flagged.

13:32We can read about this. Right? Yeah, I think this is pretty good.

13:36I remember this was one prompt. Right? So I am kind of tempted to just post this on the App Store.

13:44Uh, we're just gonna fix this. So I just took a screenshot, and I said to Claude Code, fix the pro tip part.

13:52I didn't say it, but I'm gonna do that. So let's see if we can fix that, rebuild this, and maybe we should just publish this. Uh, it will probably be a free app, I think.

14:05Or maybe we could do like a $1 app or something, $1.99 or something like that just for fun. And we can maybe we can maybe iterate on it later.

14:17So let's see if we can fix that. Let's see if we can do a link to the know your ingredients as our thing here.

14:26Okay. So you can see we made those two changes. We fixed the kind of the pro tip thing and the home view.

14:31So if you go back to the simulator now, let's see if this has a link. Yeah.

14:36We just go straight to the that's good. And there was something about, let's say, house cleaning and the pro tip. Yeah.

14:43That's fixed too. So now you can see we have a clear here. So I think this is kinda ready to just publish now the first iteration of this.

14:50Nontoxic. That is the app. So, um, I'm gonna just go back to I'm gonna say something like, use surf agent.

14:59This is a a package I have built to open a new crown CDP browser. Don't close the personal because on the app, when you're gonna submit the app, there's a lot of things we have to fill out.

15:11So I tried to automate this. So maybe I can show you a bit. So this should pop up.

15:17Okay. So we need to just check that. And this should pop up like a new Chrome instance where we can automate this.

15:23So now you can see we are logged into our App Store Connect, and I can just say to, uh, Claude Code here. We are logged in on the App Store in, uh, CDP Chrome.

15:33Start work on the upload of the new app, nontoxic. So this is something that is really nice to automate because now we don't have to fill out everything on the Apple Store Connect page because that takes a lot of time.

15:47And since our Chrome here, Cloud Code, mean, have all the information about the app, we should be able to to fill out everything.

15:59So I'm just gonna probably just speed this up for you now, and I'm gonna take you back when you can see how far we have come here.

16:14So now we should be basically done here. All I have to do is add the build. Yeah.

16:21Missing compliance. Okay. So this is none.

16:25Right? And let's see now if we save this, if we are able to add this for review. I also created this app icon.

16:33That looks pretty cool. Right? So that's the app icon.

16:38Let's just click add for review throughout submissions.

16:45Let's see. Yeah. One point o, submit, and that's it.

16:52That took me yeah. I was out and stuff, but, uh, basically, for this full thing now, and we actually published it for review, I will say about one hour with the research and everything we did in the previous segments.

17:07So I would say basically around one hour here of, yeah, research, building, using Claude Coda and everything to set this up.

17:16So basically, all they have to do now is wait for this to complete. This could take like twenty four hours, and we should be up and running on the App Store.

17:25Sometimes we have to do some small changes, iterations, and stuff. But, uh, basically, this is what I wanted to share today. And I think we are gonna head back to the Mac Mini and just to end this video and do my final thoughts on this and how I plan to maybe scale this going forward.

17:42So, yeah, I hope this video gave you some inspiration of how you can start getting into kind of farming, if you wanna call it that, the Apple iOS App Store.

17:52Because there are opportunities out there if you wanna put in some time, learn AI tools, cloud code, codex, open code, study a bit on kinda Xcode, how you can do this, set up your account. The only downside is that you have to pay the Apple dev fee that is $70, but you can see I already earned back, like, 10 x of that so far.

18:11And I've been only doing this for, like, one month now, I think. Can I see, like, the full thing? Last ninety days.

18:18Yeah. You can see I started exactly a month ago, and you can kinda see the growth here. So for me, it's been really good.

18:25Right? So I think there's opportunities for you too. So I'm gonna keep you posted on how the nontoxic app is gonna do and how we are doing going forward.

18:35So look out for that if you're interested. Also, remember this video is a white rabbit video. You can see it in the corner there.

18:42I'm gonna leave a link at the description to how it works if you wanna join in on the hunt and find yourself a nice price or reward somewhere. Also, I have started doing rewards for clips. If you clip up this video and you post it somewhere, and I will give a reward to anyone that gets x amount of views.

19:03Uh, but if you go to my homepage slash clips, you can submit your clip, and we'll figure something cool out if you do that. So look out for that too. But, uh, basically, today, just wanted to share these nice things.

19:16Basically, I wanna cover what you can use AI for today to, yeah, make some money or whatever your goal is. So don't forget to follow me on x if you wanna join the hunt, like I said.

19:27And thank you for tuning in today. Hope this gave you some inspiration, and I'll see you again, uh, soon.

The Hook

The bait, then the rug-pull.

Open the dashboard on any given morning and the number keeps climbing: 319 paid app units, $789 in sales, around $60 to $70 per day now — from two apps the creator spent a combined four hours building. The rest of the video is the proof of work.

Frameworks