I Built a $10K/Month iOS App With Paid Ads｜Starter Story – finance.biggo.com

Frederick James details how his habit tracker app reached roughly $25K over 90 days by pairing a polished App Store page with a daily Apple Ads optimization loop, TikTok creative testing, and LTV-minus-CPA math to scale from $5K to $10K monthly.
Frederick James, a London-based software developer, says a habit-tracking app he launched about three months before this episode reached roughly $13,000 in monthly revenue and now fluctuates around $10,000, built almost entirely on paid user acquisition. The conversation is mostly a tactical walkthrough of that paid-advertising pipeline — Apple Ads keyword campaigns to reach the first $5,000 a month, then TikTok Ads with target-CPA bidding to double it — plus the AI-assisted build stack and the unit economics behind it. The title’s “$10K/month playbook” is fair; the episode spends most of its runtime on exactly that.
Gus, the host of Starter Story, presses for the revenue dashboards, the exact campaign settings, and the profitability math. The reader should come away with a concrete, reproducible sequence for app-store advertising, a cost model for judging whether paid acquisition can work at all, and a clear-eyed view of the risks Frederick himself concedes: thin margins, upfront losses, and dependence on a single channel.
Frederick attributes his first revenue almost entirely to Apple Ads, which he calls “my breakthrough.” The mechanics are unglamorous and specific. He starts with the $100 free credit Apple provides, so no large budget is needed to test. Before spending anything, he invests in the App Store listing itself — AI-generated screenshots or a paid designer at $50–100 — because the listing functions as the creative, the equivalent of a video ad on another platform. He then sets up attribution with “a few lines of code” in the app and a few clicks in the RevenueCat dashboard.
The campaign structure is deliberately narrow: search match disabled, exact-match keywords only, and those keywords are the names of large competitors. When a user searches for Duolingo, his app appears as an ad. He bids Apple’s recommended maximum, waits a couple of days, and watches whether the budget is spent.
“If after a few days the funds still haven’t been spent, it means someone is betting more than you, so you need to increase your bid to match them.”
Once data exists, he exports spend from Apple and revenue from RevenueCat, combines them, and hands the file to his AI agent to find the best-performing terms — a routine he repeats every morning with his coffee.
The second leg of growth took about a week and came from TikTok Ads. Frederick’s method was to find videos in his niche that had already gone viral organically and re-shoot them himself. He launched five formats, one post each, at roughly $50 of ad spend per post. He accepts early losses deliberately, framing it as buying signal for the targeting algorithm rather than buying conversions.
“If you can afford account-level losses — not in-app losses, but losses for the sake of gathering signals for targeting algorithms — I generally think that’s the right way to go.”
He kept ads that converted or were only slightly unprofitable and killed the rest. Budgets rose by no more than 20% every few days — a ceiling he repeats emphatically and notes TikTok itself recommends. The real inflection came from switching to target-budget increases with a target cost of acquisition: he tells TikTok the price he will pay per customer and lets the platform manage spend. His campaign had been running about $200 a day; after the switch it scaled to roughly $1,000 a day, which is when revenue moved from about $5,000 to about $10,000 a week.
Frederick is direct that this model is not free money. He was not profitable for the first few weeks, and he warns a first-timer could burn $1,000 without seeing results. Ongoing paid advertising, he says, should be modeled at roughly a 30% margin. His morning routine includes a unit-economics calculation rather than just a spend report.
The calculation he describes: take realized LTV (total paying customers divided by revenue), subtract CPA, then subtract the App Store’s 15% cut (he urges enrolling in Apple’s small business program, or “you are paying 5% for nothing”), a 1% income limit, and VAT. Selling directly into the US avoids VAT at the start; an EU subscription typically carries around 20%. He advises targeting profitability on initial transactions only, and running this statement daily alongside the ad reports.
He also argues against relying on a single channel. If TikTok blocks the account — “maybe you didn’t even break anything” — the entire acquisition engine disappears. He frames the choice of channel as a choice of currency: influencers and UGC armies can yield higher margins but are paid for in time and management overhead, while paid ads are paid for in money.
Frederick, who spent five years at large London technology companies before going independent, describes a three-stage build. Stage one is preparation in Figma: an interface mockup, a basic entity-relationship diagram, an infrastructure diagram showing how data flows through the app and services, and MVP App Store screenshots. He says drawing even a wireframe first makes the AI agent “much more likely to get it right the first time” instead of iterating endlessly.
Stage two is asset generation: he pulls imagery from Pinterest, makes it unique with GPT Images (which he says is now version 2.5 and included with an OpenAI subscription), and hands layouts to ChatGPT to generate the full flow and UI. Stage three, post-launch, is A/B testing and iteration, which he says applies mostly to paywalls and pricing.
He argues the single purchased font is worth the money because it makes his products “stand out a little from the rest,” and advises founders to find one font they love and buy it.
Frederick opens his dashboards on request. The numbers are self-reported and he does not name the app, which Gus acknowledges will draw criticism in the comments.
He says the app is “nothing special” — a habit tracker in the health and fitness niche whose main value proposition is premium design and a polished interface. The idea came from an onboarding teardown of another app he saw posted on X; he concluded he could rebuild it his way, add a few obvious features, and improve the design.
Gus identifies Frederick’s posting on X as the reason he was booked, and calls building in public “a springboard for everything he’s going to build next.” Frederick says he decided early to go all in on the app and share the experience on X, and his posts — including one on reaching $13,000 a month in roughly 88 days — are how the audience finds him. Gus argues this authenticity stands out “in an age of automation, AI-powered imagery, fake people, and the ‘dead internet’ theory,” and notes the tension that the same guest builds with AI-generated assets.
Frederick’s own framing of the journey is about volume of attempts. Before apps he resold Supreme hoodies at about 15, did dropshipping, and worked as a music producer — successful, he says, but “there’s no money in music.” His stated lesson:
“Success is really just failure over a long enough period of time. You have to be willing to fail and fail fast, and you have to be willing to fail even faster.”
He also calls entrepreneurship a marathon and warns about the physical and mental toll, saying he had health problems the previous year that pushed him to take control of his life rather than continue a 9-to-5. He is now applying the same playbook to a second app.
The episode’s implicit argument is that execution — premium design, a disciplined ad pipeline, and daily unit-economics review — matters more than idea originality. Frederick’s habit tracker sits in a crowded niche, and he says so. The unresolved tension is that the business runs on roughly 30% margins, requires upfront capital before any return, and depends on two ad platforms he does not control. Frederick concedes all three and recommends a multi-channel mix, but the episode shows only the paid-only version working. The next data point to watch is whether the second app replicates the first, and whether the TikTok target-CPA scaling holds once budgets move beyond about $1,000 a day.
I started with Apple Ads. This was my breakthrough in this application. You get $100 free to get started. You don't need some crazy budget.
Never, never, never, never, never increase your advertising budget by more than 20% every few days. TikTok themselves say not to do this, even though they are interested in you spending more money.
Paid advertising requires a lot of money up front. I wasn't profitable for the first few weeks. You need to have a bit of a risk tolerance when you get into this because you may not see any profit at all.
I created an app that generated about 13,000 in revenue per month. Now it fluctuates around 10 thousand. I launched it about 3 months ago and it has given me financial freedom ever since.
I knew I could take this app, make the design premium, add a couple of features, and it would probably sell itself.
So, the journey from zero to about $5,000 a month was almost entirely Apple Ads.
Frederick James details how his habit tracker app reached roughly $25K over 90 days by pairing a polished App Store page with a daily Apple Ads optimization loop, TikTok creative testing, and LTV-minus-CPA math to scale from $5K to $10K monthly.
Frederick James, a London-based software developer, says a habit-tracking app he launched about three months before this episode reached roughly $13,000 in monthly revenue and now fluctuates around $10,000, built almost entirely on paid user acquisition. The conversation is mostly a tactical walkthrough of that paid-advertising pipeline — Apple Ads keyword campaigns to reach the first $5,000 a month, then TikTok Ads with target-CPA bidding to double it — plus the AI-assisted build stack and the unit economics behind it. The title’s “$10K/month playbook” is fair; the episode spends most of its runtime on exactly that.
Gus, the host of Starter Story, presses for the revenue dashboards, the exact campaign settings, and the profitability math. The reader should come away with a concrete, reproducible sequence for app-store advertising, a cost model for judging whether paid acquisition can work at all, and a clear-eyed view of the risks Frederick himself concedes: thin margins, upfront losses, and dependence on a single channel.
Frederick attributes his first revenue almost entirely to Apple Ads, which he calls “my breakthrough.” The mechanics are unglamorous and specific. He starts with the $100 free credit Apple provides, so no large budget is needed to test. Before spending anything, he invests in the App Store listing itself — AI-generated screenshots or a paid designer at $50–100 — because the listing functions as the creative, the equivalent of a video ad on another platform. He then sets up attribution with “a few lines of code” in the app and a few clicks in the RevenueCat dashboard.
The campaign structure is deliberately narrow: search match disabled, exact-match keywords only, and those keywords are the names of large competitors. When a user searches for Duolingo, his app appears as an ad. He bids Apple’s recommended maximum, waits a couple of days, and watches whether the budget is spent.
“If after a few days the funds still haven’t been spent, it means someone is betting more than you, so you need to increase your bid to match them.”
Once data exists, he exports spend from Apple and revenue from RevenueCat, combines them, and hands the file to his AI agent to find the best-performing terms — a routine he repeats every morning with his coffee.
The second leg of growth took about a week and came from TikTok Ads. Frederick’s method was to find videos in his niche that had already gone viral organically and re-shoot them himself. He launched five formats, one post each, at roughly $50 of ad spend per post. He accepts early losses deliberately, framing it as buying signal for the targeting algorithm rather than buying conversions.
“If you can afford account-level losses — not in-app losses, but losses for the sake of gathering signals for targeting algorithms — I generally think that’s the right way to go.”
He kept ads that converted or were only slightly unprofitable and killed the rest. Budgets rose by no more than 20% every few days — a ceiling he repeats emphatically and notes TikTok itself recommends. The real inflection came from switching to target-budget increases with a target cost of acquisition: he tells TikTok the price he will pay per customer and lets the platform manage spend. His campaign had been running about $200 a day; after the switch it scaled to roughly $1,000 a day, which is when revenue moved from about $5,000 to about $10,000 a week.
Frederick is direct that this model is not free money. He was not profitable for the first few weeks, and he warns a first-timer could burn $1,000 without seeing results. Ongoing paid advertising, he says, should be modeled at roughly a 30% margin. His morning routine includes a unit-economics calculation rather than just a spend report.
The calculation he describes: take realized LTV (total paying customers divided by revenue), subtract CPA, then subtract the App Store’s 15% cut (he urges enrolling in Apple’s small business program, or “you are paying 5% for nothing”), a 1% income limit, and VAT. Selling directly into the US avoids VAT at the start; an EU subscription typically carries around 20%. He advises targeting profitability on initial transactions only, and running this statement daily alongside the ad reports.
He also argues against relying on a single channel. If TikTok blocks the account — “maybe you didn’t even break anything” — the entire acquisition engine disappears. He frames the choice of channel as a choice of currency: influencers and UGC armies can yield higher margins but are paid for in time and management overhead, while paid ads are paid for in money.
Frederick, who spent five years at large London technology companies before going independent, describes a three-stage build. Stage one is preparation in Figma: an interface mockup, a basic entity-relationship diagram, an infrastructure diagram showing how data flows through the app and services, and MVP App Store screenshots. He says drawing even a wireframe first makes the AI agent “much more likely to get it right the first time” instead of iterating endlessly.
Stage two is asset generation: he pulls imagery from Pinterest, makes it unique with GPT Images (which he says is now version 2.5 and included with an OpenAI subscription), and hands layouts to ChatGPT to generate the full flow and UI. Stage three, post-launch, is A/B testing and iteration, which he says applies mostly to paywalls and pricing.
He argues the single purchased font is worth the money because it makes his products “stand out a little from the rest,” and advises founders to find one font they love and buy it.
Frederick opens his dashboards on request. The numbers are self-reported and he does not name the app, which Gus acknowledges will draw criticism in the comments.
He says the app is “nothing special” — a habit tracker in the health and fitness niche whose main value proposition is premium design and a polished interface. The idea came from an onboarding teardown of another app he saw posted on X; he concluded he could rebuild it his way, add a few obvious features, and improve the design.
Gus identifies Frederick’s posting on X as the reason he was booked, and calls building in public “a springboard for everything he’s going to build next.” Frederick says he decided early to go all in on the app and share the experience on X, and his posts — including one on reaching $13,000 a month in roughly 88 days — are how the audience finds him. Gus argues this authenticity stands out “in an age of automation, AI-powered imagery, fake people, and the ‘dead internet’ theory,” and notes the tension that the same guest builds with AI-generated assets.
Frederick’s own framing of the journey is about volume of attempts. Before apps he resold Supreme hoodies at about 15, did dropshipping, and worked as a music producer — successful, he says, but “there’s no money in music.” His stated lesson:
“Success is really just failure over a long enough period of time. You have to be willing to fail and fail fast, and you have to be willing to fail even faster.”
He also calls entrepreneurship a marathon and warns about the physical and mental toll, saying he had health problems the previous year that pushed him to take control of his life rather than continue a 9-to-5. He is now applying the same playbook to a second app.
The episode’s implicit argument is that execution — premium design, a disciplined ad pipeline, and daily unit-economics review — matters more than idea originality. Frederick’s habit tracker sits in a crowded niche, and he says so. The unresolved tension is that the business runs on roughly 30% margins, requires upfront capital before any return, and depends on two ad platforms he does not control. Frederick concedes all three and recommends a multi-channel mix, but the episode shows only the paid-only version working. The next data point to watch is whether the second app replicates the first, and whether the TikTok target-CPA scaling holds once budgets move beyond about $1,000 a day.
I started with Apple Ads. This was my breakthrough in this application. You get $100 free to get started. You don't need some crazy budget.
Never, never, never, never, never increase your advertising budget by more than 20% every few days. TikTok themselves say not to do this, even though they are interested in you spending more money.
Paid advertising requires a lot of money up front. I wasn't profitable for the first few weeks. You need to have a bit of a risk tolerance when you get into this because you may not see any profit at all.
I created an app that generated about 13,000 in revenue per month. Now it fluctuates around 10 thousand. I launched it about 3 months ago and it has given me financial freedom ever since.
I knew I could take this app, make the design premium, add a couple of features, and it would probably sell itself.
So, the journey from zero to about $5,000 a month was almost entirely Apple Ads.
The news and data on this website are for reference only and do not constitute investment advice or an offer to buy or sell. Information is sourced from exchanges and public sources, and may be delayed, interrupted, or updated. While we strive for accuracy, we do not guarantee timeliness, correctness, or completeness. Content may include external links for which we are not responsible. By using this site, you agree that we and our partners are not liable for any losses. Investment carries full responsibility; please carefully assess risks and consult professionals. If there are errors in the content, please contact us for correction.

source