Loading page…
AI flirting-reply generator + scan-a-barcode-see-the-skin-impact app: one person, zero ad spend, $1.5M a year
700K+ downloads · 90%+ margin
He paid one small streamer $120 and got back 2M views and tens of thousands of dollars — “Virality isn't luck, it's deterministic: post 100 videos a day, one blows up — do you still call that luck?”
Fewer bars = easier, cheaper, or faster for an AI-assisted solo builder. Editorial judgments based on the case details.
[Video] Kletchi, 22, came to the US from Nigeria with $100. The idea came from a genuinely awkward personal moment: he struck out with a girl → searched the App Store for “life advice” and found nothing → wrote himself a NestJS script using OpenAI to help reply to messages → at a Homecoming party, friends showed real interest in something he'd built for the first time → only then did he package it as an app (Social Wizard: screenshot a girl's IG story, drop it in, get reply lines). First sale of $10 in early 2024; cracked the distribution method 4 weeks later — $3.5K MRR in 2 weeks, $25K MRR in a month, roughly $250K collected over 6 months. His second app, Clean Eats (scan a food barcode → learn how it affects your skin and weight; core users are women 18-28), did about $10K in its first 2 weeks and was sold to a UK company 4 months later. Over 12 months the two apps totaled $1.5M and 700K+ downloads — with zero ad spend the entire way.
The first users were friends at a homecoming party. Before it was an app it was a NestJS script Kletchi wrote for himself after searching the App Store for "life advice" and finding nothing, and handing it around at that party was the first time "my high school friends cared about anything I was working on." The App Store launch followed, then four weeks of hunting for a channel. What broke it open was a single format, found by making the content himself first: screenshot a girl's Instagram story, drop it into the app, watch the reply lines appear on screen. Once that worked he took it to micro streamers who had never had a viral moment — videos landing at 100K or 200K views, posting Fortnite clips and reactions, nothing to do with social skills, but sitting on exactly the 16-to-24 male audience he needed. The outreach is a volume exercise: "you might have to reach out to 100 creators to get one that will work for you," and that creator's video then becomes the proof that unlocks the next batch.
The first sale was about $10, very late in January 2024, and it was a weekly subscription rather than a one-off purchase. The pricing runs in three tiers — roughly $10 a week, $20 a month, $80 a year — led by the weekly, which is the standard shape for a US consumer app whose buyer is a 16-to-24-year-old man making an impulse decision about a message he wants to send tonight. That first $10 signalled nothing, and the four weeks after it produced nothing either. Then the format landed, and the same pricing that had produced $10 produced $3,500 MRR two weeks later, crossed $25,000 MRR inside a month, and scaled to roughly $60,000 a month, with about $250,000 cash collected by month six. What lets those numbers keep almost all of their value is the cost base underneath: $1,000-$2,000 a month on infrastructure, and margins after Apple's cut that he puts at "really over 90%."
The highest-value new case for you: proof that one person with zero ad spend can do $1M+ a year — provided you treat distribution as the product itself. The micro-streamer play is cheap enough to test right now ($120 a video).
Paste into Claude Code / Codex and get a working version of this product end-to-end.
Two production-grade prompts per idea — a full build spec and an SEO growth plan — plus 10 tailored build specs a week. Members only.
Join the Unicorn Club — $5/moCancel anytime, in one clickEvery claim traces back to a listed source — open any of them and check the number yourself. Timestamps jump to the second each figure is said.
Members read the exact words said here
0:00 · hear it in the videoThe exact line behind each figure is in the Unicorn Club. Read the quote, then jump to the second it's said. Join the Unicorn Club
Data credibility: Founder's own account in a Starter Story interview. All revenue figures are founder-reported or estimated by the video creator — unaudited.
The turning point was one video. He paid a micro streamer about $120, the clip hit 2 million views and brought in tens of thousands of dollars, and from there two loops ran in parallel. The first is proof inside the creator market: once one creator's video works, "it's much easier to get more creators on board because then you can have that creator's video and leverage it to other creators." Each success makes the next hundred outreach messages cheaper to convert, so the channel widens by itself. The second is volume on owned accounts, and he rejects the word luck outright — start with two TikTok accounts at three videos a day, scale to ten or twenty within a week or two, and "if you're putting out a 100 videos per day, that's 700 videos in a week. If one went viral, I don't think you would call that luck." The format never changes; only the number of people running it does. "We took this format and we scaled it over and over and over."
The stretch before this one was an entire failed product. Casp, a group photo-sharing app, reached a few thousand users and more than 20,000 posted pictures and still died. With Social Wizard the silence was shorter and sharper: a first sale of about $10 in late January, then four weeks of hunting for a channel with nothing landing. His rule for that period is to keep the work in your own hands — "it's very important that you don't outsource this process because you want to learn as much as possible. A lot of people, they have capital and they just outsource the whole thing and they end up burning so much money." What makes four weeks survivable is lowering the bar for what counts as a signal. "A lot of people think you need to find the viral format from day one. Not really. You just need to find something that kind of works." One clip that holds attention slightly longer than the others is enough to build on, and waiting for a hit instead is how four weeks turns into a year.
[Video] Core idea: “Distribution is a show don't tell game” — don't sell the product, demonstrate it. (1) First make content yourself to find a format that works (screenshot a pretty girl's story → generate a reply live on screen). (2) Find small stream-clip creators who “haven't blown up yet” (the 100-200K view tier: cheap, and dead-center on 16-24-year-old males). (3) Real numbers: he paid one creator $120; the video did 2M views and drove tens of thousands of dollars in revenue — “no paid ads or UGC can buy that ROI.” (4) Creators don't need to be in your vertical (they normally do Fortnite, memes, reactions) — audience overlap is enough; a fitness app could use a cooking creator. (5) It's a numbers game: contact 100 to close 1, then use that video as proof to unlock more. (6) Owned content: fresh TikTok accounts blow up more easily than IG; start with 2 accounts at 3 posts a day, scaling to 10-20 within 1-2 weeks. The money quote: “100 posts a day, 700 a week — when one blows up, do you still call it luck? Virality is deterministic.”
[Video] React Native + NestJS (Render/Heroku) + Firebase + Mixpanel. “Always pick the stack that ships fastest and can still scale.” Infrastructure runs just $1-2K/mo; after Apple's cut, margins are 90%+. “90% of your budget will go to marketing — the app itself costs almost nothing.” Pricing: $10/week, $20/month, $80/year — leading with a weekly subscription is the standard play for maximizing LTV in US consumer apps.
[Video] The restart-from-zero three steps: (1) Check whether the niche already has heavy content output — if yes, go straight to creator partnerships; if not, make the content yourself (never outsource early content: “everyone who threw money at outsourcing burned it all, because they never learned anything”). (2) Mass-produce content and hunt, together with creators, for a format that “kind of works.” (3) Copy that format and scale it endlessly. Product philosophy: “Extrapolate one layer out to what you're really selling — Clean Eats looks like it sells health, but both apps actually sell confidence.” A direction he likes: education apps are badly undervalued (boxing/swimming/persuasion). “In 2025, just building a product is no longer enough — if you're not solving distribution, don't bother building.”
That was one of the 50 breakdowns open to everyone. The Unicorn Club opens the other 1,285 at the same depth — plus pain-point filters and 10 build specs a week.
Join the Unicorn Club — $5/moCancel anytime, in one click