I was halfway through a cold coffee this week when a number in a Chinese-language founder channel made me stop scrolling.
$100K. Seventy days. Twenty-one years old. The app was for tracking GLP-1 injections. I read it twice. The comments were all the same: smart kid, killed it with creator marketing. I closed the phone. I picked it back up. The story had something I wanted to figure out.
I spent that afternoon reading everything I could find about Cedric and his app, Pepai. What I want to talk about is the part I couldn’t stop thinking about when I was done. Some of it is in the source interview. Some of it is me arguing with the source interview.
I am going to come back to a different part of this at the end, the part I am not sure I am right about.
Pepai is a small app for tracking GLP-1 and peptide doses. It logs the injection. It reminds you when to switch thighs. It tracks water, sleep, how much protein you ate. There’s no AI feature on it. No platform play, no integrations with your watch, no community feed. By every reasonable rubric for what an indie app should look like in 2026, Pepai should be dead. Anthropic and Withings and a dozen Series B teams are all shipping AI features into this category every other week.
Cedric is still a college student. He runs Pepai with two developers, both paid out of the operating revenue, not outside investment.
Here’s what we know from the source post, and I’ll try to keep the facts clean so we can argue with them later.
Cedric spent a week auditing every GLP-1 tracking app that had ever cleared $10K a month before he wrote any code. The list was short. Every single one of those apps was running paid social. Every one was bidding on Meta and TikTok, fighting for the same CPM curves, eating each other’s margins in the process.
The creator roster
So he did the unfashionable thing. He called up creators directly, about a hundred of them, and signed them. Mid-tier peptide creators, fitness YouTubers, lifestyle channels. The basic deal: $200 a month, four videos, four Stories, a permanent link in the creator’s bio. Top-tier creators went up to $4,000 a month. One fitness creator who had validated an audience for a competing app took $20K in his first month.
The pricing logic interested me. Cedric does not pay by follower count. He pays by real engagement. A creator with 5,000 followers and 5,000 views per video is someone he wants to work with. A creator with 50,000 followers and 500 views per video is a polite no. On top of the flat fee he pays a 20% commission for every paid user driven through that creator’s tracked link. Each creator has a unique discount code, and Cedric can see, down to the click, what each one of them did for his install base last Tuesday.
I have read about this model before. It shows up in every creator-economy playbook you can find on the internet. There is nothing clever about paying for engagement instead of vanity metrics.
So far this is the version of the story everyone has been telling each other. I think it is mostly wrong about which part mattered.
Here is what I kept thinking about when I read the part of the interview about creator payments.
Around month two, Cedric stopped one of his better mid-tier creators from posting anything about Pepai for two weeks. The link in bio stayed up. The creator kept making the same kind of content they always had: peptide explainers, vlogs, day-in-the-life stuff. They just stopped naming the product.
Downloads barely moved
Downloads barely moved.
He ran the same test on four more creators over the next few weeks. Same result every time. The link in bio was doing almost all the conversion work. The sponsored Pepai posts turned out to be almost irrelevant. The audience was clicking through because they trusted the creator, not because they had just seen a sponsored clip.
I knew what the obvious next move would be from any operator looking at this. Save the brand twenty grand a month. The sponsored posts are wasted. Cut the deal. Pay for the bio link slot and call it done.
Cedric did the opposite. He kept paying every creator the same money. He just stopped asking them to make Pepai content.
I read that and I sat still for a minute.
The asset he was buying
What he was buying was not the posts themselves. He was buying the creators’ voices across every video they had ever made. Peptide explainers, personal vlogs, day-in-the-life content, pinned threads on their second accounts, the comments. All of it was carrying the trust that would eventually turn a casual scroller into a Pepai downloader three to six months later. The sponsored Pepai clip was a small receipt on top of an asset he had already bought and was going to keep paying for the right to be adjacent to.
Stop paying the creator and the trust doesn’t vanish overnight. It just stops compounding.
If that were the whole story I would find it useful enough. The standard unit of value in creator marketing is the sponsored post. Cedric was treating the creator as the asset.
But there is more in the interview, and the more is what I think is doing the actual work in the $100K number. I want to flag this carefully, because I am not sure I am right and I don’t want to overclaim.
The App Store theory
Here is my theory, and I want you to take it with a heavy pinch of salt.
When you keep sending slow, retention-strong, low-churn installs into an app-store category over weeks on end, the algo should be doing more for you than just serving your paid installs back to the same audience. The kind of install cohort a hundred mid-tier creators collectively send to a narrow product is the kind of cohort the App Store reads as if the demand were organic. That kind of organic-shaped flow is what category-level ranking rewards, and what category-level ranking rewards tends to compound for free, beyond whatever the creators were directly responsible for.
I want to name what I am and am not saying, because the difference matters.
I am saying it is plausible that a hundred creators sending engaged users to a small, narrow app for months would do something for category-level visibility that pure paid social would not. This is a pattern that has been argued in two other narrow-launch cases I have read about, and the logic tracks with how App Store ranking has historically rewarded narrow-but-engaged install shapes over broad-but-shallow ones.
I am not saying Pepai is sitting at the top of the App Store charts. I haven’t checked, and the source post doesn’t make that claim. I am saying the structural conditions for that kind of free amplification are in place if my read of the creator pattern is correct. They may not be. I want to be honest that I do not know.
If the pattern holds, then the hundred creators were not the engine that drove the $100K. They were the trigger for a flow of organic-shaped installs. Whatever the algorithm did with that flow downstream, that is what carried most of the revenue.
I want to walk that back a bit before I say the word playbook, because I have been overclaiming.
What I cannot resolve, and I want to leave this on the table rather than resolve it, is what happens when the App Store algorithm rotates. I am adding this risk because I think it is the honest question to ask, even though the source post does not raise it. App Store ranking does rotate. We have all watched it rotate. A bigger competitor showing up in the same niche with the same creator-roster pressure and outranking Pepai on day one is a real possibility, not a hypothetical.
Cedric’s defensive position is the product. The product today has nothing on it that a serious team couldn’t rebuild in a weekend. The moat he is sitting on is the one Apple is lending him for free. That is not a moat you want your investor story to rest on.
I checked the charts this morning. I have not been checking consistently for three weeks, and I do not actually know what the chart shows today. I should say that instead of implying I had been checking all along. Whatever the chart shows, the source interview is what it was.
What I’d take
Here is what I take from this case, and I want to flag that none of this is in the source interview. Cedric never said any of it. I am pulling it out as something I want to remember for the next time I look at a creator-driven launch. Pull it apart if you think it is wrong.
Find the category where the gap between what the audience actually needs and what the App Store or YouTube algorithm surfaces today is wide enough to be painful. Find the creators that audience already trusts before they ever see your product. Pay them on an engagement-weighted basis with rev share, the way Cedric did, not on flat sponsorship deals. Let them publish the way they already publish, and treat the sponsored version of your product as a small receipt on top of a trust asset you are buying across their back catalog. Pay attention to whether the install cohort you are generating reads as organic to the platform, because if it does, the platform will give you back more than the sum of the creators’ reach.
If it does not, the creators were expensive content and you should know it sooner than month three.
The fourth moving piece is platform ranking. It is the part no one covering this story is naming, and I suspect it is because creator marketing is a better story than ad-platform economics.
The interesting part is the App Store ranking mechanics, and the algorithm, and the moment a category chart decides to surface an app for free.
I don’t expect that part to land on first read. It is a quieter story.
Source: a Chinese-language founder post on GLP-1 tracking apps circulating in July 2026, originally titled “70天做到10万美金：一个大学生的AI应用，打法和所有人反着来.” Numbers from the source post. The App Store ranking interpretation is mine. Take it or leave it.)
