Building a passion project to 1k users after a $120k exit

Maddox Schmidlkofer sold his first business for $15k, then grew another business to $15k MRR and sold that too. Now, he's building something for the long-term — a health app called Tallow. It's still early, but he has 1,000 users, and it brought in $650 last week alone.

Here's Maddox on how he's doing it.

Two flips and a long-term business

I'm Maddox Schmidlkofer, a CS student at Purdue. I've been coding for about five years and grew up building on the internet.

When I was 16, I started Duckmath — an unblocked games platform — because I wanted to play games at school. It grew to approximately 1.5 million monthly users and about $15k/month. Last November, I sold it to my biggest competitor, Freezenova, for $120k. I was 20 years old.

Before that, I also built Maddoxcloud — a browser-based cloud gaming platform — and sold it for $15k. I've shipped many small projects along the way, but I don't like building projects that I don't care about.

Currently, I'm working on Tallow, the project I care most about. It's a food scanner app for people who want to know what's inside their food. You point your iPhone at any grocery product and receive a full breakdown in seconds: seed oils flagged by name, a NOVA ultra-processing score, harmful additives called out against EU bans and FDA flags, and lab-tested contaminant data (microplastics, PFAS, heavy metals) cited back to the source.

It's available on the App Store with over 1,000 users and a 4.9-star rating. It's still early, but it has done a little over $2k in the last year through Superwall; the last 7 days alone brought in around $650, and the trend is finally pointing up. Paywall conversion is around 84%, which I'm really happy with.

Finding an important problem

When I started Tallow, I had just sold Duckmath. That was great, but it was a gaming site that I started as a kid to play games at school. It never meant much to me beyond the money and the challenge. I decided my next venture would be in a niche I genuinely cared about: health.

Over the last couple of years, I delved deeply into clean food and seed oils. I started reading labels and realized how hard it is to know what's really in your food. That frustration sparked the idea for Tallow. Scanner apps like Yuka exist, but most just provide a traffic-light score without showing their work. I wanted something that actually tells you why.

I validated the idea through content. I didn't write a business plan and sit on it; I just started posting. I made videos about seed oils and clean food, observed what resonated, and saw that there was a real audience that cared about this as much as I did.

Financially, I was in a decent spot to take the swing. As a CS student with $120k in the bank from my acquisition, I wasn't building this out of desperation. I could afford to play the long game and build the product right instead of flipping it. I call this my third and final startup. This time, the goal isn't a quick sale; it's to build something that matters and that I'll actually use myself every day.

Data pains

I built Tallow mostly solo. I've been coding for about five years, so the app side wasn't scary; the data and methodology were the hard parts. Tallow's pitch is that it shows its work. Anyone can slap a score on a product, but I wanted to defend every claim.

The scoring system was the biggest time sink. The Tallow score, from 0-100, builds on four weighted pillars: seed oils, additives, processing level (NOVA), and plastic exposure. I then adjust it using sourcing signals like organic, grass-fed, wild-caught, and so on. Iterating to get that right took a long time.

For data, I pull ingredient info from open food databases, cross-reference additives against EU bans and FDA flags, and surface real lab data on microplastics, PFAS, heavy metals, and pesticide residue from sources like EWG, Consumer Reports, FDA testing, PubMed studies, and oasishealth.app. I cite every figure back to its origin.

Building it wasn't expensive since I did the engineering myself; costs were just the usual app stuff: Apple Developer account, infra, and tooling.

A mobile stack

Tallow is primarily an iOS app, with a native iPhone front end. Barcode scanning, product pages, score breakdowns, and the restaurant map all reside there. I kept the client fairly lean, pushing the heavy lifting — the scoring logic and data — to the backend. This allows me to update methodology without shipping a new app version every time.

AI helps with descriptions and explanations, turning raw ingredient lists into plain-English breakdowns. However, I carefully keep the actual scoring rules-based and transparent.

For monetization, the stack uses Superwall, which allows me to build and A/B test paywalls and onboarding flows without redeploying. This has been huge because revenue is closely tied to content and traffic; quickly testing paywalls is one of the few levers I have beyond simply "posting more videos."

But this all started as a simple scanner and score. The stack has grown because I added Kid-Safe Mode, Smart Swaps, lab-tested data, and the crowd-sourced restaurant map. The restaurant map especially pushed me to handle user-submitted/crowd data, a different problem than scanning a barcode. Next, I plan to move beyond iOS-only. Android and web are the most requested features, and this will mean rethinking parts of the stack to share more logic across platforms.

Pricing and expansion

As far as the business model, Tallow is a freemium subscription app. The scanner and a basic score are free, so users get value immediately upon download, and the subscription unlocks deeper features. I have three tiers:

• $29.99/yr

• $19.99/yr with a 3-day free trial

• $9.99/mo

I started charging from the start, but revenue growth wasn't driven by pricing; content and the paywall drove it. Paywall conversion is around 84%, which is unusually high because Tallow users already care about this topic. They don't casually browse. They come from videos about seed oils or clean food, ready to pay for answers.

My expansion levers are clear. First, more installs; since the funnel converts well, more top-of-funnel traffic generates revenue. Second, platform expansion; I am iOS-only today, and Android and web are my most requested features, representing a whole new user pool. Third, features that deepen the subscription: kid-safe mode, smart swaps, and the restaurant map all encourage repeated use of Tallow, which helps retention and provides more reasons to stay subscribed.

A funnel that builds itself

Tallow's growth comes entirely from short-form videos I create. TikTok, Instagram Reels, and similar platforms. I didn't do a big launch; I simply started posting about seed oils, clean food, and what's in people's food, letting the content find its audience. The niche is loud right now; people care a lot about this stuff, so good content travels.

I view the video as the funnel. Someone sees a video about, say, microplastics or a sketchy ingredient; they care, and Tallow offers the obvious next step to check their own food. That's why my paywall converts so high (~84%); the content pre-qualifies incoming users; they aren't random installs.

I intentionally run multiple accounts. I warm up new accounts and test content across them to avoid dependence on a single one and to switch the flow if an account gets throttled or videos start getting suppressed — which happens more than you'd think. Having backups means a bad week on one account doesn't kill me. I also repurpose into other channels like Reddit and Facebook to squeeze more out of each piece.

A key lesson is that consistency and volume matter more than any single viral hit, and you must observe what resonates and lean into it. I pay attention to which angles land (the political / MAHA / 'what's really in your food' framing is strong) and create more of that, while carefully caveating factual content to avoid making unsubstantiated claims. Sensationalism is easy, but credibility isn't, so I prefer caution over being viral-but-wrong.

My revenue tracks my content almost one-to-one. Good video weeks lead to good revenue weeks; bad video weeks lead to bad revenue weeks. That's the current reality of the model.

Beyond content, a force outside of my control helped too: timing. The whole clean food conversation is loud right now. People care about this stuff more than they did a couple of years ago, and this cultural wave means good content travels further than it otherwise would. I didn't create that demand; I just built the tool for it and showed up with content while the attention was there.

My growth advice? Pick a niche where the audience already cares and become the content creator. Don't outsource it. The content is the moat and the funnel at the same time. And build the product so the video leads to an obvious next action. If your content and product address the same pain, the funnel builds itself.

Niche down and create backups

And here's my advice overall:

• Build in a niche you care about. This sounds like generic advice, but it's crucial. The code isn't the hard part. The hard part is showing up daily to create content and improve the product for months before it pays off. And you cannot fake caring for that long. Duckmath taught me that building something you don't care about gets exhausting; the money's nice, but you burn out.

• Content is the funnel; don't treat marketing as a separate task you do after building. For me, a video and the app address the same pain point, so the content qualifies people before they install, which is why my paywall converts so well. If your content and your product don't address the same problem, you'll fight your funnel forever.

• Don't build entirely on someone else's platform without a backup. Platforms will throttle you, suppress your videos, and delete your account — and it will feel personal, even though it isn't. Run multiple accounts, keep backups ready, and plan for when your main channel gets hit, because it will. I learned that the hard way.

• Start posting before you build the whole thing. You don't need a fancy validation process; post content about the problem and see if anyone cares. If it resonates, you've found your audience, and you can build for them. For me, the audience came before the product.

• Finally, journal. Write down what you're doing and what's going wrong daily. It's boring advice, but looking back at my own thinking has saved me from repeatedly circling the same problems. Half of indie hacking is not lying to yourself about what's working.

What's next?

My main goal is to make Tallow a real, lasting business, not just another thing I flip. I've called it my third and final startup for a reason; I don't want a quick exit this time. I want to build a product that becomes the default way people check what's in their food.

In the near term, that means expanding beyond iOS-only. Android and web are my most requested features by far, representing a whole new pool of users I'm currently missing. It also means making the product something people open repeatedly; the restaurant map, kid-safe mode, and smart swaps are all steps toward Tallow becoming part of someone's routine instead of a one-time scan.

I also want to continue raising the bar on credibility. Tallow's whole point is to show its work. As I grow, I want the data and methodology to become deeper and more bulletproof: more cited lab data, better ingredient coverage, and a score people genuinely trust. In a niche full of hype, credibility is the long-term moat.

More broadly, I care about the clean food problem beyond just one app. I've explored related ideas (like a directory of clean products), and long term, I'd love Tallow to be the trusted layer for "Is this good for me?" across food and the places you eat.

And personally, I want to keep building things I care about; this is the first one where the mission and the business align, and I want to keep it that way.

Follow along

The best place to check out the app is tallow.app. It's live on the App Store there (iOS for now; Android and web are on the way). You can scan your own food and poke at the methodology yourself, which is the whole point.

You can find me and the rest of my projects at maddoxschmidlkofer.com. I post most of my content under foundbymaddox on socials.