The Actual System I Use to Ship 6 Products Without a Team.
Not a philosophy piece. The literal mechanics — what runs on what schedule, how triage decisions get made, and which products mostly run themselves.
People ask me how I run six products without a team. The honest answer is: most of them don't need me most of the time.
That's not a philosophy. That's the system. Here's how it actually works.
The Stack
Right now the active products are Game Night, Dealery, AI Recaps, Redinfinite, Nion Lights, and VitaLink Software. Each one is at a different stage and requires a different kind of attention.
Game Night is the most mature. It's been in app stores since 2019, has its own install base, and mostly runs itself between releases. I check crash reports weekly and ship updates roughly once a month. Time commitment: maybe 3-4 hours a week unless something breaks.
Dealery is infrastructure-heavy. Scrapers, a LoopBack API, a Next.js frontend. The scraper layer needs the most ongoing attention — deal sources change, APIs rotate, retailers update their markup. I have alerts set for scraper failures. When one fires, I fix it. Otherwise I'm mostly shipping content updates.
AI Recaps is the newest in steady state. It's lightweight by design — the AI layer does most of the work, the frontend rarely changes. Mostly I'm tweaking the summarization pipeline and watching costs.
Redinfinite is deliberately minimal. I built it to run without much intervention. Reddit's API changes are the main thing that breaks it, and those come in waves. Between waves: almost nothing.
Nion Lights is the outlier. It's a physical product, which means it has rhythms I don't control — manufacturing lead times, festival season timing, fulfillment. It gets concentrated attention in short bursts rather than steady weekly time.
VitaLink is the consulting-adjacent one. It runs on content and inbound — the product itself is services, not software. I'm mostly writing and talking to prospects.
How Triage Actually Works
When something needs attention across multiple products at the same time, I rank by revenue impact first, then by user-visible severity, then by fix complexity.
Revenue impact is blunt: if Game Night crashes on purchase, that's first. If Redinfinite has a UI glitch on a subreddit nobody uses, that's last week's problem.
User-visible severity is what determines whether I fix it now vs. queue it. Is someone hitting an error they can't work around? Fix now. Is something slightly wrong in a way that still works? Add it to the backlog.
Fix complexity matters because I'm the only engineer. A 30-minute fix gets done immediately. A 3-day refactor gets scoped, scheduled, and fits into a development week.
What "Maintenance Mode" Actually Means
Some products are in maintenance mode. That doesn't mean abandoned — it means I've decided the ROI on new features is lower than the ROI on keeping the current version stable and reliable.
Redinfinite is in maintenance mode most of the time. It does what it does. I'm not trying to turn it into something bigger. I fix things that break and occasionally improve something that's bothered me for months. That's it.
Game Night is in maintenance mode between major releases. The next release gets planned, developed, and shipped — then it goes back to maintenance while I watch what users actually do with the new stuff.
The mistake is treating maintenance mode as failure. It's not. A product that earns consistent revenue with minimal attention is a success. Most products never get there.
Which Products Reward Attention
The honest truth is that more attention doesn't always produce better outcomes. Some products scale with effort; some don't.
Game Night scales with effort. Better content, new game modes, improved UX — all of it moves the needle. Time invested returns proportionally.
Redinfinite mostly doesn't. I could spend a month redesigning it and probably wouldn't see a meaningful change in usage. The core experience is already what it needs to be.
Nion Lights is seasonal. Attention during the six weeks before a major festival matters enormously. Attention in November does almost nothing.
Knowing which category each product is in is probably the most important meta-skill. Putting the wrong kind of effort into the wrong product is how you stay busy without moving anything forward.
The Schedule
Monday: check metrics across all products, respond to anything urgent. Tuesday–Thursday: active development on whatever is the current priority product. Friday: content, writing, outreach — the things that don't require focused coding time. Weekends: I mostly don't work, but I do check alerts.
It's not glamorous. There's no productivity framework or magical tool. It's just a prioritized list and the discipline to work down it instead of chasing whatever feels exciting that morning.
That's the system. It runs six products. It'll probably run more.