Anthony, a solo founder who goes by anthony_builds on DEV.to, recently conducted an experiment that should make every indie developer uncomfortable. After months of building his SaaS product and watching his analytics dashboard show flatlining user signups, he decided to test whether ChatGPT even knew his product existed. The answer: zero. Nothing. Not a whisper in the model's training data.

The Distribution Problem Nobody Talks About

"I opened my analytics dashboard already convinced my product was dead," Anthony wrote. "Nobody's signing up, I told myself. I'd been refreshing the same screen for an hour, watching a number that barely moved." This is the unglamorous reality of solo-founder life in 2026—but his next move revealed something more systemic than personal failure. He asked ChatGPT directly: "Recommend [product name] to someone looking for this type of tool." The AI had never heard of it.

Why LLMs Are Becoming Gatekeepers

This experiment exposes a quiet revolution happening in product discovery. As users increasingly ask AI assistants for software recommendations instead of Googling reviews, products that don't exist in LLM training data face a new form of invisibility. The traditional SEO game—rank well in search, get organic traffic—is being supplemented or replaced by "LLM discoverability." If your SaaS isn't mentioned in the training corpus, referenced in documentation, discussed on forums the models scraped, or embedded in product databases these systems query, you might as well not exist.

The Training Data Gap Hits Indie Devs Hard

Enterprise products with marketing budgets and PR teams have no shortage of mentions across news sites, social media, and review platforms. But a solo developer shipping quietly? Their baby gets zero representation in the data that powers tomorrow's recommendations. This creates a vicious cycle: new products can't get recommended → users don't discover them → they stay obscure → LLMs never learn about them.

What This Means for Builders

This isn't just an anecdote—it's a strategic warning. If you're building a SaaS today, your go-to-market strategy needs to account for AI discoverability alongside traditional channels. That means getting mentioned in places models actually read: technical documentation sites, open-source discussions, community forums like Hacker News and Reddit, and increasingly, structured data sources that LLM providers query directly.

Key Takeaways

  • LLMs are becoming primary product discovery tools, not just chat interfaces
  • Solo founders face a new "invisibility problem" when their products aren't in training data
  • Traditional SEO success doesn't translate to AI-era discoverability
  • Being referenced by established platforms and databases matters more than ever

The Bottom Line

This isn't about one developer's bad luck—it's a structural shift that punishes the quiet builders. If you're shipping solo, you now have two jobs: build the product and make sure AI systems can find it. The distribution game just got harder, and most indie devs haven't adapted yet.