Over the past year, I looked closely at ~100 AI / devtool launches (mostly small teams, pre–Series A) to understand why GTM often stalls after initial traction.

A few recurring patterns stood out: - Early users come from networks, communities, or launch spikes — but don’t convert to sustained demand - Content and “build in public” create visibility, not revenue - Teams overestimate how quickly distribution compounds once the product is “good enough”

What did correlate with early revenue (but wasn’t obvious upfront): - Extremely narrow ICP + problem framing - Distribution loops tied to the product itself (not external channels) - Founder-led sales lasting much longer than planned

I’m sharing this mainly to sanity-check the patterns with other builders shipping AI tools today.

Curious: - Which of these resonates with your experience? - What did you expect to work — but didn’t? - What surprised you once you tried to get the first paying users?

Happy to share more detail if useful.