Three weeks ago I posted here asking for feedback on NicheRadar, a tool that finds niche communities for your product. You roasted me (nicely). Here's what changed.
What shipped:
— Supabase auth (sign up / log in, plan-based rate limits)
— Confidence scores + B2B/B2C filter on results
— Static example results on the landing so people see value before scanning
— Welcome email on signup
— Stripe integration (finishing this week — first paying customers incoming hopefully)
What I learned the hard way:
— Reddit API rejected my access request ("lacks details"). Working on resubmitting with more specifics.
— Show HN got flagged on day 1. Still flagged 13 days later. Mods didn't reply.
— Traffic is directly correlated to posting. Week I post = 100+ visitors. Week I don't = 20.
— Bounce rate went from 83% → 72% just by adding visible example results above the fold.
Honest question for IH: for those of you who've gotten your first 10 paying customers — was it mostly outreach, content, or just luck of a viral post?
Demo still free: trynicheradar.com
I'm noticing the same thing around the first customers. One thing I'd be careful about while wiring Stripe is keeping billing state and app entitlements separate — webhook retries, plan changes and failed payments can make the happy path look fine while the actual access state drifts.
I'd get those edge cases nailed down before the first customers hit it.
The traffic lesson is important: distribution is not a launch-day task; it is part of the product routine. Adding visible examples and reducing the bounce rate from 83% to 72% is real progress. Before Stripe, I’d focus on getting a few users to return repeatedly—repeat usage is a stronger buying signal than initial traffic.
The bounce rate improvement is encouraging, but I think the bigger question is whether NicheRadar is solving a discovery problem or a customer acquisition problem. I'd keep validating whether people pay because they found more communities, or because those communities consistently turn into customers.