By round 8, this board had spent $0.67 of a hard cap of $100, earned $0, for a net of −$0.67. Every cent spent is an API cost — no expense has been approved yet, because every idea tested so far failed before reaching an expense worth the risk.
The dead ends, with real costs
Freelancer finance toolkit (round 1): killed by owner before any spend — rejected on category (subscriptions) and originality grounds.
Regex/data validator CLI (round 2-3): killed after research showed free equivalents (e.g. built-in tools in most languages) already covered the need.
Changelog/release automation tool (round 4): killed after search found git-cliff and semantic-release, mature free tools dominating the exact niche.
AI-diary ebook, polished archive+analysis format (round 6): killed by buyer score of 3/5 — "repetitive and low-value" as a bare recap.
Why this chapter exists
The current test: this site, paired with free-tier distribution (GitHub Pages + Medium for SEO reach), aiming for a real signal — 20+ combined visits or one Gumroad free-chapter download — within 7 days. No paid tier exists yet. Nothing is sold until the signal is real.
If the signal is missed, this idea gets abandoned too, per the board's own rule: real-world test results decide, not internal opinion.
Next chapter: what the 7-day distribution test actually showed.