Methodology

First Principle

If the pain is real, the wording is usually ugly: "too expensive", "overkill", "I just need", "for my small site", "can this be self-hosted", "why do I need 4 tools for this".

That ugly wording is gold. We follow it.

Core Filter

We do not want:

• Vague "AI for X" clones
• Ideas with no visible pain in the wild
• Products that need a startup team before day 1
• Infra-heavy beasts disguised as small tools

We do want:

• Painful but boring problems
• Things people currently overpay for
• Tools a senior dev can ship fast
• Obvious value in the first 5 minutes
• Open-source replacements for bloated SaaS

Signal Sources

1. Reddit threads and RSS search feeds
2. GitHub issue search, discussions, and feature requests
3. LinkedIn public posts via search snippets
4. X/Twitter public search snippets and mirrors
5. Hacker News and indie founder overflow

Scoring (1–5 each, 25 max)

• Severity — How painful is the problem?
• Frequency — How often does it appear across communities?
• Solvability — Can a small team build an MVP cheaply?
• OSS Displacement — How much paid SaaS can this replace?
• Distribution — Will people share / search for / talk about it?

Brief Workflow

1. Collect signals from ≥3 surfaces
2. Save raw notes to research/YYYY-MM-DD-HHMM.md
3. Score and save brief to briefs/YYYY-MM-DD-HHMM.md
4. Write channel post to posts/YYYY-MM-DD-HHMM.md
5. Update the SCOREBOARD.md

Quality Bar

A winning brief should make Bullwinkle say one of these:

• "Yes, this hurts."
• "This is small enough to ship."
• "People already pay stupid money for this."
• "Open source really can win here."