The market moved.¶
Recurring Intelligence Cycles
The market moved. You found out three months later.
Outcome: Recurring research cycles — scored, filtered, compared — so you never fall behind.
- Area: Research
- Level: advanced
- Roles: Head of Marketing, Product Manager
- Time savings: 8 hours manual research → 15 minutes automated
- Try it: "Run a research cycle on procurement-ai"
- Compounding: Each cycle builds on previous — trends emerge over time
- Frequency: Weekly per domain (cron scheduled)
- Published: 2026-03-27
Inputs¶
- ⚙️ domain-config.md
- 📊 previous_kb_latest.md
- 🎯 scoring_calibration.md
What BOB does¶
- Load domain config + previous KB
- nlm-skill → deep research (50-70 sources)
- Import & query NotebookLM notebook
- Map entries → score with calibrated rubric
- Delta detection vs previous cycle
- Review findings, adjust weights
- Write KB + changelog + briefing
Outputs¶
- 📗 kb_2026-03-23.md
- 📋 changelog.md
- 📬 inbox_briefing.md
- 📕 report.pdf
Benefits¶
- 32× faster
- Change detection between cycles
- Weighted scoring surfaces what matters
Built on the continuous-research skill.