Intelligence-Research-OS
wiki/sources/competitive-intel-proof.md
Competitive Intel Proof Gate
Summary
Competitive Intelligence is the first executable Intelligence Research OS vertical because it exercises missions, source packs, signals, deltas, evidence, battlecards, and decision-ready output packs.
Products Reviewed
- Kompyte — real-time tracking across websites, reviews, content, social, ads, and job postings, filtered into actionable updates and battlecards.
- Klue battlecards — competitive enablement workflow with many-source intel collection, centralization, battlecards, and revenue-team usage.
- Crayon — competitor monitoring, relevant alerts, battlecard updates, engagement data, and sales enablement.
- Crayon battlecards — keeping pricing/product activity current inside competitive assets.
- RivalBeam and Signal Labs — newer signal-oriented positioning around AI briefs, auto-updated battlecards, source attribution, and recommended actions.
What to Copy
- Mission setup: competitors/entities, watched surfaces, source families, output intent.
- Source packs: competitor websites, pricing pages, changelogs, docs/blogs, job posts, GitHub, Reddit/HN, reviews, news/blogs.
- Signal triage: launches, pricing changes, positioning changes, complaints, hiring signals, repo/tool movement, analyst/news mentions, anomalies.
- Battlecards connected to proof points and source evidence.
- Output packs: competitor digest, battlecard, positioning shift report, sales talk tracks, product implications.
What to Avoid
- Noisy monitoring feeds without decision synthesis.
- LLM summaries treated as evidence.
- Separate bespoke engines for each use case.
- Dashboard-only output that cannot be called from workflows.
- Internal scheduling infrastructure.
Our Wedge
Intelligence Research OS should be a source-backed mission execution layer:
- Every signal has source/evidence refs.
- Every battlecard proof point cites recorded evidence.
- Multi-POV synthesis is native to the engine.
- Deltas explain what changed and why it matters.
- Outputs are available as JSON/Markdown for CLI, skill, API, and external workflow calls.
Build Decision
Build Competitive Intelligence vertical-first, then generalize the shared mission engine.