About CatalystMap

Methodology

Every relationship in CatalystMap is manually curated or AI-proposed and human-approved. Each link between a catalyst company and a publicly traded counterparty requires at least one verifiable source — SEC filings, earnings call transcripts, official press releases, or top-tier financial journalism. Relationships are typed (supplier, customer, infrastructure, partnership, investment, thematic, speculative), strength-classified (direct, indirect, speculative), and scored using a transparent, deterministic formula. The curator-in-the-loop principle means AI assists in identifying candidates but never writes data without explicit human approval.

What this is not

  • Not a stock screener — no P/E, market cap, or dividend yield filters.
  • Not a fundamental analysis tool — no financial statements, ratios, or DCF models.
  • Not a price chart or technical analysis surface.
  • Not a portfolio tracker — we do not maintain user positions.
  • Not financial advice — this is supply-chain research.
  • Not real-time data — relationships are curated periodically, not streamed.

Scoring overview

Each relationship receives a relevance score from 0 to 100, computed as:

score =
    0.40 * direct_score        (type x strength matrix)
  + 0.20 * revenue_exposure    (% of revenue tied to catalyst)
  + 0.15 * source_quality      (best source tier + agreement bonus)
  + 0.10 * recency             (days since last verification)
  + 0.05 * momentum            (neutral in v2)
  + contract_size_bonus        (0–10, for disclosed contract values)
  + gov_procurement_bonus      (0–3, for government contracts)
  - hype_risk_penalty          (0/5/15)

Clamped to [0, 100]. Scoring version: v2.

For example, a confirmed supplier with a direct relationship, 15% revenue exposure, an SEC filing source, recent verification, and a $500M government contract scores 91.5 — the highest tier. A thematic-speculative link with only blog sources and high hype risk scores 5.

Documentation