Methodology

How we detect, score, and synthesize intelligence

Transparency is foundational to credible intelligence. This page explains our data pipeline, scoring algorithms, confidence calibration, and the specific regulatory frameworks we analyze.

Intelligence Pipeline

From raw data to actionable signals

STEP 01

Data Ingestion

  • FDA FAERS, 510(k), BLA, and Guidance databases
  • ClinicalTrials.gov registry and results
  • PubMed / PMC with MeSH term mapping
  • FTC enforcement actions and consent decrees
  • SEC filings, Crunchbase, and venture funding feeds
  • Patent filings via USPTO and WIPO
  • State attorney general enforcement databases
STEP 02

Signal Extraction

  • Named entity recognition for drugs, companies, and regulators
  • Regulatory event classification (approval, rejection, guidance, hold)
  • Temporal extraction with event dating and sequencing
  • Cross-reference linking between signals from different sources
  • Duplicate detection and signal deduplication
STEP 03

AI Analysis & Scoring

  • Severity classification: Critical, High, Medium, Low
  • Confidence scoring (0-100) based on source quality and corroboration
  • Technology momentum composite from 5 weighted dimensions
  • Regulatory risk assessment with specific CFR citations
  • Trend detection using rolling window analysis
STEP 04

Synthesis & Delivery

  • Multi-signal intelligence briefs with narrative synthesis
  • Real-time alerting with urgency-based routing
  • Technology Momentum Index rankings updated continuously
  • Weekly digest compilation and distribution
  • API access for enterprise integration

Scoring Model

Technology Momentum Index (TMI)

The TMI is a composite score ranging from 0 to 100 that quantifies the current momentum of a biotech technology vertical. It aggregates five dimensions, each weighted by its predictive signal strength.

25%

Research Velocity

Publication rate, citation momentum, and institutional convergence from PubMed/PMC.

25%

Clinical Activity

Trial registrations, phase transitions, enrollment rates, and results reporting.

20%

Venture Funding

Round size, investor quality, deal frequency, and sector capital flow.

20%

Regulatory Signals

FDA actions, guidance documents, pathway designations, and enforcement patterns.

10%

Commercial Traction

Patent filings, partnership deals, product launches, and market formation signals.

Confidence & Limitations

How we calibrate confidence

Confidence Scoring

Every signal carries a confidence score from 0 to 100, derived from source reliability (government databases score higher than news), corroboration (signals confirmed by multiple sources score higher), recency, and specificity.

90-100Government primary source, multiple corroboration
70-89High-quality source, single corroboration
50-69Moderate source, pattern-based inference
0-49Preliminary signal, low corroboration

Known Limitations

  • AI analysis is educational, not legal advice. Always consult a qualified regulatory attorney.
  • Signal latency varies by source — FDA databases update faster than academic publications.
  • Confidence scores reflect data quality, not certainty about future regulatory outcomes.
  • International regulatory coverage is currently limited to FDA and FTC frameworks.
  • TMI scores reflect momentum trends, not investment recommendations.

See the methodology in action

Try the claim analyzer or explore the live intelligence feed to see how our pipeline transforms raw data into actionable insights.