Build a Company Profile with FDA Data MCP

Regulatory intelligence teams routinely need a single view of a company's FDA footprint: which names does it operate under, where are its facilities, has it been subject to recalls, and what products does it have on the market? This walkthrough shows how to assemble that profile using FDA Data MCP, in one primary call plus targeted drill-downs.

Step 1 — One call to get the full picture

fda_company_full resolves a company name against our alias table, then fetches facilities, enforcement actions, 510(k) clearances, PMA approvals, and drug applications in a single request.

fda_company_full({ "company": "Pfizer" })

The response is structured into sections, each independently paginated:

{
  "summary": {
    "company_id": "pfizer",
    "resolved_via": "exact",
    "input_normalized": "pfizer",
    "facility_count": 52,
    "enforcement_count": 187,
    "clearance_510k_count": 14,
    "pma_count": 3,
    "drug_app_count": 412
  },
  "aliases": [
    { "alias": "Pfizer Inc", "source": "alias_pack", "confidence": 0.95 },
    { "alias": "Pfizer Labs", "source": "enforcement", "confidence": 0.8 },
    ...
  ],
  "facilities": [
    { "fei": "3002807852", "firm_name": "Pfizer Inc", "city": "New York", "state": "NY", ... },
    ...
  ],
  "enforcement": [ ... ],
  "clearances_510k": [ ... ],
  "pma_approvals": [ ... ],
  "drug_applications": [ ... ],
  "company_risk_summary": {
    "total_inspections": 89,
    "nai": 62,
    "vai": 22,
    "oai": 5,
    "last_inspection_date": "2024-11-20",
    "warning_letters": 3,
    "seizures": 0,
    "injunctions": 0
  }
}

Step 2 — Understand the risk summary

The company_risk_summary gives an aggregate view of FDA inspection outcomes and compliance actions across all facilities associated with the company. The three inspection outcome codes in the response are:

• NAI (No Action Indicated) — no significant violations found
• VAI (Voluntary Action Indicated) — violations found, firm expected to correct voluntarily
• OAI (Official Action Indicated) — significant violations requiring regulatory action

A company with a high OAI count or recent warning letters warrants closer monitoring. The summary saves you from looping through individual facilities to build this picture.

Step 3 — Review aliases

Large companies operate under many names: subsidiaries, former names, and regional variants. The aliases array maps all known names for this company ID. Each alias includes its source (how it was discovered) and confidence score. Use high-confidence aliases for automated watchlists; review lower-confidence ones manually.

Step 4 — Drill into a specific facility

Each facility in the response has an FEI (FDA Establishment Identifier). FEI is the most stable join key across FDA datasets. To get detailed facility information including inspection history and device products:

fda_get_facility({ "fei": "3002807852" })

The response includes the facility's own facility_risk_summary with per-facility inspection counts and compliance actions, plus linked device products.

Step 5 — Paginate large sections

Each section in the fda_company_full response is independently paginated. If a company has 187 enforcement actions but the default limit is 25, fetch the next page with:

fda_company_full({ "company": "Pfizer", "enforcement_offset": 25 })

The same pattern works for facilities_offset, clearances_510k_offset, pma_offset, and drug_apps_offset.

Putting it together

A complete company profile workflow typically looks like:

1. Call fda_company_full to get the overview and risk summary.
2. Review aliases to confirm subsidiary coverage.
3. Check company_risk_summary for OAI counts and warning letters.
4. Drill into high-risk facilities with fda_get_facility.
5. Paginate any section that has more results than the default limit.

This gives you a structured, auditable company profile built entirely from FDA data — no manual research or external assumptions required.