generated: '2026-08-14' method: derived source: >- Derived from the 35 distinct endpoint URLs and 42 published request bodies in Tracxn's official public Postman collection (postman/tracxn-api-production.postman.json, harvested from https://www.postman.com/tracxnapi/tracxn-api), the 11 published MCP tools (mcp/tracxn-mcp.yml), and the Tracxn Help Center database documentation indexed by llms/tracxn-help-center-llms.txt. Tracxn publishes no machine-readable schema, so entity FIELDS are not enumerated here — only the entities, their endpoints and the reference edges the published filter keys prove. description: >- Entity graph of the Tracxn private-market dataset. Company is the hub: nearly every other entity either describes a company, links two companies, or classifies them. Relationships below are evidenced by real filter keys seen in Tracxn's own example request bodies (companyId, practiceAreaId, feedId, country, companyStage) and by the endpoint structure itself, not by a schema. Note the v2.2 -> v3.0 rename of the transaction entities, and that Legal Entity is a first-class database on the platform and in MCP but has no public REST endpoint. identifiers: format: 24-character hexadecimal object id examples: companyId: 685d395b56851c1154788da6 practiceAreaId: 5409cb72e4b0efef66413044 feedId: 52ff2cfce4b0b193ed664d1a resolution: >- Every entity is keyed on an internal Tracxn id. Resolving an external identifier (a company website/domain) to that id is a first-class MCP tool (resolve_entities) but has no dedicated REST endpoint — on REST a caller filters /companies by domain and reads the id from the result. entities: - name: Company endpoints: - POST /api/3.0/companies - POST /api/2.2/companies - POST /api/2.2/companies/search mcp_tool: search_companies description: >- Startup / private company profile — the core entity. Carries a `lastUpdatedDate` freshness field. Documented facets include stage, funding, investors, competitors, financials, cap table, filings and news. known_filter_keys: [companyId, country, companyStage, practiceAreaId, feedName, latestFundingDate] - name: LegalEntity endpoints: [] mcp_tool: search_legal_entities description: >- Registered legal entity linked to a company, with corporate structure, risk indicators, leadership and workforce, financial performance, cap table, trademarks and patents. A full platform database and an MCP tool, but NO public REST endpoint appears in the collection. rest_available: false - name: FundingRound endpoints: - POST /api/3.0/fundingrounds - POST /api/2.2/transactions mcp_tool: search_funding_rounds description: Investment round linking investors to a company; amount, date, stage, participants. renamed_from: Transaction (v2.2) - name: Acquisition endpoints: - POST /api/3.0/acquisitions - POST /api/2.2/acquisitiontransactions mcp_tool: search_acquisitions description: M&A event linking an acquirer company to an acquired company. renamed_from: AcquisitionTransaction (v2.2) - name: IPO endpoints: - POST /api/3.0/ipos mcp_tool: search_ipo_events description: IPO / public-listing event for a company. v3.0 only — no v2.2 equivalent. - name: Investor endpoints: - POST /api/2.2/investors - POST /api/3.0/investors/metric mcp_tool: search_investors description: >- VC, PE, corporate or angel investor with a portfolio. Note the asymmetry: v3.0 exposes only investor AGGREGATION; investor search remains a v2.2 endpoint. - name: TimeSeriesPoint endpoints: - POST /api/3.0/timeseries/revenue - POST /api/3.0/timeseries/netprofit - POST /api/3.0/timeseries/ebitda - POST /api/3.0/timeseries/valuation - POST /api/3.0/timeseries/balancesheetassets - POST /api/3.0/timeseries/balancesheetequity - POST /api/3.0/timeseries/balancesheetliability - POST /api/3.0/timeseries/cashflowoperatingactivity - POST /api/3.0/timeseries/cashflowinvestingactivity - POST /api/3.0/timeseries/employeecountlabourfilings - POST /api/3.0/timeseries/employeecountannualreport - POST /api/2.2/companies/timeseries/revenue - POST /api/2.2/companies/timeseries/valuation - POST /api/2.2/companies/timeseries/employeecount - POST /api/2.2/companies/timeseries/ebidta - POST /api/2.2/companies/timeseries/netprofit mcp_tool: search_time_series description: >- Year-over-year metric series attached to a company — financials (revenue, net profit, EBITDA, balance sheet, cash flow), valuation, and employee count. Always filtered by companyId. note: >- Values are raw per year; YoY growth percentages are NOT provided and must be computed client-side. In v3.0 employee count splits by source of record (labour filings vs annual report), which means two series can legitimately disagree for the same company. - name: Feed endpoints: - POST /api/3.0/feeds - POST /api/2.2/feeds mcp_tool: search_sectors description: >- Sector / industry feed — the primary classification taxonomy (3,000+ feeds). Carries publishStatus and compsCurationStatus. Business Models are queried through the same endpoint with a `dataset` discriminator. - name: PracticeArea endpoints: - POST /api/3.0/practiceareas - POST /api/2.2/practiceareas mcp_tool: search_sectors description: Top level of the sector taxonomy, above Feed (e.g. Fintech, Retail). - name: Metric endpoints: - POST /api/3.0/companies/metric - POST /api/3.0/fundingrounds/metric - POST /api/3.0/investors/metric - POST /api/3.0/acquisitions/metric mcp_tool: null description: >- Not a stored entity but an aggregation projection: a filter plus a `metric` (e.g. "count") plus a `bucket` field list, returning grouped counts. Reachable only over REST — no MCP tool returns aggregates. Rate-limited far more tightly than the rest of the API (10/hour). - name: BookmarkList endpoints: - POST /api/2.2/bookmarkslist mcp_tool: null description: >- A user's saved/custom list inside the platform — a workspace object rather than market data. v2.2 only, and unreachable from MCP. - name: Person endpoints: [] mcp_tool: null description: >- Founders, CXOs, senior management, board members and angel investors, reachable only THROUGH company records — Tracxn states that direct or standalone querying of the People database with broad filters is not supported, and that People access requires the Standard API plan above the 50K company tier. rest_available: via-company-only - name: NewsEvent endpoints: [] mcp_tool: null description: >- News items, filterable by feedName to scope to a sector. Referenced by Tracxn's FAQs as the "News API" and subject to the 500-ids-per-request cap, but no news endpoint appears in the public Postman collection. rest_available: undocumented-publicly relationships: - from: Company to: FundingRound kind: has_many via: funding rounds raised by the company - from: FundingRound to: Investor kind: has_many via: investors participating in the round - from: Investor to: Company kind: has_many via: portfolio companies - from: Company to: Acquisition kind: has_many via: acquisition / exit events involving the company (as acquirer or target) - from: Company to: IPO kind: has_many via: public listing events - from: Company to: TimeSeriesPoint kind: has_many via: companyId filter on every /timeseries/* endpoint - from: Company to: LegalEntity kind: has_many via: legal entities registered for the company - from: Company to: Person kind: has_many via: founders, CXOs, board members surfaced on the company record - from: Company to: Feed kind: belongs_to via: feedName / feedId classification - from: Feed to: PracticeArea kind: belongs_to via: practiceAreaId — the taxonomy is two-level, PracticeArea above Feed - from: Company to: Company kind: has_many via: competitors (derived by Tracxn; a dedicated endpoint returns names without consuming unique-company credits) - from: Company to: NewsEvent kind: has_many via: news coverage, filterable by feedName - from: BookmarkList to: Company kind: has_many via: user-curated membership taxonomy_note: >- Tracxn states the taxonomy itself is proprietary intellectual property and is deliberately NOT exposed via the API — the /feeds and /practiceareas endpoints let a caller FILTER by it and look up individual nodes, but the classification tree cannot be exported wholesale. coverage_note: >- 7.7M+ companies across 2,000+ sectors, 3K+ feeds and 55K+ taxonomies (company description in apis.yml); the API developer page cites "over 8M+ companies".