Skip to main content
Use a single endpoint and describe what you’re looking for in plain English: “high-growth fintech companies in Germany founded after 2022” or “VC investors focused on climate.” Specter handles the rest, mapping your query to the right filters and returning the matching records.

​
What’s covered

  • All eight products are searchable through the single AI search endpoint: companies, people, investors, talent signals, investor-interest signals, funding rounds, acquisitions, and IPOs.
  • Every filter a product supports can be expressed in the query: geography, sector, growth stage, funding, headcount, revenue, seniority, dates, “similar to” a company, “backed by” an investor, and so on.
  • Set product to force the product, or omit it to let Specter infer it; the response reports the product actually searched.

​
Return types

Each result is the same rich object the product’s own endpoints return, so you can chain and parse it identically.

​
How to use

  • Send query (required) and optionally product. Results page at 50; walk further with page.
  • total_count (and the X-Total-Count header) reports how many records matched in total, so you can tell when there is more than the page you were given. To work through a larger set, keep paging, or save the search in the app and read it with the saved-search results endpoints.

​
Notes

  • Charged per block of results returned rather than per result, so a partial page is cheaper than a full one and a query that matches nothing is free. See the AI search endpoint’s “Costs” note for the block size and the exact figure.
  • User-scoped filters are not supported: your network, CRM, or saved lists can’t be resolved (the API authenticates as an organisation, not a user); such a query returns 422.