Skip to main content
POST
Search database companies

Use Case:

Search cached company data with database filters.

Pricing

Full rows are billed: you are charged for each result returned. If no results are found, the request is free. A request without limit_by returns at most 100 rows, so that is the most it is billed for. data.results_count is the whole match; page through the rest with offset_by.

Free thin rows

Send "detail": "thin" and the same search returns thin rows at no charge. A thin row tells you what the company is and where: id (the LinkedIn company id), name, industry, headcount_range, hq_country, hq_city, company_type and founded_year. It has no domain or website, no LinkedIn URL and no contacts. Keep the ids that fit and pay only for what you take further (enrich the company or search its people).
  • Thin rows are free up to a daily quota per account, currently 2,000 rows per UTC day. meta.free_tier on every thin response reports daily_rows, rows_used_today and rows_left_today, and a page is cut to the rows you have left today.
  • When the day’s rows are used up, a thin search returns 429 with a Retry-After header until 00:00 UTC. It is never billed instead: "detail": "full" keeps working, billed per row.
  • meta.amount_charged is 0 on every thin response, and a thin search works on a zero balance.
  • Leaving detail out, or sending "detail": "full", returns full rows billed per row exactly as before. Existing integrations are unaffected.
  • Accounts on a custom API contract are priced by their contract, so a thin search there returns 400 unless the contract includes it.
  • The Generect MCP asks for thin rows by default.

Companies without a website

By default the search leaves out every company that has no website. Send "include_companies_without_website": true to also get companies that have a LinkedIn page but no site, such as many hospitals, clinics and local businesses.
  • They have no website and no domain, so there are no corporate emails to find for their people, and lead matching by domain does not work for them. An exclude_domains entry cannot remove one; use exclude_ids.
  • Each one is returned and billed as a result like any other. Many of them have few or no leads on LinkedIn, so expect more companies without leads.
  • Companies with zero employees on LinkedIn are still left out, and the blocked-domain list still applies to companies that have a website.
  • results_count counts them, here and on the count endpoint.
Your price decreases automatically as your lifetime spend grows. Check your current tier and pricing →

Authorizations

Authorization
string
header
required

Use the required Token prefix. Example: Authorization: Token xxxxxxxxx

Body

application/json
industries
string[]

Industries of the target companies; a company is included if it matches at least one. Values follow LinkedIn's industry taxonomy. Examples: "Software Development", "Financial Services", "Information Technology".

exclude_industries
string[]

Industries to exclude.

sub_industries
boolean
default:false

Controls how the industries filter maps onto LinkedIn's industry taxonomy. By default (false), a selected industry also matches companies in its sub-industries. Set to true to match only the industries you specify and exclude their sub-industries, for finer sub-category targeting — for example, "Manufacturing" without "Chemicals".

headcounts
string[]

Company size, as employee-count ranges; a company is included if it falls into at least one. Allowed values: "1-10", "11-50", "51-200", "201-500", "501-1000", "1001-5000", "5001-10000", "10 000+".

locations
string[]

Company headquarters locations — matches across cities, states/regions and countries; a company is included if its HQ matches at least one. Examples: "United States", "San Francisco, California, United States", "Germany".

exclude_locations
string[]

Locations to exclude.

company_types
string[]

Company types: "Public Company", "Privately Held", "Non Profit", "Government Agency", "Educational", etc.

exclude_ids
string[]

Exclude companies by LinkedIn ID or URN.

exclude_domains
string[]

Exclude companies by domain.

limit_by
integer
default:100

Maximum number of companies to return, from 1 to 10000. Without it, a search returns 100 rows: results_count is the whole match, and offset_by reaches the rest.

offset_by
integer

Number of companies to skip (0-9999, for pagination).

strict
string[]

Fields to apply strict matching on (e.g. "company_locations").

apply_domain_blocklist
boolean
default:true

Apply Generect's blocked-domain list, which removes staffing agencies, directories and similar low-signal records that match broad industry filters. Set to false to receive them. On by default on the company search endpoints, in both database and realtime mode; the company-leads endpoints do not apply it. In database search the exclusion happens before results_count, so the count always matches what the pages can deliver; in realtime the records are dropped after retrieval, so a page may hold fewer companies than results_count implies.

include_companies_without_website
boolean
default:false

Also return companies that have a LinkedIn page but no website, such as many hospitals, clinics and local businesses. Off by default: the company search endpoints drop every company without a website, in both database and realtime mode. Such a company has no website and no domain, so there are no corporate emails to find for its people and lead matching by domain does not work for it; an exclude_domains entry cannot remove it (use exclude_ids). Each one is returned and billed as a result like any other, and many of them have few or no leads on LinkedIn. The other quality rules stay: a company with zero employees on LinkedIn is still dropped, and the blocked-domain list still applies to companies that have a website. In database search these companies are counted in results_count; the realtime count comes from LinkedIn and always included them. The company-leads endpoints do not read it.

detail
enum<string>
default:full

Row shape. full (the default) returns full rows, billed per row. thin returns the company's name, industry, size and location at no charge, up to a daily quota per account (currently 2,000 rows per UTC day); a thin row has no domain, website, LinkedIn URL or contacts. A spent quota is refused with 429 until 00:00 UTC, never billed. Accounts on a custom API contract get 400 unless their contract includes thin search. The count endpoints ignore this field: counts are always free.

Available options:
full,
thin
founded_year

Founding year of the company — the "company age" filter. Accepts either an explicit list of years, or a {"min", "max"} range of founding years; both bounds are inclusive and either may be omitted (an omitted max means "up to the current year").

Database mode only. Realtime company search has no founded-year facet upstream, so it rejects this field rather than accepting it and returning a wider result set than you asked for.

Coverage matters here: the founding year is known for roughly 57% of companies, and a company without one never matches. Filtering by company age therefore also drops records that simply lack the data, not only those outside your range.

Example:

Response

Company search results

data
object

Search results.

meta
object

Operation metadata returned in every data/meta response.