Find Companies

Describe the company profile you want, get back a list of matching companies.

What it does

• Finds companies based on the filters you pass and returns matching company profiles.
• No required parameter. You must pass at least one filter or there is nothing to search on.
• Filters cover identity, category, location, type, size, revenue, funding, social following, age, headcount growth, department mix, technographics and competitors — see the tables below.
• Cost scales with results — 0.5 credits per company returned. A default page of 25 costs 12.5 credits; a full page of 100 costs 50.

This is how you build an account list. Enrichment tools like enrich_organization work on domains you already have — this call produces them.

Use cases

• Build a TAM list from scratch. Industry, geography, size and type in one call is the fastest route from an ICP definition to a list of real accounts.
• Source funded accounts to work this quarter. Combine last_funding_round_types with last_fundraise_after to get companies that have just raised and are spending.
• Target on growth rather than size. Headcount-growth filters find the 80-person company scaling fast, which a static headcount range would rank alongside a flat one.
• Find lookalikes of your best customers. Feed competitor domains or a technographic filter and get accounts that resemble the ones already closing.
• Qualify on team shape. Department filters separate a 300-person company with 60 in sales from a 300-person company with six.

How to use it

Find privately held SaaS companies in the US with 51-200 employees.

Narrow that to companies whose last funding round was series_a or series_b and
closed after 2025-06-01.

Now only keep companies with min_headcount_growth_percent_6m above 20.

Return 100 per page and keep paging with the cursor until I have 300 companies.
Report credits used.

Example prompts

Best prompting practices

• Always pass at least one filter. There is no required parameter, so an unfiltered ask returns whatever the search defaults to rather than what you meant.
• Use the range enums, not free text. employee_count_ranges accepts fixed buckets — "51-200", "1001-5000". “Mid-market” is not a value.
• Dates are YYYY-MM-DD. last_fundraise_after: "2025-06-01". “Last six months” will not parse.
• Stack growth windows deliberately. min_headcount_growth_percent_1m catches a sudden spike; _12m catches sustained expansion. Percent filters favour small companies, absolute filters favour large ones.
• Paste the docs URL into the prompt. “Parameter reference: https://docs.syncgtm.com/mcp_server/tools/find_companies — read it first and use the exact parameter names and value formats.” With this many filters, a client left to guess invents company_size and location and silently drops half your criteria.
• Exclude early. exclude_industries costs nothing and keeps agencies, recruiters and consultancies out of a list you would otherwise clean by hand.
• Filter on the team, not the company. department plus min_department_headcount finds accounts that actually staff the function you sell to. Headcount alone does not — a 300-person company with six people in sales is not a sales-tooling buyer.
• top_technologies beats technologies for qualification. technologies matches any detection, including a tag left on one page. top_technologies matches what the company actually runs.
• One department per call. The department filters all read the same department value, so run separate calls to compare sales-heavy and engineering-heavy accounts.
• Set limit explicitly. Default is 25. limit is the cost dial at 0.5 credits per company, so ask for volume only when you will use it.
• Ask for the raw columns. “Return company_name, primary_domain, employee count, country, last round and date — as CSV.” Left to itself your client writes a paragraph about the market and you lose the fields you paid 0.5 credits each for.
• Search before you enrich. 100 companies here is 50 credits; guessing 100 domains and enriching them at 2 credits each is 200.

Output

{
  "companies": [
    {
      "company_name": "Example Inc",
      "primary_domain": "example.com",
      "employee_count_range": "51-200",
      "countries": ["United States"]
    }
  ],
  "next_cursor": "eyJwYWdlIjoyfQ"
}

Credits and limits

0.5 credits per company returned. limit is 1–100 with a default of 25, so a default call costs 12.5 credits and a full page costs 50. Run check_credits (free) before a long paging run.

Parameters

No parameter is required, but a call with no filters has nothing to match on. Combine filters from any of the families below.

Filters combine as AND across families and any-of within a single array filter. industries: ["Software Development", "Financial Services"] matches either industry; adding countries narrows that set rather than widening it.

Identity

Category

Location

Type

Size

Revenue

Funding

Social

Age

Growth

Department headcount

Filters on the company’s employee mix rather than its total size — a 300-person company with 60 in sales is a different account to a 300-person company with 6.

Accepted department values: accounting, administrative, arts_and_design, business_development, community_and_social_services, consulting, customer_success_and_support, education, engineering, entrepreneurship, finance, healthcare_services, human_resources, information_technology, legal, marketing, media_and_communication, military_and_protective_services, operations, product_management, program_and_project_management, purchasing, quality_assurance, real_estate, research, sales, support.

Technographics

Competitors

Data freshness

Paging

Filter reference

Several filters only match values from a fixed list. A near-miss does not error — it silently matches nothing, which reads like “no companies found”. These are the full accepted value lists, hosted as JSON.

Pass the name, not an id — largest_headcount_countries: ["United States"], not ["USA"].

To search one locally:

curl -s https://dyyboia6z5eq9.cloudfront.net/mcp/filters/find_companies/static-linkedin-industries.json \
  | jq -r '.[]' | grep -i software

industries and exclude_industries are a separate, differently-worded taxonomy from linkedin_industries — do not mix strings between the two.

Related tools

Next steps

• All MCP tools — the full catalogue with credit costs
• TAM Sourcing — workflows that turn a company search into a working list

Keep learning

Two courses take these tools past the reference page — clustered by GTM job, then chained into workflows.