PitchBook Data Without a Subscription: Look Up Companies and Investors

PitchBook data without a subscription means the free public preview PitchBook shows for a profile, read as JSON instead of in a browser. Send a profile id and get a company's description, status, address, industries and parent company, or an investor's status and types. Valuation and funding sections stay paywalled and come back empty. The PitchBook company profile preview page is the reference for exact fields, request shape and prices.

The short version

• —What you get: the public preview of a PitchBook company or investor profile as JSON, plus a name search that returns profile ids. Over REST or as MCP tools.
• —What you do not get: round amounts, valuations, cap tables, portfolio lists or fund performance. Those sit behind the licence. This is not a cheaper copy of a subscription and does not claim the same coverage.
• —Who it is for: teams that need to identify a company or investor, check whether it is still independent, and fill gaps in a list or CRM, from code or from an agent.
• —What it costs: 5 points for a company or investor preview, 1 point per profile a search returns. A full find-and-read is 6 points when the search returns one hit.

PitchBook data without a subscription, step by step

"PitchBook data" can mean very different things. A deal professional means the full database: round-level financing, valuations and fund performance. A developer building a lead list or a research agent usually means something much smaller: which entity is this, what does it do, where is it and who owns it. The three sources below answer the second question and stay quiet on the first.

From the source pages, at the time of writing. The exact parameter names and defaults are on those pages and in the reference at docs.neuralverge.ai.

1. Search by name to find the profile id

The profile search takes a keyword, an optional type of company, investor or fund, and a max_items cap. It returns identifiers only: a name, a type, a profile id, a link and a logo. It does not return a description or a status. The id is the code at the end of a profile address, such as 439834-78 in the address of a company profile. If you already hold ids, for example from a spreadsheet or a CRM field, skip this step.

2. Check the type and the link

Matching is on the profile name, so unrelated entities that share a word appear together. For the keyword stripe capped at two records, the captured response returned two company profiles and a total_points of 2, and total_found came back null because the platform did not report a total. Read type and url before you trust a hit, and prefer a more specific keyword to a short one.

3. Read the preview

Pass the id or the profile link to the company preview or the investor preview. The page is read when you ask, not served from a stored copy, so the response reflects the profile as it stands at that moment. The response also reports how it went: blocked is false when data came back and true if the platform refused the request, pages_opened says how many pages were read, and proxy_traffic_bytes reports the data transferred, for information only.

4. Follow the ownership trail

For a company, status and parent_company tell you whether it was acquired and by whom. The parent comes back as its own id, name and link, so you can pass that id to the same company preview and walk up one level.

A worked example: Acme Oy and a target list

Consider a hypothetical analyst at a Helsinki venture fund who keeps a list of Finnish software companies in a spreadsheet. Acme Oy is an illustrative company in that list, invented to show a workflow and not to describe a real business. No figures here come from a real profile.

She has company names but no ids. For each name she runs the profile search with a small max_items, so a bad match costs a point or two. She reads back the type and link, rejects anything that is not a company, and keeps the id she trusts. For the companies that survive, she runs the company preview, which adds the fields her sheet was missing: founding year, status, employee count, primary industry and address. For any company whose ownership_status shows it was acquired, she reads parent_company and notes the parent, which changes who she should be talking to.

She does not expect the preview to say how much anyone raised. For that she goes to the licensed database or a second source. The startup and funding data guide covers the other platforms in this category, which are separate sources with the same request format.

One real request: a company preview

Here is one request and the response that came back, taken from the company preview source page. It asks for the company with profile id 439834-78.

curl -X POST https://api.neuralverge.ai/functions/v1/run-pitchbook-companies \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"company":"439834-78"}'

The response below is copied from the source page. The session_id is a placeholder in the page's sample, and the sample is a real captured response for a small company, not an Acme Oy example.

{
  "session_id": "11111111-1111-1111-1111-111111111111",
  "kind": "pitchbook_companies",
  "company": "439834-78",
  "company_id": "439834-78",
  "human": "# PitchBook — company profile preview\n\n- **Id:** 439834-78\n- **Pages opened:** 1\n- **Proxy traffic:** 228.0 KB\n- **Name:** Stripe Theory\n- **Founded:** 2015\n- **Status:** Acquired/Merged\n- **Employees:** 25\n- **Industry:** Media and Information Services (B2B)",
  "machine": {
    "company": {
      "id": "439834-78",
      "name": "Stripe Theory",
      "url": "https://pitchbook.com/profiles/company/439834-78",
      "description": "Provider of data monitoring services intended to help brands define and execute on unique positioning. The company provides global monitoring across social media, search results, news media, and the d",
      "logo": "https://image.pitchbook.com/llGaQS51jLDbvGoisogvTOq35qI1785304888993_200x200",
      "website": "http://www.stripetheory.com",
      "year_founded": 2015,
      "status": "Acquired/Merged",
      "address": {
        "lines": [
          "1200 Spring Street Northwest",
          "Atlanta, GA 30309"
        ],
        "street": "1200 Spring Street Northwest",
        "city": "Atlanta",
        "region": "Georgia",
        "postal_code": "30309",
        "country": "United States"
      },
      "social": {
        "linkedin": "https://www.linkedin.com/company/stripetheory",
        "twitter": "https://twitter.com/StripeTheory"
      },
      "faqs": [
        {
          "question": "When was Stripe Theory founded?",
          "answer": "Stripe Theory was founded in 2015."
        },
        {
          "question": "Where is Stripe Theory headquartered?",
          "answer": "Stripe Theory is headquartered in Atlanta, GA."
        }
      ],
      "details": {
        "formerly_known_as": "Stripe Reputation",
        "ownership_status": "Acquired/Merged (Operating Subsidiary)",
        "financing_status": "Corporate Backed or Acquired",
        "primary_industry": "Media and Information Services (B2B)",
        "other_industries": "Consulting Services (B2B)"
      },
      "counters": {
        "year_founded": 2015,
        "employees": 25,
        "financing_rounds": 1
      },
      "verticals": [],
      "sections": {
        "valuation_and_funding": {
          "title": "Valuation & Funding",
          "total": null,
          "shown": null,
          "paywalled": true,
          "intro": null,
          "tables": [
            {
              "label": null,
              "total": null,
              "columns": [
                "Deal Type",
                "Date"
              ],
              "rows": []
            }
          ]
        }
      },
      "patents_summary": null,
      "related_research": [],
      "news": [],
      "employees": 25,
      "latest_deal_type": "M&A",
      "financing_rounds": 1,
      "formerly_known_as": [
        "Stripe Reputation"
      ],
      "ownership_status": "Acquired/Merged (Operating Subsidiary)",
      "financing_status": "Corporate Backed or Acquired",
      "primary_industry": "Media and Information Services (B2B)",
      "other_industries": [
        "Consulting Services (B2B)"
      ],
      "parent_company": {
        "id": "340778-08",
        "name": "Acceleration Community of Companies",
        "url": "https://pitchbook.com/profiles/company/340778-08"
      }
    },
    "blocked": false,
    "pages_opened": 1,
    "proxy_traffic_bytes": 233424
  },
  "total_points": 5
}

Reading the response

Start with the envelope. kind names the source, company and company_id echo what you sent, and total_points is 5, the price of one successful company preview. blocked is false and pages_opened is 1, so one page was read and the platform answered.

Inside machine.company, the identity fields are id, name, url, logo, website and description. The description in this sample ends mid-word, because the source returned a truncated string. The preview gives you the public text as the page shows it, and it is not always complete.

The business fields are year_founded, status, employees, financing_rounds, latest_deal_type, ownership_status, financing_status, primary_industry and other_industries. In this sample the company is an Atlanta-based provider of data monitoring services, founded in 2015, with status Acquired/Merged, 25 employees, one financing round and a latest deal type of M&A. The count of rounds and the latest deal type are there. The rounds themselves are not.

That gap is visible in sections.valuation_and_funding: paywalled is true, the table has the columns "Deal Type" and "Date", and rows is empty. Nothing is filled in to hide the gap, and nothing here should be read as the contents of that table.

The ownership trail is in parent_company, with the parent's id, name and profile link. Former names sit in formerly_known_as. Location is in address, split into lines, street, city, region, postal code and country. Social links are under social, and faqs carries the question and answer pairs the profile shows.

The investor preview and the search

The investor preview works the same way with run-pitchbook-investors and an investor argument. In its captured sample, the investor is a private equity firm in Brooklyn, New York, founded in 2022, with status Actively Seeking New Investments, a primary investor type of PE-Backed Company, another type of PE/Buyout and one professional listed. Its sections object came back empty, so there was no portfolio or deal list in that response. One detail worth knowing: in that sample, investor_type reads Strategic Acquirer while primary_investor_type reads PE-Backed Company. Both are returned as PitchBook shows them, and your code should not assume they agree.

The profile search uses run-pitchbook-profiles-search with a keyword and a max_items cap, and each item has an id, name, url, type and logo.

REST and MCP

Each source is reachable over REST and as a tool on the hosted MCP server (https://mcp.neuralverge.ai/mcp). The request above is the REST form. Over MCP, the same call is a tool named run_pitchbook_companies with the arguments {"company": "439834-78"}. The tool names use underscores where the REST paths use hyphens: run_pitchbook_companies, run_pitchbook_investors and run_pitchbook_profiles_search, and the arguments match the REST body.

That makes an entity-resolution step easy to give an agent. Asked about a company, it can search by name, pick the right id and read the preview, in the same request format it uses for every other source. For setup in Claude and other clients, see the MCP server guide. AI research is REST-only, so if the agent needs open-ended questions answered once the entity is confirmed, send those to the AI research product over REST.

Cost and limits

Pricing is in points, and the plans on the pricing page are monthly allowances. At the time of writing, the Base plan is $50 for 50,000 points a month and the Core plan is $100 for 100,000, with larger plans above them. Check the pricing page for current terms.

From the source pages:

• —Company preview: 5 points per successful run, stated as $5 per 1,000 lookups.
• —Investor preview: 5 points per successful run, stated as $5 per 1,000 lookups.
• —Profile search: 1 point per returned profile, stated as $1 per 1,000 records, capped by max_items.
• —Misses: a lookup is charged only when the profile resolves, and a search that returns nothing costs nothing.

A find-and-read for one company is one search record plus one preview, or 6 points when the search returns a single hit. A search that returns five candidates and one preview is 10 points. Those totals follow from the per-source prices; work them through for your own list before you commit.

Pitfalls

• —Public information only. Gated sections, including the valuation and funding table, come back empty.
• —Fields vary by entity. Several fields come back empty or null for a given profile (in the company sample: verticals, news and patents_summary). Parse defensively and do not treat a missing field as a statement that the thing does not exist.
• —Labels are PitchBook's. Status, ownership status, industries and investor types are PitchBook's own classifications, returned as given, and may differ from another database's labels for the same entity.
• —Truncated text. A description can end mid-sentence, as in the sample above.

How to decide

Check each of these against your own job.

1. —Which fields do you actually need? List them. If any are round amounts, valuations, cap tables, portfolio lists or fund performance, a public preview will not supply them.
2. —Is the job identification? Confirming what a company does, where it is and whether it is still independent before it enters a CRM or a sequence is what the preview is for.
3. —Do you need ownership checks? status and parent_company show whether an acquired business sits inside a larger group.
4. —Are you enriching a list or screening investors? Industry, headcount and founding year for ids you hold, or status, investor types and location for investors a founder is considering, all fit. Deciding who to research further is the next step.
5. —Do you already hold ids? If yes, the preview is a direct lookup at 5 points. If not, budget for the search step as well.
6. —Can your code tolerate empty fields? Run a handful of real profiles and see which fields are null or empty.
7. —What is your volume? Multiply expected searches and previews by their point prices and compare with the monthly allowance on the pricing page.
8. —What is your second source? If funding history matters, decide where it will come from, a licensed database or another source from the catalog, and how you will reconcile disagreements.

Frequently asked questions

Is this the same data as a PitchBook subscription?

No. The three PitchBook sources return the free public preview of a profile. Sections such as valuation and funding are paywalled by PitchBook and come back with empty tables, and the investor sample comes back with an empty sections object. The sources help you identify an entity and read its public description. They are not a substitute for a subscription, and this article does not claim the same coverage.

How do I get a profile id if I only have a name?

Use the PitchBook profile search with a keyword, an optional type of company, investor or fund, and a max_items cap. Each returned profile carries an id, a type, a link and a logo. Pass the id to the company or investor preview. Matching is on the profile name, so check the type and the link before you trust a hit.

What does a lookup cost?

A company or investor preview costs 5 points per successful run, which the source pages state as $5 per 1,000 lookups. A profile search costs 1 point per returned profile, or $1 per 1,000 records. Points come out of the monthly allowance of your plan.

Do I pay when the profile is not found?

The source pages state that a lookup is charged only when the profile resolves, and that a search which returns nothing costs nothing. Searches are charged per profile actually returned, capped by max_items.

Can an agent call these sources over MCP today?

Each source is reachable over REST and as an MCP tool on the hosted MCP server. The tool names are run_pitchbook_companies, run_pitchbook_investors and run_pitchbook_profiles_search, and the arguments match the REST body.

Does the investor preview list an investor's portfolio or deals?

Not in the captured sample. The investor preview returns the description, website, founding year, status, address, investor types and the number of professionals, and its sections object came back empty for the sampled investor. Portfolio, deals and co-investors were not populated.