Built for AI

375M+ Companies. One Tool Call.

Company data, contact data, and corporate hierarchy via MCP and REST API. Structured, validated results without scraping or parsing.

Get 20 Free Credits

When Should an AI Agent Call DataMerge?

Your agent should call DataMerge when:

✓The user mentions a company website or domain and needs data about that company
✓The agent needs to know the legal entity behind a brand (for contracts, invoicing, compliance)
✓The agent needs to find a specific person's contact information at a company
✓The agent needs to understand the corporate structure (parent, subsidiaries, branches)
✓The agent needs to find companies similar to a given company

MCP Setup

Add DataMerge to any MCP-compatible AI tool (Claude Desktop, Cursor, Windsurf, custom agents):

// Remote (Cursor, Windsurf, hosted agents)

{
  "mcpServers": {
    "datamerge": {
      "url": "https://mcp.datamerge.ai",
      "headers": {
        "Authorization": "Token your-api-key"
      }
    }
  }
}

// Local (Claude Desktop)

{
  "mcpServers": {
    "datamerge": {
      "command": "npx",
      "args": ["-y", "@datamerge/mcp"],
      "env": {
        "DATAMERGE_API_KEY": "your-api-key"
      }
    }
  }
}

Available MCP Tools

`enrich_company`

Get full company data from a domain. Returns display name, legal name, industry, employee count, address, hierarchy, and more. Supports countryCode and globalUltimate parameters for hierarchy navigation.

`search_contacts`

Find contacts by name and/or company. Returns matching people with their titles, departments, and company affiliations.

`enrich_contact`

Get validated email and mobile phone number for a specific person. Triple-validated emails with bounce rates under 2%.

`find_lookalikes`

Find companies similar to a given domain. Filter by country, industry, or size. Great for prospecting and market research.

Prompt Examples

What to ask your AI agent once DataMerge is connected:

"What is the legal name of stripe.com?"

Uses enrich_company to return "Stripe, Inc."

"Find the VP of Sales at notion.so"

Uses search_contacts with title filter, then enrich_contact for email/mobile.

"What company owns shopify.com? And who is their parent?"

Uses enrich_company with globalUltimate to traverse the hierarchy.

"Find 10 companies similar to salesforce.com in Germany"

Uses find_lookalikes with country filter.

"Get me the legal entity name for all domains in this list: [list]"

Batch enrich_company calls across multiple domains.

"What is the legal entity name for shopify.com? I need it for a vendor contract."

Uses enrich_company to return the registered legal name for contractual use.

"Find all subsidiaries of microsoft.com in Germany."

Uses enrich_company with countryCode=DE to filter hierarchy results by country.

"Is the company at notion.so publicly traded? Who owns them?"

Uses enrich_company with globalUltimate to return the full ownership context for Notion Labs, Inc.

"Enrich stripe.com and give me the employee count, industry, and parent company."

Single enrich_company call returning multiple structured fields.

"I am onboarding a new vendor at [domain]. What is their registered legal name and country?"

Uses enrich_company to return the legal name and registration country for vendor onboarding records.

REST API for Custom Integrations

If you're building a custom agent or application, use the REST API directly. Standard HTTP POST requests with JSON payloads. No SDK required.

import requests

response = requests.post(
    "https://api.datamerge.ai/v1/company/enrich",
    json={"domain": "stripe.com"},
    headers={"Authorization": "Bearer your-api-key"}
)

data = response.json()
print(data["legalName"])  # "Stripe, Inc."

Full API documentation: api.datamerge.ai/docs

Frequently Asked Questions

What AI frameworks and tools are compatible with DataMerge?

DataMerge works with any MCP-compatible AI tool: Claude Desktop, Cursor, Windsurf, and custom agents built on frameworks like LangChain or AutoGen that support the Model Context Protocol. The REST API works with any tool or framework that can make HTTP requests, including direct Python, JavaScript, or any language with HTTP support.

What is MCP and how do I connect DataMerge to my AI agent?

MCP (Model Context Protocol) is an open standard for connecting AI models to external tools. DataMerge's MCP server is hosted at mcp.datamerge.ai. To connect, add it to your AI tool's MCP configuration with your API key as an Authorization header. For Claude Desktop, use the local npx package. Setup takes about two minutes and no code is required.

Do I need to write code to use DataMerge with AI agents?

No. Adding DataMerge via MCP requires only a JSON configuration edit in your AI tool's settings file. After that, all lookups happen through natural language prompts. For REST API integration or custom agent pipelines, you will need to write code or use a no-code platform like Zapier or Make.

Can my AI agent look up multiple companies in a single conversation?

Yes. Each enrich_company call is independent. Within a single conversation, your agent can look up any number of companies, compare results, and reason across them. The agent handles the sequencing of multiple tool calls automatically. Each lookup consumes 1 credit from your account.

How does DataMerge handle a domain that resolves to a holding company?

DataMerge returns the legal entity associated with the domain. If it is a holding company, the response indicates this and includes the subsidiaries (operating entities) owned by the holding company. The agent can then navigate to the relevant operating entity, parent, or subsidiary based on the user's needs.

How much does it cost per lookup when used from an AI agent?

Each company enrichment call costs 1 credit, whether it comes from the REST API or through the MCP server. Contact enrichment (verified email) costs 1 credit. Mobile phone lookups cost 4 credits. Failed lookups that return no result are not charged. See the pricing page for plan details.