Exact Match
Guide

Exact Match MCP Server for Consumer Data: Tools & Setup

What the server does, every public tool, how to add it to Claude, the limits, and how it compares with the B2B data vendors' MCP servers.

September 2026Updated November 22, 20269 min read

The Exact Match MCP server is a remote Model Context Protocol server that connects AI assistants such as Claude to Exact Match's 178M+ U.S. consumer profiles and 13M+ U.S. business records. Once it's connected, the assistant can search 60,000+ consumer targeting attributes, preview audience counts, resolve and enrich contact lists, and export files, all from a plain-English request.

Last updated: November 2026

What the Exact Match MCP server does

An MCP server publishes a set of tools, each with a name and a description. An AI client reads that list and decides which tool to call when you ask it something. You type "find families in these ZIP codes who are interested in outdoor recreation," and the client turns that into tool calls against the server instead of writing an answer from memory.

Behind Exact Match's tools sits its consumer data graph: 178M+ U.S. consumer profiles across 9 people data domains (affinity, demographic, education, employment, financial, household, intent, political and purchase), with geography as a separate layer, plus 13M+ U.S. business records. The data is updated with each data release. The server endpoint is https://mcp.exactmatch.io/mcp/v2, and it signs you in with OAuth using your own Exact Match account.

The tool list

These are the tools Exact Match publishes in its public MCP tools reference, grouped the way that reference groups them. Parameters and response formats are documented there, not here, because they change faster than a blog post does.

  • Resolution: entity_resolve matches emails, phones, addresses or device IDs to Exact Match profiles and returns a confidence score for each match. resolve_and_enrich_rows takes an uploaded CSV and returns it enriched, one row out for each row in. entity_relations looks up relationships such as employer and household.
  • Enrichment: entity_enrich fetches profiles and data domains for known people or businesses.
  • Discovery: trait_search finds audience segments from a natural-language description. trait_get returns size, prevalence and skew for segments and resolves geographic boundaries (state, ZIP, county, DMA and more). entity_find finds people or businesses that match a boolean expression of traits, with a preview mode and an export mode. entity_traits checks which traits a set of people or businesses carry.
  • Analysis: group_entities_by_trait totals trait frequencies across an audience, and calculate_trait_lift compares an audience with the general U.S. population to show which traits are over- or under-represented.
  • Utility: generate_upload_url, generate_download_url, geocode_location, check_export_status, list_exports, get_export_download_url and submit_feedback.
  • Billing and account: check_balance, purchase_credits (it returns a checkout link, so you pay on the website, never inside the chat), get_upgrade_options, get_user_profile, get_subscription, get_billing_overview, list_billing_history, get_usage_history, list_credit_usage_all, get_usage_by_module, list_audit_logs, get_available_modules, list_organizations, switch_organization, and three export-template tools (list_export_templates, get_export_template, get_default_export_templates).

Every tool tells the AI client whether it only reads or can spend credits, so the client knows when to ask you before calling it. The server also ships guided workflow prompts, including audience_build, icp_analysis and contact_enrich, that you can ask Claude to run by name. If you want the same capabilities from the REST side, the consumer data API post covers the four products behind them.

How to add the Exact Match MCP server to Claude

Claude adds remote MCP servers as custom connectors. Anthropic's help article on custom connectors using remote MCP says they're available on Free, Pro, Max, Team and Enterprise plans (Free accounts can add one), and it has the current menu names if they change. For a personal account:

  1. In Claude, open Customize > Connectors.
  2. Click + Add, then Add custom connector.
  3. Name it "Exact Match" and enter the server URL https://mcp.exactmatch.io/mcp/v2.
  4. Keep the default sign-in options and click Add.
  5. On the new connector, click Connect, sign in with your Exact Match account, and click Allow on the "Connect to Exact Match?" screen.

To use it in a chat, click the + button, choose Connectors, and switch Exact Match on. A good first message is "What can Exact Match do? Check my account status."

Claude Team or Enterprise. Members can't add custom connectors themselves. An Owner adds Exact Match once under Organization settings > Connectors, and each member then finds it under Customize > Connectors and clicks Connect, signing in as themselves. Exact Match's organization setup guide walks an Owner through it.

Claude Desktop and Claude Code. Claude Desktop uses the connectors on your claude.ai account. In Claude Code, run claude mcp add --transport http exactmatch https://mcp.exactmatch.io/mcp/v2, then run /mcp inside Claude Code and sign in, as the Claude Code setup page describes.

Exact Match also publishes setup steps for a Slack bot (enabled per workspace on request) and for direct programmatic access with API keys, which is on the Enterprise plan. Its docs don't give setup steps for other assistants, so the paths above are the ones we can vouch for. The full menu of paths is in the Claude integration overview.

A worked example: size an audience, then export it

Here's a realistic request from a marketer planning a spring campaign for an outdoor-gear retailer:

"Find households with children in the Columbus DMA who are interested in camping or hiking. Show me how many there are and a few sample records. If the count looks right, export it as a CSV."

This is what the assistant does with the tools, following Exact Match's published audience build workflow:

  1. Find the segments. It calls trait_search for "camping or hiking interest" and "households with children" and picks the matching segments from the results.
  2. Pin the geography. It calls trait_get to turn "Columbus DMA" into a geographic trait, rather than guessing at a radius.
  3. Preview the count. It combines the traits into one expression (children AND the DMA AND (camping OR hiking)) and calls entity_find without an export format. That returns the audience size and sample records, and it costs no credits.
  4. Stop and ask. It shows you the count and the samples and waits. The workflow tells the assistant not to export without your explicit yes. This is the moment to tighten or loosen the definition, and every re-count along the way is free.
  5. Check the balance. Before exporting, it can call check_balance to confirm your remaining credits cover the file.
  6. Export. On your go-ahead, it calls entity_find again with the same expression and format: "csv". That starts a background export job and uses credits, which are charged when you download the file, for the rows delivered.
  7. Deliver. It polls check_export_status until the job finishes and hands you a download link. Links expire, so download the file promptly. If a saved export's link lapses, list_exports and get_export_download_url get you a fresh link without re-running the export.

One habit worth keeping: treat the CSV, not the chat summary, as the deliverable. A model can round a count or describe a trait more loosely than the data defines it. Open the file and check a handful of rows. For more on how agents run this loop, see the AI agent audience data API guide.

Limits, stated plainly

  • Plans: every Exact Match plan includes MCP access, along with every product. Plan prices are on the pricing page.
  • Rate limit: 30 requests per minute per user. The limit is per user, not per connection, so Claude, Claude Code and any other client signed in to the same account share it. That's plenty for conversational research. For long lists, use resolve_and_enrich_rows, which processes a whole file in one call instead of one call per row.
  • Credits: searches and counts are free. Exports and lookups use credits, including lists you upload for resolution or enrichment. At zero credits, new exports and lookups pause until you add a top-up pack or the next cycle starts; searches and counts keep working.
  • API keys: Enterprise plan only. Signing in to the MCP server through Claude uses OAuth with your own account and doesn't need an API key. API-key limits can be raised by agreement.
  • Coverage: U.S. only. Exact Match describes its data handling as CCPA compliant; what that covers, and what stays with you after an export, is in our guide to the CCPA compliant consumer data platform.

Agencies and resellers. Subaccounts give each downstream client a scoped child identity under one parent API key, with its own fair-share rate limit and export concurrency, data isolation between subaccounts, and billing and credit tools hidden from them. Subaccounts run through the API, which is on the Enterprise plan, so they work with API keys, not from inside an assistant app like Claude. The best consumer data platform for agencies guide covers the other agency checks.

Reselling or white-labeling Exact Match under your own brand runs under a short signed Reseller Addendum. Contact sales to set it up.

How it compares with other data vendors' MCP servers

Several data vendors now run official MCP servers. The useful distinction isn't which one is better. It's what kind of data each one serves. ZoomInfo, Apollo and Lusha are B2B prospecting databases: companies, decision-makers, work emails and direct dials. Exact Match is built around U.S. consumer audiences. If your job is reaching a VP of Sales at a software company, a B2B tool fits better than Exact Match. If your job is building an audience of households, Exact Match fits better. Here's what each vendor's own pages said when we checked in November 2026:

VendorOfficial MCP server?Data it servesHow you connect
ZoomInfoYes, ZoomInfo MCPB2B company, contact and technographic data, plus intent signalsConnector in Claude, ChatGPT, Perplexity and Replit; sign in with ZoomInfo credentials
ApolloYes, Apollo MCPB2B people and company records, plus Apollo's outreach tools (sequences, tasks, lists)Built-in connectors in Claude, ChatGPT, Perplexity, Replit and Cursor; per-user OAuth, or an API key for server-side use
LushaYes, Lusha MCPB2B contacts and companies, including verified emails and phone numbersOAuth in Claude, ChatGPT and Codex; API key for other clients, drawing on the same credit pool as Lusha's API
SalesgenieYes, Salesgenie MCPBusiness and consumer data, searchable in the same conversationCustom connector in Claude or ChatGPT; sign in with Salesgenie credentials; enrichment needs a paid subscription
Exact MatchYes, Exact Match MCP178M+ U.S. consumer profiles and 13M+ U.S. business recordsCustom connector in Claude, Claude Desktop or Claude Code; OAuth with your Exact Match account; included in every plan

Salesgenie is the closest comparison, because it's the only one of the four that also serves consumer data. Buying a list there returns a checkout link on Salesgenie's site. On Exact Match, exports draw on your plan's monthly credits, and the tool set leans toward audience analysis, such as finding which traits define your existing customers with calculate_trait_lift and then finding more people like them. For the wider picture of how B2B and B2C data vendors differ, see the B2B vs B2C data provider comparison.

Is it worth setting up?

For a team that builds consumer audiences regularly, yes, with one caveat: the gain is speed of iteration, not better judgment. The server lets you test a segment idea in a minute instead of a ticket cycle, and the counts that tell you whether an idea is viable cost nothing. It won't tell you whether the idea was good. You still need someone who knows the customer to look at the traits and say "no, that's not them."

Frequently Asked Questions

What is the Exact Match MCP server?

It's a remote Model Context Protocol server that connects AI assistants such as Claude to Exact Match's 178M+ U.S. consumer profiles and 13M+ U.S. business records. The assistant can search targeting attributes, preview audience counts, resolve and enrich contact lists, and export files from a plain-English request.

How do I add the Exact Match MCP server to Claude?

In Claude, go to Customize > Connectors, click + Add, choose Add custom connector, and enter https://mcp.exactmatch.io/mcp/v2 as the server URL. Then click Connect, sign in with your Exact Match account, and click Allow. On a Claude Team or Enterprise plan, an Owner adds the connector for the organization and each member clicks Connect.

Which plans include Exact Match MCP access, and what uses credits?

Every Exact Match plan includes MCP access along with every product, and plan prices are on the pricing page. Searches and counts are free. Exports and lookups use credits, and when credits run out, new exports and lookups pause until a top-up pack or the next cycle.

What are the rate limits on the Exact Match MCP server?

MCP and API access is limited to 30 requests per minute per user, and every client signed in to the same account shares that limit. API keys are on the Enterprise plan only, and API-key limits can be raised by agreement.

How is it different from the ZoomInfo, Apollo and Lusha MCP servers?

ZoomInfo, Apollo and Lusha all run official MCP servers, and all three serve B2B company and contact data for sales prospecting. Exact Match's server is built around U.S. consumer (B2C) audiences. Salesgenie's MCP connector is the closest comparison, since it serves both business and consumer data.

Do I need a developer to use a consumer data MCP server?

Not for day-to-day use. Connecting Exact Match to Claude takes a server URL and a sign-in, and after that you ask for audiences or enrichment in plain language. A developer is only needed for programmatic access with API keys, which is on the Enterprise plan. For the loop an agent runs once it's connected, see ai agent audience data api.

Get Started with Exact Match

Every plan includes all four products, and plan prices are on the pricing page. Schedule a consultation to talk through which plan fits.