The question comes up the first time someone hears the word MCP: does this replace the API we already planned to build? It does not. Both reach the same records, through the same entitlement checks, drawing on the same credit pool. What separates them is who does the asking.
MCP is asked by a person, in a sentence, inside an AI tool they already have open. The API is called by code you write and deploy. That difference decides almost everything else: setup time, output shape, who on your team owns it, and what it can do at volume.
The Short Version
Use MCP when a human is in the loop and the answer changes depending on who's asking. A seller researching an account before a call, an AE mapping a buying committee, a marketer sizing a segment: all MCP.
Use the API when a system is in the loop and the output feeds something else. Nightly CRM enrichment, a lead-routing service, an agent pipeline your engineers maintain: all API.
Most enterprise teams end up with both. They serve different people.
What MCP Actually Is
MCP stands for Model Context Protocol. Anthropic created it and published it as an open standard, and it is not a ZoomInfo product name. Any vendor can run an MCP server, and any AI client can connect to one. Slack, Zoom, and hundreds of other tools already do.
The protocol solves a wiring problem. Before it existed, connecting an AI model to five systems meant five custom integrations, each written to a different spec. MCP replaces that with one interface: the client connects, asks the server what it can do, and the server answers with a list of tools it exposes. That last part matters more than it sounds, and it's the real line between MCP and a REST API. With an API, a developer reads the docs and builds to a fixed spec. With MCP, the model reads the tool list at the start of every session and picks what to call based on the question in front of it.
When ZoomInfo adds a tool, it shows up in that list automatically. Nobody redeploys anything.
Three Surfaces, One Dataset
Before comparing MCP and the API, it helps to separate them from ZoomInfo Copilot, which is where the confusion usually starts.
- ZoomInfo Copilot is the AI experience built into the ZoomInfo platform. Nothing to connect, nothing to configure. You log in and it's there.
- ZoomInfo MCP is a hosted server at
https://mcp.zoominfo.com/mcp. It lets an external AI tool query ZoomInfo in natural language. Every supported client connects to that one endpoint, listed in the connector catalog. - The REST API is a set of HTTP endpoints your developers call directly, documented at docs.zoominfo.com↗.
All three read from the same graph. Copilot and MCP are optional layers on top, and a customer running Copilot today can add either without changing anything about their existing setup.
How MCP and the API Differ
| MCP | REST API | |
|---|---|---|
| Who calls it | A person, in natural language, through an AI client | Code your team writes |
| Who picks the data | The model, from the tool list it reads each session | The developer, at the time they write the call |
| Setup | Connect your ZoomInfo account in the AI tool's connector settings | Register an app, implement OAuth, build and deploy the integration |
| Access requirement | Any ZoomInfo subscription with bulk data credits enabled | ZoomInfo account with API access, which an admin assigns via the DevPortal subscription |
| Output | Conversational, with the model choosing which tool to call | Deterministic JSON against a fixed schema |
| New capabilities | Appear in the next session with no work on your side | Require reading the docs and shipping a change |
| Direction | Read-only | Reads from ZoomInfo, writes into your destination system |
| Volume | Interactive, up to 25 records per enrichment call | Batch and pipeline scale |
| Time to first result | Minutes | A developer sprint |
| Who owns it | The individual user | Engineering or RevOps |
When MCP Is the Right Call
MCP earns its place wherever the request is personal. "Research the accounts I'm meeting with this week" resolves against that rep's calendar. "Show me my renewals and what those accounts said on recent calls" resolves against their book of business. Each user authenticates with their own ZoomInfo login, so the server can answer questions that contain the word my. See User Sign-In vs Service Account for how that authentication model works.
The setup cost is close to zero. A rep connects ZoomInfo in their AI tool's settings and asks a question, with no developer and no deployment in between. That's why MCP tends to reach a team faster than an API project does. It also makes a reasonable first step while a longer integration is still being scoped.
The Getting Started doc covers requirements, and Clients has setup steps for each supported tool.
When the API Is the Right Call
The API wins on volume and write access. A nightly job enriching new CRM records has no person in the loop and no reason to phrase anything in English. Neither does a scoring service that needs the same field, in the same shape, on every call. Determinism is the point there, and a model choosing its own tools works against it.
Three requirements point at the API specifically:
- Data has to land somewhere. MCP is read-only, so anything involving CRM write-back or a warehouse sync is an API job.
- Volume exceeds interactive use. Enrichment through MCP processes up to 25 records per call, which is right for a conversation and wrong for a backlog of 40,000 accounts.
- The output feeds another system. Agents, pipelines, and internal apps need a fixed schema they can parse, not prose.
Building Audiences with the GTM Studio API is a worked example. If you want a custom app on the MCP server instead of the REST endpoints, How to Build Your Own App covers app registration and both OAuth flows.
What Each One Costs
Both draw on the same pool, so running both does not mean paying twice.
Neither carries a separate fee. MCP is included with all ZoomInfo subscriptions. API access is included as well, gated by entitlement rather than by price, since an admin assigns the DevPortal subscription that turns it on. What each consumes is your existing credit allocation: bulk data credits for enrichment, AI action credits for the research agents, and nothing at all for search, lookup, and find-similar. The Tools overview lists the credit type for every tool, free, bulk, or AI action.
MCP requires bulk data credits
MCP does not work with recurring monthly credits. That one gap causes most failed first connections. To see what an account holds, open Admin Portal → Usage → Data Credit Dashboard; if no bulk credits appear there, a ZoomInfo admin or your account team enables them. One more toggle matters. API Access, set per user under Admin Portal → Users → User Management, is what governs MCP connectivity. See Credits & Billing for the full credit model and Security & Data for that toggle.
Records Under Management is what keeps the math honest across surfaces. Once a record is enriched anywhere in the platform, it stays under management for 12 months at the organization level, and no further credit is charged for it in that window. Enrich an account through the API on Monday and pull it up in Claude on Friday, and you paid once.
Check These Before You Connect
Three things decide whether an MCP connection works on the first try.
- The AI tool has to support MCP. The connector catalog lists what connects today. Some tools require a paid tier or admin approval before connectors appear.
- Every user signs in with their own ZoomInfo credentials. Admin-only seats have no data access by design and cannot connect.
- It is read-only. Bulk exports, CRM write-back, and batch enrichment pipelines belong on the API.
Choosing
Answer two questions and the decision makes itself.
Is a person asking, in the moment, about their own accounts? MCP. Is a system asking, on a schedule, on behalf of the whole company? API.
If the honest answer is both, run both. A rep gets ZoomInfo inside Claude this afternoon while engineering scopes the pipeline that will run next quarter. Neither one waits on the other, and the credits they consume land in the same place.