Choose Your Path
The D&B Commercial Graph is accessible two ways. Choose the one that fits how your team works.
Model Context Protocol (MCP)
Access Dun & Bradstreet Data in Your Existing AI Tools
Connect the D&B Commercial Graph™ MCP server to your AI platform of choice: Claude, Cursor, Codex, Gemini, Microsoft Copilot Studio, Databricks, or any MCP-compatible tool. Your AI agent will call D&B tools directly within your existing workflow, so you can resolve entities, pull firmographic data, assess risk, and build automated skills without leaving your development environment.
This option is best for developers, data engineers, and technical teams who want to integrate trusted business data into their own AI workflows and applications.
Prompt Interface
Access Dun & Bradstreet Data Through a Prompt Interface
Explore Dun & Bradstreet data in a conversational AI environment hosted by Dun & Bradstreet. Ask questions about companies and see verified data come back in real time. No setup, no integration, no credentials required.
This option is best for business teams, analysts, and evaluators who want to explore D&B data and test use cases without the need for a technical integration.
Understanding Data Blocks
The D&B Commercial Graph packages business intelligence into Data Blocks -- structured, governed data sets designed for machine consumption. Each Data Block is entity-resolved through the Dun & Bradstreet D‑U‑N‑S® Number, so your AI always knows exactly which company it's working with.
Data Blocks cover company fundamentals, financial health, payment behavior, corporate structure, risk and compliance, contacts, sales and marketing, ESG indicators, and diversity.
The D‑U‑N‑S Number
The D‑U‑N‑S Number is D&B's unique nine-digit identifier for every entity in the D&B Commercial Graph. When an AI model resolves a company, for example "Gorman Manufacturing," to a D‑U‑N‑S Number, it knows exactly which entity it's reasoning about. No fuzzy matching, no hallucinated merges, no ambiguity.
Understanding Model Context Protocol
MCP is a standard framework that connects AI agents to the data sources they need, acting as a universal adapter that provides the context required for accurate, relevant results. It's often compared to an API, and the two do share a core purpose: both let systems exchange data. But where a traditional API is a general-purpose connector — defining how any two pieces of software talk to each other, one request and response at a time — MCP is purpose-built for AI and large language models. Rather than just moving data back and forth, it's designed to feed models the surrounding context they need and to help them hold onto it across tasks and interactions.
In short, an API hands over the data; MCP hands over the data and the context that makes it useful to an AI.