Data sourced from Dakota Marketplace, the global LP and GP intelligence platform trusted by thousands of investment professionals. Learn More | Book a Demo
Ask Microsoft Copilot how much a family office has under management, or who runs manager selection at a state pension, and you will get an answer. It will read well. It may also be wrong, and nothing in the response will tell you which.
That is not a failure of the model. Copilot has no line into institutional investor data, so it does what any model does without a source and fills the gap. The fix is to give it one. With the Dakota Marketplace Copilot Connector, Copilot answers from verified allocator records instead of approximating them, and it does so inside Teams and Microsoft 365 Copilot where your team already works.
In this article, we’ll cover what the connector gives Copilot access to, how the data stays current, the questions worth asking first, and how to connect it in about 15 minutes.
What’s Now Available With Dakota Marketplace
The Dakota Marketplace Copilot Connector gives Copilot direct access to Dakota Marketplace data: allocator accounts, verified contacts, fund strategies, holdings, transactions, institutional searches, public plan minutes, Form ADV and Form D filings, job changes, and fundraising news.
Two things are worth knowing up front. The connector is set up through Microsoft Copilot Studio rather than installed from an app directory, so one person builds the agent and publishes it to the rest of the team. And access is personal. Each person signs in with their own Dakota Marketplace credentials, so the agent returns only what that individual’s membership already covers.
How It Works With Dakota Marketplace
This is a live connection, not a one-time export. When someone asks about allocators in a strategy or which firms are fundraising now, the agent queries Dakota Marketplace in the moment and answers from what comes back. Dakota’s research team updates the data daily, and the answer reflects that, not a file someone pulled last quarter.
The difference shows up in what you can do with the answer. A guess has to be checked before you act on it. A sourced record comes with the account, the contact, the title, and the filing behind it, which means the next step is outreach rather than verification. Dakota also connects to Claude and ChatGPT the same way, so teams split across tools are working from one dataset.
Not a Dakota Marketplace member yet? Book a demo and we’ll show you the data your Copilot agent would be answering from.
What to Ask It
Prospecting and list building. “Find public pension funds in the Midwest that invest in private credit.” “Which firms have private equity buyout funds that are open and fundraising now?”
Account and contact research. “Who are the CIOs at family offices in the San Francisco Bay Area?” “Who has joined [firm name] this year?”
Meeting prep and outreach. “Summarize this month’s private equity fundraising news.” “List closed software buyout transactions from 2026.”
How to Connect It
Setup takes about 15 minutes:
- Go to copilotstudio.microsoft.com, pick your environment, and create a new agent.
- In the agent’s settings, confirm generative orchestration is turned on.
- Open Tools, add a new tool, and select Model Context Protocol. Enter https://marketplace-mcp.dakota.com/mcp as the server URL, then choose OAuth 2.0 and Dynamic discovery.
- Create the connection and sign in with your Dakota Marketplace username and password.
- Publish the agent and add it to the Teams and Microsoft 365 Copilot channel so your team can use it.
One thing to check before you start: some organizations block custom connectors through Power Platform data policies, and your Teams admin may need to approve the published agent. Worth a note to IT ahead of time. The full setup walkthrough covers every step in detail.
Copilot can reach institutional investor data. It just needs to be pointed at it. If you want to see what that looks like against your own target list, in Copilot, Claude, ChatGPT, or your CRM, book a demo.
