Looking to make your Microsoft Copilot smarter by connecting it directly to your Dataverse tables? You’re in the right place! In this step-by-step guide, we’ll show you exactly how to integrate Dataverse data into Microsoft Copilot Studio, empowering your Copilot agents to access and use your organization’s core business information securely and effectively.
By the end of this post, you’ll know how to set up Dataverse as a knowledge source and understand how Copilot respects your security privileges throughout the process.
Why Connect Dataverse to Copilot?
Microsoft Dataverse is a powerful, cloud-based storage platform that underpins many Microsoft business applications, like Dynamics 365. Integrating Dataverse into Copilot Studio allows your AI agents to:
- Access real-time, structured business data.
- Provide more personalized, relevant, and accurate responses.
- Leverage your existing investments in data security and governance.
Plus, Copilot fully respects existing security roles and permissions. This ensures that users interacting with your Copilot agent can only access the Dataverse information they’re authorized to see — maintaining your company’s data security posture.
Two Ways to Connect Dataverse to Copilot
In this tutorial, we’ll focus on adding Dataverse tables as a knowledge source — one of the simplest and most powerful ways to enable Copilot to access Dataverse data.
Step-by-Step: How to Add Dataverse Tables as a Knowledge Source
1. Open Microsoft Copilot Studio
First, head to Microsoft Copilot Studio. Select the Copilot you want to configure — for this guide, we’ll use a sample called “Coffee Copilot.”
Imagine you want your Coffee Copilot to access the Account table stored in Dataverse. Here’s how you do it.
2. Go to the Knowledge Section
Navigate to the Knowledge tab inside your Copilot project.
This section manages the various information sources — documents, websites, PDFs, and Dataverse tables — that your Copilot can use to generate answers.
3. Add Knowledge Source
Click “Add Knowledge” and you’ll see an exciting preview feature: Dataverse.
Select Dataverse to proceed.
4. Choose Dataverse Tables
Here, you can select up to 15 Dataverse tables for Copilot to access.
✅ In our example, we’ll choose the Account table.
After selection, you’ll get a preview showing:
- Top 20 records.
- First 20 columns (alphabetically ordered).
Don’t worry — this is just a preview! Copilot will ingest all fields and records from the selected Dataverse table, not just the previewed ones.
5. Name and Describe the Knowledge Source
Next, you’ll be asked to:
- Set a Name (what appears in your knowledge list).
- Provide a Knowledge Description.
Why is the Knowledge Description Important?
The description helps Copilot’s generative AI better understand how to use the data.
Think carefully about the terms and context your users would naturally use. For example, instead of saying “Account Table,” you might describe it as “a list of customer accounts and their associated details.”
➡️ Tip: Use simple, conversational language that aligns with how your users talk about your business data.
6. Finalize and Save
After adding the name and description, hit Next and Save.
Congratulations! Your Copilot can now access and reference Dataverse information in conversations and actions — all while respecting user permissions.
Important Notes on Security and Access
When you connect Dataverse to Copilot:
- Security Roles Apply: Copilot only surfaces data the user is allowed to see.
- No Extra Risk: Existing Microsoft Dataverse permissions remain enforced automatically.
- Selective Access: Choose only the tables you want Copilot to use, keeping your project organized and relevant.
Best Practices When Adding Dataverse to Copilot
✔️ Select only the necessary tables.
More tables = more complexity for the AI. Start small and scale thoughtfully.
✔️ Write clear, user-friendly descriptions.
This improves AI comprehension and response quality.
✔️ Review security permissions.
Ensure users have appropriate access before surfacing sensitive data through Copilot.
✔️ Update knowledge regularly.
Keep your data fresh to ensure Copilot delivers accurate, up-to-date answers.
In Conclusions
Connecting Microsoft Copilot to Dataverse unlocks powerful new capabilities for your AI-driven solutions. By following this guide, you’ll make your Copilot smarter, faster, and more relevant — while ensuring your data stays secure. This integration is perfect for teams looking to automate customer service, streamline internal operations, and deliver incredible AI-powered experiences without starting from scratch.
Happy Developing!