Notifications
Announcements
I'm building a case study generation Copilot using Microsoft Copilot Studio. The Copilot is supposed to fetch detailed information from Dataverse and generate a case study for a customer.
However, I'm noticing that it's often returning generic or irrelevant answers, especially when trying to extract specific fields. For example:
Business challenges are stored in the description field of the Opportunity table, but the Copilot doesn't return this info.
Instead, it generates generic content based on the project topic, ignoring what's actually stored in Dataverse.
I’ve confirmed that the data is present and accessible via the Dataverse connection.
Why might this be happening?
Is this an issue with the way I've set up the prompts or grounding?
How do I ensure the Copilot strictly uses Dataverse content and avoids hallucinations?
Any help or guidance from someone who has built similar bots would be appreciated!
Enabling "General Knowledge" means using the trained data, e.g., 2024.
Yes, disabling this allows the agent to use only the knowledge source you provided.
With an M365 Copilot license, you can enable Enhanced search results in the Generative AI setting to improve semantic search. It has a broader view and understands content patterns, so exact keywords aren't needed.
Under review
Thank you for your reply! To ensure a great experience for everyone, your content is awaiting approval by our Community Managers. Please check back later.
In our never-ending quest to improve we are simplifying the forum hierarchy…
We are honored to recognize Ajay Kumar Gannamaneni as our Community Spotlight for December…
These are the community rock stars!
Stay up to date on forum activity by subscribing.
Michael E. Gernaey 255 Super User 2025 Season 2
Romain The Low-Code... 205 Super User 2025 Season 2
S-Venkadesh 101 Moderator
