Forum Discussion

PriyaJha's avatar
PriyaJha
Helper I
1 year ago
Solved

Copilot usage and usecases

Hi All,

 

Is there any feature within Fabric/Power BI copilot that would help us with the data quality and data governace?

 

And also can it assit with data modelling aspects like relational /star schema creation? 

 

  • Anonymous's avatar
    Anonymous
    1 year ago

    Hi PriyaJha 

     

    I don't find a relevant use case for using Fabric/Power BI Copilot to help manage data quality and data governance. Based on the current documentation, this may not be supported or ideal.

     

    According to the following document, to work well with Copilot for Power BI, it requires that all relationships between tables are clearly defined and logical. This indicates that Copilot might be not able to help model the data or create appropriate relationships.  

    Update your data model to work well with Copilot for Power BI - Power BI | Microsoft Learn

     

    This article shows some supported question types for asking Copilot in Power BI. Copilot can't currently answer questions that require on-the-fly calculations. It also can't generate new insights like anomaly detection, forecasting, or finding key influencers.

    Ask Copilot for data from your model - Power BI | Microsoft Learn

     

    In Fabric, Copilot doesn't have access to all the data stored in the Fabric items. Copilot can access different data in different Fabric experiences. However, it mainly accesses some metadata related to the table schema and the data involved in the queries that have occurred, and does not access the data inside. For example, read data use section in below articles:

    Privacy, security, and responsible use of Copilot for Data Science - Microsoft Fabric | Microsoft Learn

    Copilot for Data Warehouse (preview) - Microsoft Fabric | Microsoft Learn

     

    In summary, at the moment, it seems that using Copilot to help with data governance or data modeling is not ideal. Perhaps in the future, Copilot will improve in these areas.

     

    Best Regards,
    Jing
    If this post helps, please Accept it as Solution to help other members find it. Appreciate your Kudos!

1 Reply

  • Anonymous's avatar
    Anonymous
    Not applicable

    Hi PriyaJha 

     

    I don't find a relevant use case for using Fabric/Power BI Copilot to help manage data quality and data governance. Based on the current documentation, this may not be supported or ideal.

     

    According to the following document, to work well with Copilot for Power BI, it requires that all relationships between tables are clearly defined and logical. This indicates that Copilot might be not able to help model the data or create appropriate relationships.  

    Update your data model to work well with Copilot for Power BI - Power BI | Microsoft Learn

     

    This article shows some supported question types for asking Copilot in Power BI. Copilot can't currently answer questions that require on-the-fly calculations. It also can't generate new insights like anomaly detection, forecasting, or finding key influencers.

    Ask Copilot for data from your model - Power BI | Microsoft Learn

     

    In Fabric, Copilot doesn't have access to all the data stored in the Fabric items. Copilot can access different data in different Fabric experiences. However, it mainly accesses some metadata related to the table schema and the data involved in the queries that have occurred, and does not access the data inside. For example, read data use section in below articles:

    Privacy, security, and responsible use of Copilot for Data Science - Microsoft Fabric | Microsoft Learn

    Copilot for Data Warehouse (preview) - Microsoft Fabric | Microsoft Learn

     

    In summary, at the moment, it seems that using Copilot to help with data governance or data modeling is not ideal. Perhaps in the future, Copilot will improve in these areas.

     

    Best Regards,
    Jing
    If this post helps, please Accept it as Solution to help other members find it. Appreciate your Kudos!