Skip to main content

Sometimes there are changes in the data model (I can add or remove some columns/update transformations, etc). If I have a Power BI connected to this Data Model, the connector will not recognize the change and it will continue update the data based on the previous version, I would like that the connector updates according to the Data Model version.

Hi Matheus,

Thank you for suggesting this feature. I understand your desire to have Power BI automatically adapt to data model changes, but we won't be implementing automatic version updates.

Data model versions act as contracts between our service and clients like Power BI. When you specify a particular version, it guarantees that the schema will remain stable, ensuring your queries work consistently without errors. This stability is important for maintaining reliable data pipelines.

If you create a new version of your data model with different columns or transformations, you'll need to manually update the version in Power BI and adjust your queries accordingly. While this requires extra work, it helps prevent unexpected breaking changes and gives you control over when and how to implement schema updates in your reports.


I understand. Is there a way of knowing what is connected to the Data Model then? So we can send a notification if I have a new data model version?


Hi Matheus,

Unfortunately, this is a common limitation across many platforms. For example, if you’re using REST APIs from companies like Microsoft, Google, or AWS, the client applications consuming those APIs don’t receive automatic notifications about schema changes. They remain tied to the version they initially integrated with, and updates must be communicated through release notes, documentation, or direct outreach to the users.

That being said, I do agree that having better visibility into which consumers rely on which data model versions could be helpful. Feel free to submit a product idea based on this capability. But, keep in mind that even if we do build such a feature, it would only show anonymized usage data that follows GDPR and other privacy rules—we would not reveal the identities of individual consumers. So, in the end, once you’ve deployed a new data model version, you would still need to let your consumers know so they can decide if and when to update their queries.