Forum Discussion
Unexpected behavior creating Data Product from Fabric assets
We tried creating a data product from PBI assets in Fabric. This resulted in unexpected behavior.
Case:
- We want to create a data product containing: a PBI report, PBI dataset and a PBI dataflow. They show as such in data product search.
- We create a data product, so we select the one PBI report one PBI dataset and one PBI dataflow to be added the data product. So as a user: we select 3 assets in the menu
- Purview determines that every table in the PBI dataset is also an asset (onelake delta table). A dataproduct is created containing 35 assets.
According to support this is expected behavior. However, from the perspective of a user, selecting 3 assets resulting in 35 assets being added to a data product feels like a bug. From a logical perspective, the outcome is inconsistent with the input. From the perspective of the new pricing model where a customer has to pay for every 'curated asset', this behavior doesn't really land well.
We would much rather like to see that data assets are somehow aggregated at a higher level within in the data product view (e.g. in a hierarchy). We have data products containing hundreds of tables. This would make maintaining data products a lot easier, and the overview way more user friendly compared to this 'flat' view of assets.
Hello Rik! Thank you for sharing this detailed feedback. I understand how this comes out inconsistent and challenging. Your suggestion to aggregate data assets at a higher level within the data product view makes a lot of sense to improve the feature and experience here.
I'll make sure to raise this valuable feedback with the relevant team. Your insights are crucial for improving the product, and I appreciate you taking the time to share this.
1 Reply
- milgo
Microsoft
Hello Rik! Thank you for sharing this detailed feedback. I understand how this comes out inconsistent and challenging. Your suggestion to aggregate data assets at a higher level within the data product view makes a lot of sense to improve the feature and experience here.
I'll make sure to raise this valuable feedback with the relevant team. Your insights are crucial for improving the product, and I appreciate you taking the time to share this.