Fabric link from Dataverse - decimals lost on column value

[Gpop13]

Hi,

Has anyone faced any issue with Fabric link for Dataverse, where some column in dataverse holds a decimal value but in the replicated table in lakehouse, it is a whole number?

Example - Dataverse (dynamics)

But when the table is checked on the fabric lakehouse :

Is there any setting for this to flow correctly and not losing the decimals?

[v-sgandrathi]

Hi @Gpop13,

Thank you for your question. I understand you are facing an issue where a Dataverse column containing decimal values appears as a whole number in the Lakehouse table after using the "Link to Fabric" feature.
This is a known behavior and is related to how data types are interpreted during the replication process between Dataverse and Microsoft Fabric.

Sometimes, decimal or floating-point columns in Dataverse are mapped to integer types in the Lakehouse, leading to a loss of decimal precision. This typically occurs due to metadata translation during the linking process. Currently, the "Link to Fabric" feature does not offer settings to control or enforce data type mapping.

As a possible solution, please verify that the original column in Dataverse is explicitly set to support decimal values. If the issue persists, you may want to explore alternative ingestion methods, such as Dataflow Gen2, which provide greater control over data types and transformations during data transfer.

Here are the documents from microsoft website you can have a look:
Link your Dataverse environment to Microsoft Fabric and unlock deep insights - Power Apps | Microsof...
Troubleshooting common issues with link to Fabric with Microsoft Dataverse - Power Apps | Microsoft ...
Troubleshoot the Dynamics 365, Dataverse (Common Data Service), and Dynamics CRM connectors - Micros...
Configure your environment and link to Microsoft Fabric - Power Apps | Microsoft Learn

Thank you.

[Gpop13]

Thank you @v-sgandrathi  for your response. Can you explain what do you mean by this?
"As a possible solution, please verify that the original column in Dataverse is explicitly set to support decimal values."
I do see the value holding a decimal precesion on Dynamics front end. Where else can I check this?

[v-sgandrathi]

Hi @Gpop13,

Thank you for your follow-up.

To clarify, when we referenced verifying the original column in Dataverse for decimal support, we meant checking the actual data type at the schema level,rather than just depending on its appearance in Dynamics. Even if decimals are visible in Dynamics, the field may be set as a Whole Number or Floating Point Number, which can result in inconsistent precision when connecting to Fabric.

To confirm the data type, please navigate to Power Apps > Tables > select your table > Columns, and review the specific column to ensure it is defined as Decimal Number. This is recommended for maintaining precision during replication.

Thank you.

[Gpop13]

Thanks @v-sgandrathi . I checked and the datatype that shows on Power Apps table for that column is currency. So does it not support decimals in Fabric link?

[v-sgandrathi · Accepted solution]

Hi @Gpop13,

Thank you for confirming that the column is of type Currency in Power Apps.

While the Currency data type in Dataverse does support decimal precision, we’ve observed that during the "Link to Fabric" process, these fields may sometimes be mapped in a way that causes the decimal portion to be lost in the Lakehouse table. This is because Currency fields in Dataverse are stored with additional metadata (such as currency codes), and in some cases, the value field may be interpreted as a whole number during replication.

At this time, this appears to be a limitation in how Currency fields are handled during automatic schema generation when linking Dataverse to Fabric. To retain decimal precision, we recommend considering Dataflow Gen2 as an alternative ingestion method, where you have more control over data type mapping and transformation.

Regards,
Sahasra.