Forum Discussion
Local Date Table when connecting to semantic model
Hi julsr ,
Yes, if you have multiple semantic models, you'll need to disable Auto Date Time individually for each one, as the setting only applies to the specific model. Turning it off in one model does not impact the others. Disabling this feature will not affect Incremental Refresh or disrupt your model. Incremental Refresh relies on the date column defined in the refresh policy on the fact table and does not depend on the automatically created LocalDateTable tables.
By turning off Auto Date Time, you take direct control over your date logic. Instead of Power BI generating hidden date tables, you create and manage a visible calendar table, which is then used for filtering, reporting, and time intelligence in both Power BI and Excel. This calendar table does not need to be included in the Incremental Refresh policy, and it does not change data partitioning or refresh behavior.
There is no loss in functionality your visuals and PivotTables will continue to work as before, and often with greater predictability due to the use of a single shared Date table. Additionally, the absence of LocalDateTable entries in Excel keeps the field list more organized. This approach is recommended for production semantic models, particularly when sharing and using Incremental Refresh.
Best Regards,
Tejaswi.
Community Support
Hi julsr ,
Thanks for reaching out to the Microsoft fabric community forum.
The behavior you’re experiencing when connecting Excel to a Power BI semantic model is expected. The LocalDateTable_* objects you see are not manually created tables; they are automatically generated hidden date tables produced by the Power BI engine when the model includes date columns and the Auto Date/Time feature is enabled. Excel displays these internal tables in the field list, making them more visible than in Power BI Desktop.
Currently, Excel does not provide a supported method to selectively hide or exclude only the LocalDateTable_* tables after they have been created.
To prevent these tables from appearing, it is best to disable Auto Date/Time in your semantic model and implement a dedicated Date dimension table. Create a calendar table, designate it as the official Date table, and establish relationships from your fact tables to this calendar table. This approach will prevent Power BI from generating multiple LocalDateTable_* objects, and they will no longer appear in Excel.
In summary, while it is not possible to hide these tables directly in Excel, you can avoid them entirely by turning off Auto Date/Time and using a proper Date table, which is also considered best practice for incremental refresh models.
Best Regards,
Tejaswi.
Community Support
Thank you all for your answers! So, if I have 3 different semantic models connected to my original data source with Incremental Refresh, what I would need to do is disable the Auto date/time feature for each semantic model and create a dedicated datetime table that references the attributes in my table used by the incremental refresh? I didn't understand if I disable it, how could I achieve same results?
Hi julsr ,
Yes, if you have multiple semantic models, you'll need to disable Auto Date Time individually for each one, as the setting only applies to the specific model. Turning it off in one model does not impact the others. Disabling this feature will not affect Incremental Refresh or disrupt your model. Incremental Refresh relies on the date column defined in the refresh policy on the fact table and does not depend on the automatically created LocalDateTable tables.
By turning off Auto Date Time, you take direct control over your date logic. Instead of Power BI generating hidden date tables, you create and manage a visible calendar table, which is then used for filtering, reporting, and time intelligence in both Power BI and Excel. This calendar table does not need to be included in the Incremental Refresh policy, and it does not change data partitioning or refresh behavior.
There is no loss in functionality your visuals and PivotTables will continue to work as before, and often with greater predictability due to the use of a single shared Date table. Additionally, the absence of LocalDateTable entries in Excel keeps the field list more organized. This approach is recommended for production semantic models, particularly when sharing and using Incremental Refresh.
Best Regards,
Tejaswi.
Community Support
