Power BI Semantic model refresh schedule is not updating the data

[virimian123]

We have observed that the Power BI semantic model takes approximately 2 to 3 minutes to refresh the data during scheduled or regular refresh operations. However, we have also noticed that sometimes the same data refreshes in just about 10 seconds. Additionally, despite successful data refreshes, some updates are not immediately reflected in the associated reports.

Could you please help us understand the specific factors or conditions that could cause such discrepancies in refresh duration? Furthermore, we would appreciate any guidance on best practices or steps to ensure that updates are always accurately and promptly reflected in the reports.

[V-yubandi-msft]

Hi @virimian123 ,

May I ask if you have resolved this issue? If so, please mark the helpful reply and accept it as the solution. This will be helpful for other community members who have similar problems to solve it faster.

Thank you.

[rohit1991]

Hi @virimian123 ,
We’ve noticed that the Power BI semantic model behaves inconsistently when it comes to data refresh times—sometimes completing in just 10 seconds, while other times it takes 2 to 3 minutes. Even when the refresh completes successfully, the updated data doesn’t always appear immediately in the associated reports. This discrepancy can be caused by several factors. First, Power BI uses a combination of caching and query optimization, so shorter refresh times may occur when minimal data changes are detected or when the cache is still valid. On the other hand, longer refreshes might involve larger data volumes, more complex transformations, or timeouts from external sources. 

Additionally, report visuals may show outdated data if the report page is cached or if the semantic model hasn't been reprocessed in memory yet. To ensure timely and accurate updates, consider using scheduled refreshes in combination with manual or API-triggered refreshes after data uploads, enabling query caching to be cleared when necessary, and monitoring the refresh history for errors or delays. Also, using tools like Power BI Service metrics and performance analyzer can help pinpoint where bottlenecks are occurring.

Passionate about leveraging data analytics to drive strategic decision-making and foster business growth.

Connect with me on LinkedIn: Rohit Kumar.

[V-yubandi-msft]

Hi @virimian123 ,

Thanks for posting your query in the Microsoft Fabric Community. We have examined the variability in refresh times and delays in report updates. The duration of Power BI semantic model refreshes and the delays in report updates are often affected by several factors.

1. Large datasets or complex transformations can increase refresh time. If some partitions have more data changes than others, refresh times will vary.
2. Refresh performance may vary based on the available capacity in your Power BI Premium or Fabric workspace. If multiple workloads are running simultaneously, your refresh might be throttled or queued.
3. Even after a successful refresh, Power BI Service caches (both report visual cache and tile cache in dashboards) might not immediately show the new data. Report pages often use cached visuals unless manually refreshed or interacted with.
4. If using on premises sources, a slow or overloaded on-premises data gateway can introduce variability in refresh duration.
5. APIs, cloud databases, or shared Excel/CSV files may sometimes respond slower, depending on network or load conditions.

Improve This :
1. If you’re working with large datasets, setting up incremental refresh can really help by only updating the data that has changed, not the entire dataset.

Incremental refresh for semantic models in Power BI - Power BI | Microsoft Learn

2.Try to remove unnecessary steps in Power Query, reduce the number of columns or rows, and use query folding to push calculations back to the source system whenever possible.

3. Eliminate unused columns and measures, and be mindful of columns with too many unique values (like exact timestamps) as they slow things down.

4.Review and streamline DAX calculations, and use the Premium Capacity Metrics App to monitor how your refreshes are performing over time.

If my response solved your query, please mark it as the Accepted solution to help others find it easily!

And if my answer was helpful, I'd really appreciate a 'Kudos'.