Why does Power BI Desktop take a lot of time to refresh

[Khushboobarai]

Hi Community, I'm facing this issue of power bi desktop which takes a lot of time like more than hours or 2 hours whenever I have to edit in quesry then close an dapply it takes aorund 2 hours 
My data source is Azure Storage explorer.
It has data from the year 2019 on the daily basis till now.
On the daily basis it does take time in Febric service too, I have only Power BI Pro license.

I do have problem on the Desktop as well as on febric service 

I somehow managed to do for all the back data like 19-25 year I did load and disable its laod so i will take less time in service but still its is 35-40 min and on the desktop you know even if its disabled, it will definitely take the load! idk what to do in this case.

[v-hashadapu]

Hi @Khushboobarai , Hope you're doing fine. Can you confirm if the problem is solved or still persists? Sharing your details will help others in the community.

[HarishKM]

@Khushboobarai Hey,
for desktop - you can use parameter to restrict data to 10-100 row.
refer this - Try limiting rows when creating reporting for big data in Power BI

You can use incremental row restrict full data refresh on Power BI service - Avoid the full refresh with Incremental Refresh in Power BI (Premium)

or you can deploy schema changes without needing to perform the refresh with Power BI Premium and the ALM Toolkit.
refer this - Deploy Power BI dataset schema changes WITHOUT refreshing!

If this solves your problem. Kindly give kudos this solution and accept it as well so other can leaverage the same.

Thanks

Harish M

Connect me on LinkedIn - Harish Kumar Mishra - Fractal | LinkedIn

[Khushboobarai]

I have only power bi pro license, in that case how to do the incremental refresh

[v-hashadapu · Accepted solution]

Hi @Khushboobarai , Yes, you can still use Incremental Refresh with a Pro license. The limitation is more around overall capacity and performance compared to Premium/Fabric, not the feature availability itself. But honestly, in your case the bigger problem is probably the Azure Storage file structure itself. If Power BI is reading thousands of small daily files from 2019 onward, it can become very slow because it has to open and process many files during both Close & Apply and Service refreshes.

Incremental Refresh is still worth implementing because it helps avoid fully reprocessing all historical data every time. Just keep expectations realistic since file-based sources usually do not support query folding like SQL sources do.

I would also strongly recommend keeping only a small recent date filter while developing in Desktop and if possible consolidating old daily files into larger monthly/yearly files or Parquet format. That often helps much more with Azure Storage scenarios like this.

Configure incremental refresh for Power BI semantic models - Power BI | Microsoft Learn

Configure incremental refresh and real-time data for Power BI semantic models - Power BI | Microsoft...

Creating a Dataflow (legacy) - Power BI | Microsoft Learn

Introduction to Dataflows and Self-service Data Prep (legacy) - Power BI | Microsoft Learn

Best practices when working with Power Query - Power Query | Microsoft Learn

[HarishKM]

@Khushboobarai Hey,
for desktop - you can use parameter to restrict data to 10-100 row.
refer this - Try limiting rows when creating reporting for big data in Power BI

You can use incremental row restrict full data refresh on Power BI service - Avoid the full refresh with Incremental Refresh in Power BI (Premium)

or you can deploy schema changes without needing to perform the refresh with Power BI Premium and the ALM Toolkit.
refer this - Deploy Power BI dataset schema changes WITHOUT refreshing!

If this solves your problem. Kindly give kudos this solution and accept it as well so other can leaverage the same.

Thanks

Harish M

Connect me on LinkedIn - Harish Kumar Mishra - Fractal | LinkedIn

[v-hashadapu]

Hi @Khushboobarai , 

hope you are doing great. May we know if your issue is solved or if you are still experiencing difficulties. Please share the details as it will help the community, especially others with similar issues.

[Khushboobarai]

Not resolved yet! 

[v-hashadapu]

Hi @Khushboobarai , There is not much I can say now other than to check the file structure itself in Azure Storage. If you have thousands of small daily CSV/text files from 2019 onward, Power BI can become very slow because it needs to open and process each file separately during refresh. In many cases, this becomes the main bottleneck rather than the PBIX size itself.

If possible, try consolidating older files into larger monthly/yearly files or switch older historical data to Parquet format. Also review any Merge, Group By, Sort or custom column steps in Power Query, because those operations can become extremely expensive on file based sources at this scale.

[v-hashadapu]

Hi @Khushboobarai , Thank you for reaching out to the Microsoft Community Forum.

I think it is expected the View Native Query option being unavailable, since your source is Azure Storage files. Power BI is likely re-reading and transforming a huge amount of historical file data every time you do Close & Apply, which is why both Desktop and Service refreshes are very slow. Also, Disable Load alone does not always prevent Power BI from evaluating a query if it is referenced elsewhere.

With a Pro license, you can still use Incremental Refresh and Power BI Dataflows Gen1. Incremental Refresh will help the most here because only recent data gets refreshed instead of all data from 2019 onward every time. During development, also keep a smaller date filter in Power Query to avoid loading the full history in Desktop.

These Microsoft docs are a good starting point:

Creating a Dataflow (legacy) - Power BI | Microsoft Learn

Understanding Query Evaluation and Query Folding in Power Query - Power Query | Microsoft Learn

Configure incremental refresh and real-time data for Power BI semantic models - Power BI | Microsoft...

Best practices when working with Power Query - Power Query | Microsoft Learn

[cengizhanarslan]

1) Reduce the data at the source. If you are loading raw daily files from Azure Blob/ADLS, apply filters in Power Query as early as possible. Filter by year or date range so that only the data you actually need for reporting is loaded. Go to Home → Transform data, select the date column, and filter out anything older than your reporting window.

2) Enable query folding where possible. If your source supports it (e.g. Azure SQL, ADLS with compute), make sure your filters and transformations fold back to the source rather than being processed locally in Power Query. Check this by right-clicking a step in Power Query → View Native Query. If it is greyed out, folding is not happening.

3) Use incremental refresh. Since your data grows daily, configure incremental refresh so that only recent data (e.g. last 7 or 30 days) is refreshed, and historical partitions remain untouched. This significantly reduces refresh time in the Fabric Service.

[Khushboobarai]

Hi @cengizhanarslan  Native query option is not available in my case as my data source is Microsoft Azure Storage Explorer

[rajendraongole1]

Hi @Khushboobarai  - This usually happens because Power BI is reprocessing a very large amount of historical data every time you click Close & Apply. Since your source is Azure Storage with daily data from 2019 till now, Power BI likely scans and transforms all those files again during refresh.

Recommendation would be to implement Incremental Refresh so only recent data refreshes instead of all historical years every time.

Also consider:

Moving transformations to Fabric Dataflows/Lakehouse

Using Parquet instead of many CSV files

Reducing unnecessary Power Query steps

Filtering recent data while developing in Desktop

Checking if “View Native Query” is still available (query folding)

Even if “Disable Load” is enabled, Power BI may still evaluate those queries if they are referenced somewhere.

In most enterprise scenarios, the better architecture is:

Azure Storage--> Fabric/Dataflow/Lakehouse --> Power BI Semantic Model

instead of processing all raw historical files directly inside Power BI Desktop.

Hope this helps.

[Khushboobarai]

hi @rajendraongole1  thanks for your response
I want  more detail explantion could you suggest me any video or any articale to read and learn about this way
as per my knowledge with Pro liecsense I can use data flow or lake house Idk im not sure as Im fresher here so could you pls help me more about this

[rajendraongole1]

Hi @Khushboobarai - Nop, please find the links and learn docs.

Advanced Incremental refresh in Power BI - Mark Endicott

Incremental Refresh in Power BI

Configure incremental refresh and real-time data for Power BI semantic models - Power BI | Microsoft...

Hope this helps.