Incremental Refresh with Power BI Pro License

[Khushboobarai]

Hi Community, I wanted to know how can I make my data to load for only last 1 day or 2 day in service? or in desktop 
my data like overall has data since last 6 year from 2020 to current... and data comse on the daily basis.
Data souce is  Microsoft Azure Storage Explorer and one data from the share point...I did connected as a cloud in serive with the respective credential adn working all perfectly from that.
I did schdueled the refresh too which works perfect but all the time it took 40 min to 1 hr almost which is too long to refresh... 
I did made the Range start and range end parameter as suggested by some tutaorial... but that is not visible in service like there is no any option to selcted the parameters.

How can I do that?

[v-sgandrathi]

Hi @Khushboobarai,

Thank you for confirming that the issue remains unresolved.

As previously mentioned, combining multiple folders with append is likely causing incremental refresh to fail, since it disrupts query folding.

If you prefer not to change your main query, you can try this approach:

Create a separate staging query that connects only to the latest data (2026 folders: f1cd, f2cd, f3cd)

Apply RangeStart and RangeEnd filters to a suitable Date/DateTime column in Power Query

Enable incremental refresh on this staging query (refresh the last 1–2 days)

Append this staging query to your main query

This should help reduce refresh time without affecting your model.

Also, please check:

The date column is in Date/DateTime format

Filters are applied in Power Query, not DAX

Use “View Native Query” to verify query folding

Alternative

If your source doesn’t support query folding, you can filter files by last modified date (last 1–2 days) in Power Query as a workaround.

Thank you.

[v-sgandrathi]

Hi @Khushboobarai,

we haven't heard back from you regarding our last response and wanted to check if your issue has been resolved.

Should you have any further questions, feel free to reach out.
Thank you for being a part of the Microsoft Fabric Community Forum!

[Khushboobarai]

not resolved

[v-sgandrathi]

Hi @Khushboobarai,

Thank you @rohit1991 for your prompt response to the query.

We haven’t heard from you on the last response and was just checking back to see if your query was answered.
Otherwise, will respond back with the more details and we will try to help.

Thank you.

[rohit1991]

Hii @Khushboobarai 

Incremental refresh in Power BI Pro works only in the background using RangeStart and RangeEnd, and these parameters are not visible or selectable in Service (by design). To load only the last 1–2 days, you must apply a date filter in Power Query using these parameters and enable incremental refresh (e.g., store 6 years, refresh last 1–2 days). If your refresh still takes 40–60 minutes, it usually means query folding is not happening or a full refresh is being triggered instead of incremental.

[Khushboobarai]

Cna you tell me more about this mehtod??

see my stutcaure is

3 categiorys of data it has backd ata okay

so till 2025 i kept in a seprrate folder for all 3 categiory like

f1bd

f2bd

And f3 bd

and from 2026 it anotehr folder as

f1cd

f2cd

and f3 cd

and my main wuesry has evrything appended all 6 folders okay...

and then like that main file is used for all the emasure and evryrthing...

so i dont want to touch or migrate thta otherwise my work gonna completeda to chanegs evrything....

[rohit1991 · Accepted solution]

Hii @Khushboobarai 

Your current setup (multiple folders + append + main query) is the reason incremental refresh is not working. Incremental refresh only works when a single table has a proper Date column filtered using RangeStart/RangeEnd with query folding  folder-based appends usually break this.

In your case, since you don’t want to change the main model, the best approach is:

• Create a separate staging query that directly reads only recent files (last 1–2 days) using a date filter 
• Apply incremental refresh on that staging table 
• Then append this small “recent data” into your existing model

  This way, you avoid touching your main logic, reduce refresh time, and still get near-incremental behavior.