Power BI scheduled refresh: keep every report current
Power BI scheduled refresh reloads an imported semantic model at set times, so readers see current numbers without anyone opening a file. The feature is simple to switch on, but it depends on credentials, gateways and a well-built model. When one of those pieces breaks, reports go stale quietly. This guide explains how refresh works, why it fails and how to make it dependable.
- How refresh works
- Common failure causes
- Setup in four steps
Last updated
How does Power BI scheduled refresh work?
An imported model holds a copy of the source data. That copy stays frozen until something reloads it. A schedule in the service tells the platform when to do so.
At each chosen time, the service queues a refresh. It connects to every source with stored credentials and runs the Power Query steps again. Then it loads the new rows and recalculates the model. Reports and dashboard tiles update afterwards.
So the schedule lives with the semantic model, not with the report. The number of refreshes allowed per day depends on the license or capacity, and Microsoft documents the current limits.
What does a Power BI refresh schedule depend on?
Six things must line up before a schedule runs cleanly.
Stored credentials
The service signs in to each source on your behalf. When a password expires, the refresh stops.
A data gateway
Sources on a private network need a gateway, and that machine must stay online at refresh time.
Storage mode
Only imported tables need a reload. DirectQuery tables read the source live, so they follow other rules.
Time zone and slots
Each schedule uses a time zone and specific times. Pick slots after the source systems finish their own loads.
Model efficiency
A lean model refreshes quickly. However, wide tables and heavy query steps can run past the time limit.
A responsible owner
Failure emails go to the model owner. If that person has left, nobody hears about the problem.
Why does a Power BI refresh fail?
Most failures have plain causes. A password changed, a gateway computer went to sleep or a source column got renamed. Because the error message names the step, the fix is usually quick once someone looks.
Timeouts are the harder case. As data grows, a full reload takes longer each month. Eventually it crosses the limit, and the refresh stops partway. Incremental refresh solves this, since it reloads only recent periods.
There are also quiet pauses. The service can disable a schedule after repeated failures. It may also pause one when nobody has opened the content for a long time. Therefore a report can look healthy while showing old numbers.
Good habits for Power BI scheduled refresh
These habits prevent most stale-data surprises.
- Show the last refresh time on every report page
- Send failure notices to a shared mailbox, not one person
- Run gateways on servers that stay on, never on a laptop
- Stagger schedules so large models do not all start together
- Use incremental refresh for big fact tables
- Review refresh history each week for slow or failing models
How to set up Power BI scheduled refresh in four steps
The setup takes minutes once the groundwork is in place.
Publish the model
First publish from the desktop application to a workspace. Then open the settings of the semantic model.
Connect the sources
Enter credentials for cloud sources, and map local sources to a gateway connection.
Choose times
After that, turn the schedule on, set the time zone and add the refresh slots your license allows.
Test and watch
Finally, run a manual refresh and check the history. When it succeeds, confirm who receives failure notices.
Questions about Power BI refresh and current data
How often can a model refresh each day?
It depends on the license or capacity behind the workspace. Shared workspaces allow a modest number of daily slots, while capacity-backed workspaces allow many more. Microsoft publishes the current limits. If you need fresher data than the schedule permits, consider DirectQuery for the tables that change quickly.
Does Power BI scheduled refresh need a gateway?
Only for sources the cloud service cannot reach directly. Databases and files on a private network need an on-premises data gateway. Cloud sources usually connect without one. When a model combines both kinds, the gateway configuration must include every source, or the refresh will fail.
What is incremental refresh?
Incremental refresh splits a large table into time periods and reloads only the recent ones. Older periods stay untouched. As a result, a refresh that took an hour may finish in minutes. It needs date filters in Power Query and a policy on the table, and it works on standard per-user licenses too.
Why did our Power BI refresh schedule stop?
Two causes are common. After several failures in a row, the service disables the schedule until someone fixes the error and turns it back on. It can also pause a schedule when the content has gone unused for a long time. The refresh history shows which case applies.
Can you fix unreliable refreshes for us?
Yes. We review refresh history, gateway health and the model itself, and then remove the causes one by one. A free 30-minute assessment comes first. Power BI Agency is a division of Sataware Technologies and is independent of Microsoft. We work remotely inside your own tenant.
Let's build your first dashboard together.
Tell us what you report on today. A consultant replies within 2 hours with next steps.
Trusted partners
Microsoft Solutions Partner · AWS Partner
Cloud & AI Platforms
Azure SQL, Microsoft Fabric and the data pipelines behind your reports.
AI Business Solutions
Power BI reporting on your Dynamics 365, Business Central and Power Platform data.
Security
Row-level security and Entra ID sign-in on every report and workspace.
Amazon Web Services
Power BI on your AWS data, with connectors for Redshift, Athena and S3.