How Does Power BI Refresh Data?

Keeping your Power BI reports up-to-date isn't just a best practice, it's essential for making timely, accurate decisions. The key to maintaining this accuracy is understanding and mastering the data refresh process. This article breaks down how Power BI data refreshes work, from the different connection types that define your options to step-by-step instructions for scheduling automatic updates.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

Understanding the Basics of Data Refresh

Before diving into the settings, it's important to distinguish between two types of "refreshing" that happen in Power BI. Understanding the difference clarifies what's actually being updated when you configure a refresh.

Dataset Refresh vs. Visual Refresh

When you interact with a Power BI report - clicking a visual, applying a filter, or slicing data - the charts on the page update instantly. This is a visual refresh. The report is simply re-rendering the visuals based on the dataset that has already been loaded into memory. It's fast because it's not fetching new data from the original source.

A dataset refresh, on the other hand, is the process of updating the actual data within the Power BI dataset. Power BI reaches out to the original data source (like an Excel file, a SQL database, or Google Analytics) and pulls in any new or updated information. This updated dataset then becomes the new baseline for all reports and visuals connected to it. This article focuses entirely on the process of setting up and managing a dataset refresh.

It All Starts with Your Data Connection Mode

The way Power BI refreshes data is fundamentally determined by how you connected to your data source in the first place. You choose your connection mode when you first build your report in Power BI Desktop. There are three primary modes, and each has its own rules for data freshness.

Import Mode

This is the most common mode and the one that relies heavily on scheduled refreshes.

When you use Import Mode, Power BI takes a copy of your data from the source and stores it in a highly compressed, in-memory format within the Power BI file (.PBIX). This makes your reports incredibly fast and responsive because all calculations are performed locally by Power BI's powerful engine. However, since the data is a snapshot, it becomes stale as soon as the source data changes. To see new data, you must perform a dataset refresh.

• Best for: Small to medium-sized datasets where performance is a top priority and near-perfect, real-time data is not required.
• Refresh requirement: Requires a manual or scheduled refresh to update the copied data.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

DirectQuery Mode

With DirectQuery, Power BI doesn't import or copy any data. Instead, it maintains a direct connection to the underlying data source. Whenever you interact with a report, Power BI sends live queries to the source database to fetch the necessary information. This means your report always displays the most current data.

While this sounds ideal, it comes with trade-offs. Report performance is entirely dependent on the speed and efficiency of the source database. Additionally, there are more limitations on the types of transformations you can perform in Power Query compared to Import Mode.

• Best for: Very large datasets (too big to import) or situations where real-time data is absolutely critical, such as a monitoring dashboard.
• Refresh requirement: None. The data is always live. A visual refresh sends new queries to the source.

Live Connection

Live Connection is similar to DirectQuery but is specifically used to connect to pre-built analytical models, most commonly SQL Server Analysis Services (SSAS), Azure Analysis Services (AAS), or another Power BI dataset. Like DirectQuery, no data is imported, Power BI is simply a visualization layer on top of a governed, centralized model. The responsibility for data refreshing falls on the source model, not the Power BI report itself.

• Best for: Enterprise scenarios where a certified, single-source-of-truth data model already exists.
• Refresh requirement: Handled at the source model level (e.g., within Analysis Services).

Configuring Data Refreshes in Power BI Service

Since Import Mode is the most common setup, let's walk through how to configure its refreshes in the Power BI Service (app.powerbi.com), which is where your dashboards and reports live online.

Manual Refresh ("Refresh Now")

The simplest way to update your data is with a manual refresh. This is an on-demand option that immediately tells Power BI to go out and grab the latest data from your source.

When to use it:

• Right after you've made a significant update to the source data and want to see it reflected immediately.
• As a quick test to ensure your data source credentials and connections are working correctly before setting up a schedule.

To start a manual refresh, simply navigate to your workspace, find your dataset in the list, and click the circular "Refresh now" icon.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

Scheduled Refresh

For most business reports, you'll want your data to update automatically. After you publish your report from Power BI Desktop to the Power BI Service, you can set up a refresh schedule.

Follow these steps to schedule a refresh:

1. Go to Dataset Settings

In your Power BI workspace, find the dataset you want to schedule. Instead of clicking the refresh icon, click the three-dot menu (...) and select Settings.

2. Configure Data Source Credentials

The first thing Power BI needs is permission to access your data. In the settings, expand the Data source credentials section and click Edit credentials. You'll be prompted to sign in to your data source (e.g., your Microsoft 365 account for an Excel file in SharePoint, or a database user/password for SQL Server). Power BI securely stores these credentials to use during scheduled refreshes.

Note: If your data source is on a local file system or a private network server (and not in the cloud), you will see a message about needing a Gateway connection. This is a critical component for on-premises data.

3. Set Up Your Data Gateway (If Needed)

If your sources - like a local SQL Server or an Excel file on a company-shared drive - are not accessible via the public internet, you need to install and configure an on-premises data gateway. The gateway acts as a secure bridge, allowing the cloud-based Power BI Service to communicate with your private, on-premises data without exposing it to the web.

• Install the standard gateway on a computer within your network that is always on.
• Connect your on-premises data source to the gateway in the dataset settings.

You do not need a gateway if all your data sources are cloud-based, such as SharePoint Online, Azure SQL Database, or Google Sheets.

4. Set the Schedule

Once your credentials are set and your gateway is running (if necessary), expand the Scheduled refresh section. Here, you can:

• Toggle the schedule on: Keep data up to date according to the schedule below.
• Set Refresh frequency: Typically "Daily," though "Weekly" is also an option.
• Select your Time zone: Ensure your refreshes run at the correct local time.
• Add times: Click Add another time to specify when the refresh should occur.

The number of refreshes you can schedule depends on the capacity your workspace is in:

• Shared Capacity (Pro license): Up to 8 refreshes per day.
• Premium Capacity (Premium license): Up to 48 refreshes per day.

5. Set Up Failure Notifications

Finally, it's wise to enable refresh failure notifications. Check the box to "Send refresh failure notifications to me." This way, if a refresh fails for any reason (e.g., the gateway is offline, or credentials have expired), you'll receive an email and can resolve the issue quickly.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

Advanced Refresh Scenarios for Power BI Premium

Organizations with a Power BI Premium license gain access to more advanced refresh capabilities that are vital for handling large and fast-moving datasets.

Incremental Refresh

By default, a standard refresh re-imports your entire dataset every time. For a massive dataset with years of historical data, this is inefficient. Incremental refresh provides a smarter solution. It partitions the data (usually by date) and only refreshes the most recent partitions, leaving the historical data untouched. For example, you can configure it to refresh only the last seven days of data while archiving old data once it's complete. This dramatically reduces refresh times and resource consumption. You configure incremental refresh policies within Power BI Desktop before publishing.

Automatic Page Refresh

This feature applies primarily to DirectQuery sources. It allows you to set a specific report page to automatically refresh its visuals at a set interval - from every second to every few days. This turns your regular report page into a real-time monitoring dashboard, ensuring visuals are updated without any user interaction. It's perfect for manufacturing floors, social media monitoring, or any operations center display.

Troubleshooting Common Refresh Failures

Even with a perfectly configured schedule, refreshes can fail. Here are a few common culprits and how to fix them:

• Gateway is Offline: The machine hosting your gateway might be turned off or has lost its internet connection. The solution is simple: ensure the machine is on and connected. This is why gateways are usually installed on dedicated servers.
• Expired Credentials: A source system password changed, but you forgot to update it in Power BI Service. Go back to the dataset settings, edit the data source credentials, and re-authenticate.
• Refresh Timeout: Refreshes on a Pro/shared capacity have a 2-hour-time limit. If your dataset is large and complex, the refresh might time out. You can optimize your data model in Power BI Desktop or, if needed, move the workspace to a Premium capacity, which supports a refresh duration of up to 5 hours.
• Source Data Change: An underlying column name in the source Excel file was changed from "Revenue" to "Sales." Because your transformations in Power Query are looking for the original name, the refresh will fail. The fix here is to open the PBIX file in Power BI Desktop, try refreshing there to pinpoint the error in the Power Query Editor, correct the broken step, and republish the report.

Final Thoughts

Successfully configuring your data refreshes is the final step in moving from a static report to a dynamic, reliable business intelligence tool. Whether you're running a simple manual refresh or a complex, incrementally scheduled one, understanding how Power BI communicates with your sources is key to delivering trustworthy data for decision-making.

While Power BI is a powerful tool, managing refresh schedules, gateways, and credentials across multiple platforms can quickly become complex. At Graphed, we simplify this entirely. By connecting your sources - like Google Analytics, Shopify, Facebook Ads, and dozens of others - directly, we create real-time dashboards that update automatically. There are no schedules to configure or gateways to install, just live data ready for you to analyze using simple, natural language.