Why Won't My Power BI Refresh?

Nothing is more frustrating than meticulously building a Power BI report only to find that it won't refresh with the latest data. When your dashboards serve stale information, it defeats their entire purpose. This guide walks you through the most common reasons why your Power BI refresh might be failing and shows you exactly how to fix each one.

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

First, Check Your Refresh History for Clues

Before you start tearing apart your data model, let the Power BI service tell you what's wrong. The error message in your refresh history is the number one clue for diagnosing the problem. It will point you in the right direction and save you from guessing.

• In the Power BI service, navigate to your workspace.
• Find the dataset you're having trouble with (look for the orange warning icon).
• Click the three dots (...) next to the dataset name and select Settings.
• Go to the Refresh history tab and look at the status of the recent refresh attempts.

You'll likely see a specific error message, such as "Invalid Connection Credentials" or "On-premises gateway is offline." Use that message to find the relevant section below and get your data flowing again.

Common Problem #1: Data Source Credential Errors

This is easily the most frequent cause of refresh failures. Your credentials are the username and password or authentication token Power BI uses to access your data source. If they expire or change for any reason, Power BI can no longer get in.

Why Credential Errors Happen:

• Password Changes: Your organization's policy might require you to update your password every 90 days. When you change the password for the account used to connect to a source (like a SQL database), you must also update it in Power BI.
• Expired OAuth Tokens: For cloud services like Salesforce, Google Analytics, or SharePoint, Power BI uses a token (OAuth2) to connect. These tokens have a lifespan and will eventually expire, requiring you to re-authenticate.
• Permissions Revoked: If your access to the underlying data source (like a specific database or folder) was changed or revoked, the refresh will fail because the credentials are no longer valid for that resource.

How to Fix It:

1. Navigate to your dataset settings in the Power BI service as described above.
2. Find the Data source credentials section and expand it.
3. You’ll likely see a notice asking you to update your credentials. Click the Edit credentials link.
4. A pop-up window will appear. Carefully re-enter your new password or sign in again to re-authenticate with the cloud service.
5. Once updated, go back to your workspace and trigger a manual refresh to confirm it's working.

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

Common Problem #2: Gateway Issues (For On-Premises Data)

If you're pulling data from a source that is located within your company's private network (like a local SQL Server or a file on a shared network drive), you need an on-premises data gateway. This gateway acts as a secure bridge allowing the cloud-based Power BI service to reach your internal data sources.

Is Your Gateway Online?

A gateway is software that runs on a server computer inside your network. If that computer is turned off, loses internet connectivity, or the gateway service stops running, the bridge is broken, and your refresh will fail.

• In Power BI, click the settings gear icon in the top-right corner and select Manage connections and gateways.
• Find your Gateway Cluster name in the list. You'll see a green status indicator if it's online and connected properly.
• If it’s offline, you'll need to work with your IT team to ensure the machine running the gateway is on, connected to the internet, and that the gateway software is running. It might just need a simple reboot.

Is the Gateway Up-to-Date?

Microsoft regularly releases updates to the gateway software. In some cases, an out-of-date gateway can lose compatibility with the Power BI service. It's good practice to keep the gateway updated to the latest version to prevent these kinds of issues.

Is the Data Source Correctly Added to the Gateway?

For a refresh to work, you must specifically tell Power BI how to use the gateway to find the data source. First, the data source needs to be added within the gateway settings itself. Then, your dataset in the service must be mapped to use that specific gateway connection. If this mapping is missing or incorrect, the refresh will have no path to your data. Double-check your settings under Manage connections and gateways to confirm everything is linked correctly.

Common Problem #3: Changes to Your Data Source

Your Power BI report is sensitive to changes in the structure of its source data. A simple modification to an Excel file or a database table by a well-meaning colleague can bring your scheduled refreshes to a halt.

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

Column Renamed or Deleted

Imagine your report uses a column from an Excel file named "Monthly Sales." Later, a teammate changes that column header to "Sales - Aug. 2024" to be more specific. The next time Power BI tries to refresh, it will look for "Monthly Sales," fail to find it, and the entire refresh will stop with an error like "The column '[Column Name]' of the table was not found."

Data Type Changes

Another common issue is a data type mismatch. Let's say a column contains numerical product IDs but someone accidentally adds a new product with an ID of "T-shirt-01". The column's data type switches from number to text. If your Power Query has a step that relies on that column being a number (like performing a calculation), the refresh will fail when it hits that text value.

How to Fix It:

The best way to diagnose these issues is in Power BI Desktop.

1. Open your .PBIX file.
2. Go to the Home ribbon and click Transform data to open the Power Query Editor.
3. On the Home ribbon within Power Query, click Refresh Preview.
4. Power Query will attempt to re-run your transformation steps and will show you exactly which step is causing the error, often highlighting it in red or yellow.
5. From there, you can adjust the step. You might need to change the column name in your formula to match the source or correct a "Changed Type" step to handle the different kind of data being loaded.

Common Problem #4: Power Query Refresh Timeouts

Sometimes, the refresh starts but never finishes. The Power BI service will automatically cancel any data refresh that takes too long. On a Power BI Pro license, individual datasets are limited to 2 hours per refresh. For Premium capacities, the limit is typically 5 hours.

Why Timeouts Happen:

• Large Data Volume: You might simply be trying to pull too much data at once.
• Inefficient Transformations: Complex or poorly optimized steps in Power Query ("M" code) can significantly slow down the refresh process. Filtering data at the end of your query instead of the beginning is a classic example.
• Slow Data Source: The problem might not be Power BI, but the source database. If it is overloaded or slow to respond, it will cause the Power BI refresh to run slowly and potentially time out.

How to Fix It:

• Filter Early: The single most effective way to speed up a query is to remove unnecessary rows and columns as early as possible. Filter your rows in the very first steps. Use the "Choose Columns" function to get rid of any data you aren't actually using in your report.
• Leverage Query Folding: When connecting to sources like SQL databases, try to ensure your transformations are "folded." This means Power BI translates your Power Query steps (like filtering and grouping) into the native language of the source (like SQL) and sends one efficient query to the database, rather than downloading the entire table and transforming it within Power BI. You can check if a step is folded by right-clicking it, if "View Native Query" is an option, it's folding.
• Incremental Refresh: For very large datasets, set up an incremental refresh. This tells Power BI to only refresh the most recent data (e.g., the last 7 days) instead of re-importing the entire dataset every time, dramatically reducing refresh times.

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

A Sneaky Suspect: Data Privacy Levels

Data privacy levels are a security setting within Power BI that dictates how different data sources are allowed to interact with each other. A mismatch here can prevent data from being combined during a refresh, leading to a sometimes cryptic error.

The settings are:

• Private: Isolates the data source. It cannot be combined with or see any other data source.
• Organizational: Allows the source to be combined with Public sources as well as other Organizational sources.
• Public: Can be freely combined with any other data source.

An error occurs when Power BI needs to send data from a 'Private' source to another source for processing. To prevent leakage of sensitive data, the refresh is blocked. The fix is to ensure your sources are set to appropriate levels, usually setting all internal business sources to 'Organizational'. You can adjust these in Power BI Desktop's options or in the data source credential settings in the service.

Final Thoughts

Troubleshooting a failed Power BI refresh is usually a process of elimination starting with the simplest explanations. Begin by checking the error messages, then your data source credentials and gateway status, as these account for the vast majority of issues. From there, you can dig deeper into changes at the data source or opportunities to optimize your query.

When you're tired of managing gateways, updating credentials across platforms, and writing code, you might find that much of this friction goes away with modern tools. We built Graphed to solve this very problem by connecting directly to your marketing and sales data sources in just a few clicks. It handles the data syncing automatically, pulling real-time information into live dashboards without you ever needing to configure a gateway, re-authenticate a token, or troubleshoot a failing query.