How to Change Data Source Credentials in Power BI Service

Nothing stops a productive morning faster than opening Power BI and seeing a "refresh failed" alert on your most important dashboard. Often, this isn't a complex data modeling issue but something much simpler: an outdated password. This guide walks you through the straightforward process of changing your data source credentials in the Power BI Service so your data flows smoothly and your reports stay current.

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

Why Do Data Source Credentials Need to Be Updated?

You set up your data connections once and everything works perfectly... until it doesn't. Your data source credentials aren't static, they change for many valid reasons, most of which are tied to security and account administration. Understanding why this happens can help you anticipate and quickly resolve refresh failures.

Here are a few of the most common reasons you'll need to update credentials:

• Standard Password Expiration Policies: Many organizations enforce regular password updates for security compliance. When you change your password for a database or service that Power BI connects to, you also have to update that password inside the Power BI Service for your scheduled refreshes to continue working.
• An Employee Leaves the Company: If a report's data sources are connected using a specific person's credentials, you'll have a problem when that person leaves and their account is deactivated. This is a powerful argument for using dedicated service accounts for data connections instead of individual user accounts, as they are not tied to a single employee's tenure.
• Switching to a Service Account: Related to the point above, you might realize connecting with your personal login was a quick solution but not a sustainable one. As a best practice, businesses often create a dedicated, non-personal user account (a "service account") with the necessary permissions. You'd then need to switch the data source credentials in Power BI from your account to the new service account.
• Changes in Permissions: Maybe a database administrator tightened security, and the account Power BI was using no longer has "read" access to a specific table. While this isn't strictly a credentials update, the troubleshooting process often starts here, and updating to an account with the correct permissions is the solution.
• Expired OAuth Tokens: When you connect to cloud services like Salesforce, Google Analytics, or SharePoint Online, Power BI often uses an authentication method called OAuth2. Instead of storing a password, Power BI gets a secure access token. For security reasons, these tokens expire periodically (e.g., every 90 days). When they do, you simply need to sign in again to "re-authorize" the connection and get a new token.

Staying on top of these credentials is a fundamental part of reliable reporting. Otherwise, your seemingly automated dashboards can quickly become filled with stale, inaccurate 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

Finding Your Data Source Settings in Power BI Service

"Refresh needs updated credentials" is a helpful error message, but it does not provide a button labeled "click here to fix." Before updating your login details, you first need to find where they're stored. The key is knowing whether your data is coming from a Dataset or a Dataflow.

• A Dataset is the data model behind your reports and visuals. It connects directly to the source.
• A Dataflow is a shared collection of data prepared in Power Query Online that can be used by multiple datasets.

If your report uses a dataflow, you must update the credentials on the dataflow first. If it's a direct connection, you'll update the dataset. Here's how to get to the settings for either.

For Datasets:

1. Log in: Open your web browser and navigate to app.powerbi.com.
2. Navigate to your Workspace: On the left-hand navigation pane, select Workspaces and choose the workspace containing the dataset you need to fix.
3. Locate the Dataset: Within the workspace, you'll see a list of all your assets: reports, dashboards, datasets, and dataflows. Find the dataset that's failing to refresh. Datasets are labeled with the type "Dataset".
4. Open Settings: Hover your mouse over the dataset, click the three-dot menu (...)", for more options, and then select Settings from the list.

This will take you directly to the settings page for that specific dataset.

For Dataflows:

The process is nearly identical.

1. Follow steps 1 and 2 above to get to the correct workspace.
2. In the list of assets, find the relevant dataflow. It will be labeled with the type "Dataflow."
3. Hover over the dataflow in question, click the three-dot menu (...)", and select Settings.

Once you're on the Settings page for either your dataset or dataflow, you're in the right place to make the fix.

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

Step-by-Step Guide to Changing Credentials

Now that you've navigated to the Settings page, finding and updating the credentials is straightforward.

On the Settings page, look for the section called Data source credentials. Here you'll see a list of every unique data source that powers your dataset or dataflow. Maybe you connect to a SQL Server database and pull data from a separate SharePoint list, you would see both listed here.

Find the data source whose credentials need updating and click the Edit credentials link next to it.

A dialog box will pop up. Here's where things can vary slightly depending on the type of data source you are using.

Understanding Authentication Methods

Power BI offers several ways to authenticate, and the one you need depends on your data source.

• Basic: The simplest method. This is typically used for sources like SQL Server or other databases where you manually enter a hardcoded Username and Password.
• OAuth2 / Organizational Account: This is the standard for most cloud-based SaaS applications (SharePoint Online, Salesforce, Dynamics 365, etc.). Instead of you providing a password directly to Power BI, it redirects you to the service's own sign-in page. You sign in with your work account, and the service gives Power BI secure, temporary access. This is why you often see a "Sign in" button instead of a password field.
• Web API: Used for custom web data sources, this method often requires a security Key to be entered. You get the key from the web service you're trying to access.
• Windows: For on-premises data sources accessed through a gateway, this allows you to pass through Windows credentials, which are typically in the format of DOMAIN\username.
• Anonymous: As the name implies, no credentials are required. This is for publicly accessible data from the web. You probably won't be updating this type of connection.

To Update Your Credentials:

1. Select the correct Authentication method: Ensure this matches what is required by the data source.
2. Enter details:
3. Set the Privacy level: This setting tells Power BI how to handle data security between different sources.
4. After entering the updated information and settings, click "Save" or "Connect." Power BI will test the connection immediately. A success message means you have successfully updated the credentials! You can now trigger a manual refresh or wait for the next scheduled one to succeed. It can take up to a few minutes for this sync to work, so don't stress if your first attempt fails.

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

Troubleshooting Steps

"Updated successfully" is the ideal outcome. But what happens when you receive an error? Here are a few quick checks:

• Invalid credentials: The most obvious but often overlooked problem. Double-check every character in your username and password. Make sure there are no extra spaces, and confirm that you've copied them correctly.
• Gateway is Off or Misconfigured: When connecting to an on-premises data source like a SQL Server in your office, Power BI uses the On-Premises Data Gateway to access the data. If that gateway service isn't running on the server or is not configured correctly, no credential update will succeed. Check that your gateway is installed and configured properly.
• Permissions Issue: The account you're using might have valid logins, but it may not have permission to view new data in the source. Check with the database administrator to ensure the account you're using has assigned correct permissions.
• OAuth Token Expired: This is a common error with cloud services. It simply means that the temporary security token Power BI has for your cloud service has expired. Just go through the sign-in process again to refresh the token.

By systematically checking the credentials and evaluating permissions in the data source itself, you can resolve most common data refresh failures.

Final Thoughts

Updating credentials is a quick but vital maintenance task for any Power BI-driven organization. When you understand the need to update these credentials and know how to navigate towards dataset or dataflow settings, you can prevent refresh failures and make your reports useful and trustworthy.

This important process of managing settings and credentials is part of the reason we built Graphed. We wanted to make connecting data sources, like Google Analytics, and household source setups a few clicks easier. Our tool automates troubleshooting and minimizes time spent on setting new datasets from old data. With Graph, you're not hunting down settings changes, and you're streamlining your data connection experience.