Graphed LogoGraphed
Graphed LogoGraphed
How It WorksAgentsIntegrationsPricing

How to Change Privacy Level in Power BI

Cody Schneider••8 min read

When you're trying to combine data from different sources in Power BI, you might run into an error message about privacy levels that stops you in your tracks. This isn't just a technical glitch, it's a critical security feature designed to prevent sensitive data from accidentally being exposed. This guide will walk you through exactly what these privacy levels are, what each one means, and how to change them step-by-step in Power BI Desktop and the Power BI Service.

GraphedGraphed

Your AI Data Analyst to Create Live Dashboards

Connect your data sources and let AI build beautiful, real-time dashboards for you in seconds.

Watch Graphed demo video

What Are Power BI Privacy Levels (and Why Do They Matter)?

First, it's essential to understand that Power BI privacy levels have nothing to do with user permissions or who can access your reports. Instead, they are an isolation setting that controls whether a data source is allowed to share information with other data sources in your model.

Why is this necessary? Power BI is intelligent. To speed up performance, it tries to do something called query folding. This means it combines steps and pushes calculations back to the original data source (like a SQL server) to do the heavy lifting before the data even gets to Power BI. This is much faster than pulling all the raw data into your computer's memory and then processing it.

However, for query folding to work, Power BI might need to send data from one source to another. For example, it might need to send a list of customer IDs from your internal Salesforce report to a public web API to enrich the data.

Here’s the catch: what if your Salesforce data is confidential? You wouldn't want to accidentally send that sensitive information to a public server. Privacy levels prevent this from happening by letting you define the rules of engagement for each data source.

Free PDF Guide

AI for Data Analysis Crash Course

Learn how to get AI to do data analysis for you — the best tools, prompts, and workflows to go from raw data to insights without writing a single line of code.

The Trade-Off: Security vs. Performance

You can think of it as a balance. The more you restrict your data sources (higher privacy), the more secure your data is from potential leaks. However, this restriction can also prevent query folding from occurring, which may slow down your data refresh performance. Understanding how to set these levels correctly allows you to find the right balance for your specific project.

Understanding the Four Main Privacy Levels

Power BI offers four levels of privacy for you to assign to each data source. Selecting the right one ensures your data is handled correctly.

1. Public

This is the least restrictive level. Information from a data source set to 'Public' can be freely shared with any other data source, including those marked as Organizational or Private.

  • When to use it: Use this for data that is genuinely public and contains no sensitive information. Think of data from government websites, public APIs, web pages, or a public census database.
  • Example: A table of US state holidays pulled from a Wikipedia page.

2. Organizational

This level is for internal, business-related data. Data from an Organizational source can be shared with Public sources and other Organizational sources, but it cannot be shared with a Private source. This prevents sensitive company data from being exposed to strictly confidential sources.

  • When to use it: This is best for most internal company data sources that don't contain personally identifiable information (PII) or top-secret financial data.
  • Example: A SharePoint list of company office locations, a sales pipeline report from your CRM (like Salesforce or HubSpot), or product inventory data from an internal database.

3. Private

This is the most restrictive and secure level. Data from a Private source cannot be shared with any other data source. However, other data sources (Public or Organizational) can share their data with the Private source for processing.

  • For example, Power BI can take state names from a Public web source and use them to filter a Private sales table, but it will never send your private sales figures back to the Public source.
  • When to use it: Use this for any data containing sensitive or confidential information, such as PII, employee salary data, patient records, or detailed financial statements.
  • Example: An Excel spreadsheet on a local drive containing employee performance reviews and bonus information.

4. None

The None level isn't truly an isolation level. Setting a source to 'None' tells Power BI to simply ignore it for the purposes of the Data Privacy Firewall check. The data source is effectively invisible to the privacy check rules. While this might sound like an easy fix, it's rarely the right answer and can lead to performance issues or unintended data exposure. It's often better to correctly set the other sources as Public, Organizational, or Private.

GraphedGraphed

Your AI Data Analyst to Create Live Dashboards

Connect your data sources and let AI build beautiful, real-time dashboards for you in seconds.

Watch Graphed demo video

How To Change Data Privacy Levels: A Step-by-Step Guide

Now that you know what the levels mean, let's look at how to change them. The process is slightly different depending on whether you're working in Power BI Desktop or setting up a gateway in the Power BI Service.

In Power BI Desktop

This is where you'll spend most of your time building reports and transforming data. Here’s the most common way to access the privacy settings:

  1. Navigate to the File tab in the ribbon at the top-left corner.
  2. Click on Options and settings, and then select Data source settings.
  3. A dialog box will appear, listing all the data sources currently used in your file. Select the source you wish to modify.
  4. With the source highlighted, click the Edit Permissions button at the bottom of the window.
  5. Another dialog box will appear. Here, at the very bottom, you'll see a dropdown menu labeled Privacy Level.
  6. Click the dropdown menu and select the appropriate level for your data source (Public, Organizational, or Private).
  7. Click OK to save your changes, and then Close the Data source settings window.

Alternatively, if you are working within the Power Query Editor, you can access the same menu by clicking on the Data source settings button located in the Home ribbon.

In the Power BI Service (for On-Premises Gateways)

If you're refreshing data in the Power BI Service from an on-premises source (like a SQL Server inside your company network), the privacy levels must be configured on the gateway connection, not just in the PBIX file.

  1. Log in to your Power BI Service account at app.powerbi.com.
  2. Click the Settings gear icon in the top-right corner of the screen.
  3. From the dropdown, select Manage connections and gateways.
  4. At the top of the page, switch to the Connections tab.
  5. Find the specific data source connection you need to update. Click the three dots (...) menu next to its name and select Settings.
  6. In the settings panel, expand the On-premises data gateway section.
  7. You will see a Privacy setting dropdown. Here you can choose between Private, Organizational, Public, or None for this specific gateway source.
  8. Once you've made your selection, the settings are saved automatically.

Common Errors and Best Practices

Understanding the "how" is good, but knowing the "why" and "when" is even better. Here are some tips to help you manage privacy levels effectively.

Best Practice 1: Start with "Private" as Your Default

When in doubt, always assign the Private level to a new data source. This is the safest setting. If you then encounter a formula firewall error because two sources can't be combined, you can make a conscious decision to lower the privacy of one of them (e.g., to Organizational), knowing the potential implications.

Free PDF Guide

AI for Data Analysis Crash Course

Learn how to get AI to do data analysis for you — the best tools, prompts, and workflows to go from raw data to insights without writing a single line of code.

Best Practice 2: Fix the Right Error Message

If you see an error like "Formula.Firewall: Query references other queries or steps, so it may not directly access a data source. Please rebuild this data combination," this is your cue. The error means Power BI's privacy firewall is stopping one source from sending its data to another. To fix it, go to your Data source settings and ensure that the privacy levels are set appropriately for what you're trying to do.

Best Practice 3: When to Ignore Privacy Levels (Proceed with Caution)

For development purposes on your local machine, and only when using non-sensitive sample data, you can choose to bypass the privacy checks entirely. To do this:

  • Go to File > Options and settings > Options.
  • Under the Global section, click Privacy.
  • Select the option "Always ignore Privacy Level settings."

Warning: Be extremely careful with this setting. While it can make development faster by eliminating firewall errors, it can cause accidental data leaks if you use it with confidential data. It’s generally recommended to leave this setting on "Combine data according to your Privacy Level settings for each source."

Final Thoughts

Mastering Power BI privacy levels is a key step from being a beginner to an intermediate user. It's about taking control of your data's security and your report's performance by thoughtfully defining how your sources can interact, instead of just reacting to error messages.

While managing these configurations is a powerful skill, we created Graphed to simplify the entire analytics workflow. Instead of getting bogged down in connection settings, gateways, and privacy firewalls, you can securely connect all your key data sources - like Google Analytics, Shopify, HubSpot, or Salesforce - in a few clicks. We handle the technical complexity so you can skip straight to asking questions in simple English and get real-time, shareable dashboards built for you automatically.

Related Articles