Fabric and Azure

How to Connect Power BI to Snowflake

Connect Power BI to Snowflake in five steps: pick the connector, enter the warehouse, sign in, choose tables, then Import or DirectQuery.

Austin Levine

Updated

September 23, 2026

In Power BI Desktop, select Get Data, choose the Snowflake connector, enter your server and warehouse name, then sign in and pick your tables. The whole process takes five steps and works the same way whether you plan to import the data or query it live. Before you start, you need an existing Snowflake account with a server hostname and a warehouse already set up; this guide does not create either for you.

Snowflake is a common data warehouse for Power BI reports because it separates storage from compute, so a report's queries do not compete with the jobs loading new data. Below are the five steps, and what to consider when you choose how Power BI connects to it.

Step by step: connect Power BI to Snowflake

Step 1: Get Data

From Power BI Desktop, select Get Data on the Home ribbon, choose Database, then select Snowflake from the list of connectors.

Power BI Desktop Get Data dialog with the Database category open and Snowflake highlighted in the connector list

Get Data > Database > Snowflake in Power BI Desktop.

Step 2: Enter the server and warehouse

In the Snowflake dialog, enter your Snowflake server and the warehouse you want to query.

Power BI Snowflake connection dialog with Server and Warehouse fields and an Advanced options section collapsed

The Snowflake connection dialog. Server and Warehouse are required; Advanced options are not.

Expand Advanced options to set a role name, a command or connection timeout, a specific database, or a native SQL statement to run instead of browsing tables in the Navigator.

Power BI Snowflake advanced options panel showing Role name, Command timeout, Connection timeout, Include relationship columns, Database, and a SQL statement field with a sample query

Advanced options: role, timeouts, a specific database, and a native SQL statement.

Select OK once the fields you need are filled in.

Step 3: Sign in

Power BI Snowflake credentials dialog with Snowflake and Microsoft Account tabs, and Username and Password fields

The credentials dialog. Snowflake (username and password) is one option; Microsoft Account signs in with Microsoft Entra ID.

Power BI Desktop reuses the credentials you enter here for later connections to the same server. To change them, go to File > Options and settings > Data source settings.

Step 4: Select the tables to import

In the Navigator, select the table or tables you want. The panel on the right shows a preview so you can confirm you picked the right one.

Power BI Navigator window showing a Snowflake database tree on the left and a preview of the DIMPRODUCT table on the right

Navigator, with a table selected and previewed on the right.

Select Load to bring the data in as is, or select Transform Data to open Power Query Editor first and remove columns or filter rows before loading.

Step 5: Choose the connection type

Select Import to copy the data into Power BI, or DirectQuery to query it live in Snowflake, then select OK.

Power BI Connection settings dialog with Import and DirectQuery radio buttons

Connection settings: Import copies the data in; DirectQuery leaves it in Snowflake.

Import versus DirectQuery for Snowflake

With Import, Power BI copies the selected data into its own compressed model. Reports run against that copy, which is why Import is usually the faster option to browse and filter. The copy is only as fresh as the last scheduled refresh. Its size is also capped by your license: 1 GB per semantic model on Power BI Pro, 100 GB on Premium Per User, and up to the capacity's own limit on a Fabric capacity workspace with large semantic model storage format turned on.

With DirectQuery, Power BI sends a query to Snowflake every time a visual needs data. There is no copy to refresh, so the report always reflects the current state of the warehouse, at the cost of depending on Snowflake's response time for every interaction. For a full comparison with worked examples, see our guide to Import versus DirectQuery.

Use DirectQuery when the report needs data that changes by the minute, such as an operations dashboard reading directly from a warehouse that a pipeline updates continuously, and the dataset is too large or too fast-moving to reimport on a schedule. Use Import for everything else. It is faster for the reader and does not depend on Snowflake being reachable at report-viewing time.

FAQs

Why do people connect Power BI to Snowflake?

Snowflake stores data and runs warehouse compute separately, so you can scale query capacity up or down without moving the data itself, and multiple teams can run workloads against the same tables without contending for the same compute. Power BI does not do data warehousing. Connecting the two lets a data team keep transformation and storage in Snowflake and use Power BI only for modeling measures and building reports.

Is Snowflake the same kind of tool as Power BI?

No. Snowflake is a cloud data warehouse: it stores tables and runs SQL against them. Power BI is a semantic modeling and reporting tool: it connects to a data source, defines relationships and DAX measures, and renders visuals. They solve different problems and are commonly used together, with Snowflake as the source and Power BI as the reporting layer on top of it.

Does the Snowflake connector support single sign-on?

Yes. Selecting the Microsoft Account tab at sign-in authenticates with Microsoft Entra ID instead of a Snowflake username and password, provided your Snowflake account has Microsoft Entra ID integration configured. Entra ID sign-in only supports DirectQuery, not Import.