Does Snowflake Connect to Looker?

Yes, Looker connects directly to Snowflake, and it's one of the most powerful and common pairings in the modern data analytics stack. This combination allows you to leverage Snowflake’s scalable cloud data warehouse with Looker’s robust business intelligence and data exploration capabilities. This article will walk you through why this connection is so beneficial, how it works, and a step-by-step guide to get it set up.

Graphed

Still Building Reports Manually?

Watch how growth teams are getting answers in seconds — not days.

Watch the Demo

Why Connect Looker and Snowflake?

Connecting Looker to Snowflake isn't just about making two popular tools work together, it's about creating a business intelligence environment that is fast, scalable, and governed. First, let's briefly define what each tool does.

• Snowflake is a cloud-native data warehouse. It’s designed to store and analyze massive amounts of data with incredible speed. Its unique architecture separates data storage from the computing resources (called "virtual warehouses") used to run queries, which allows for immense flexibility and cost-efficiency.
• Looker (now part of Google Cloud) is a business intelligence and data analytics platform. Its strength lies in its semantic modeling layer, LookML, which creates a reusable, centralized, and version-controlled definition of your business logic. This allows everyone in the company, from data analysts to marketers, to explore data and get reliable answers.

When you combine them, you get the best of both worlds:

1. Centralized Data Meets Self-Service BI: Snowflake acts as your single source of truth, holding all your structured and semi-structured data. Looker sits on top of it, providing a user-friendly interface that lets business users ask questions and build reports without needing to write a single line of SQL.
2. Exceptional Performance: Looker doesn't extract or store your data. Instead, it pushes live queries directly to Snowflake. This means you’re taking full advantage of Snowflake’s powerful processing engine for every dashboard refresh and every report, ensuring your insights are based on the latest available data.
3. Strong Governance and Consistency: Through LookML, you can define your key business metrics (like ‘revenue’, ‘customer lifetime value’, or ‘monthly active users’) once. These definitions are then applied consistently across all reports and dashboards, eliminating the common problem of different teams reporting different numbers for the same metric.

How the Connection Works Under the Hood

Understanding the architecture is straightforward and reveals why this duo is so effective. The entire process relies on Looker's "in-database" approach.

Here’s what happens when you view a dashboard in Looker that is connected to Snowflake:

1. A user clicks to load or interact with a visualization in Looker.
2. Looker's LookML layer translates the user's request into a highly optimized SQL query specifically written in the Snowflake dialect.
3. Looker sends this SQL query to your Snowflake instance using a standard Java Database Connectivity (JDBC) connection. This is a secure and standard way for applications to talk to databases.
4. Snowflake’s virtual warehouse (its compute engine) executes the query against the relevant data stored in its databases. This is the heavy lifting.
5. Snowflake returns the query results back to Looker.
6. Looker takes the raw results and renders them into the beautiful charts, tables, and dashboards your team sees.

The key takeaway is that your data remains securely within Snowflake. Looker is the "brains" that generates the questions (SQL queries), while Snowflake is the "muscle" that crunches the numbers and provides the answers.

Step-by-Step Guide: Connecting Looker to Snowflake

Setting up the connection involves a few preparation steps in Snowflake followed by configuration within the Looker interface. Let's walk through it.

Graphed

Still Building Reports Manually?

Watch how growth teams are getting answers in seconds — not days.

Watch the Demo

What You'll Need First:

• Admin access to your Looker account.
• Account admin privileges (ACCOUNTADMIN role) in Snowflake, or permissions to create users, roles, and warehouses.
• Your Snowflake account identifier (e.g., yourorg-youraccount.snowflakecomputing.com).

Step 1: Create a Dedicated Looker User and Role in Snowflake

For security and management, it’s a non-negotiable best practice to create a dedicated user and role in Snowflake just for Looker. This avoids giving Looker overly broad permissions and allows you to track its specific usage and costs.

Log into your Snowflake account and run the following SQL commands in a worksheet. Be sure to replace the placeholder values with your own.

-- Use a high-privilege role to set up the Looker infrastructure
USE ROLE ACCOUNTADMIN,

-- 1. Create a Role specifically for Looker
CREATE ROLE LOOKER_ROLE,

-- 2. Create a User for Looker. Set a strong password
CREATE USER LOOKER_USER
  PASSWORD = 'a-very-strong-and-secure-password'
  LOGIN_NAME = 'looker'
  DISPLAY_NAME = 'Looker Service User'
  DEFAULT_WAREHOUSE = 'LOOKER_WH' -- We will create this next
  DEFAULT_ROLE = 'LOOKER_ROLE'
  MUST_CHANGE_PASSWORD = FALSE,

-- 3. Grant the newly created role to the new user
GRANT ROLE LOOKER_ROLE TO USER LOOKER_USER,

Step 2: Create a Dedicated Virtual Warehouse for Looker

A virtual warehouse provides the computing power for Looker's queries. Dedicating one to Looker helps isolate its workload from other processes (like data ingestion) and makes it simple to monitor Looker-related costs.

-- 4. Create a Warehouse for Looker
CREATE WAREHOUSE LOOKER_WH
  WAREHOUSE_SIZE = 'XSMALL'
  WAREHOUSE_TYPE = 'STANDARD'
  AUTO_SUSPEND = 60
  AUTO_RESUME = TRUE
  INITIALLY_SUSPENDED = TRUE
  COMMENT = 'Virtual Warehouse for Looker',

-- 5. Grant usage permissions on the warehouse to the Looker role
GRANT USAGE ON WAREHOUSE LOOKER_WH TO ROLE LOOKER_ROLE,

Note: We’re starting with an XSMALL warehouse size, which is often sufficient. You can easily scale this up later if performance becomes a concern. The AUTO_SUSPEND = 60 setting is crucial for cost management, as it will automatically turn off the warehouse after 60 seconds of inactivity.

Graphed

Still Building Reports Manually?

Watch how growth teams are getting answers in seconds — not days.

Watch the Demo

Step 3: Grant Database Permissions to the Looker Role

Now, you need to give your new LOOKER_ROLE permission to actually read the data you want to analyze in Looker. It needs usage rights on the database and schema and select rights on the tables.

Replace YOUR_DATABASE and YOUR_SCHEMA with your actual database and schema names.

-- 6. Grant usage on your database
GRANT USAGE ON DATABASE YOUR_DATABASE TO ROLE LOOKER_ROLE,

-- 7. Grant usage on your schema(s)
GRANT USAGE ON SCHEMA YOUR_DATABASE.YOUR_SCHEMA TO ROLE LOOKER_ROLE,

-- 8. Grant SELECT permissions on existing tables in the schema
GRANT SELECT ON ALL TABLES IN SCHEMA YOUR_DATABASE.YOUR_SCHEMA TO ROLE LOOKER_ROLE,

-- 9. IMPORTANT: Grant SELECT permissions for any new tables created in the future
GRANT SELECT ON FUTURE TABLES IN SCHEMA YOUR_DATABASE.YOUR_SCHEMA TO ROLE LOOKER_ROLE,

Running step 9 is important because it ensures that if new tables are added to your schema, Looker will automatically have access to them without requiring you to manually grant permissions again.

Step 4: Configure the Connection in Looker

With all the preparation done in Snowflake, it's time to head into Looker.

1. Navigate to the Admin section in Looker.
2. In the left sidebar, go to Database -> Connections.
3. Click the Add Connection button at the top right.
4. Fill out the connection form:

• Name: e.g., Snowflake_Production
• Dialect: select "Snowflake"
• Host: your Snowflake account host, e.g., yourorg-youraccount.snowflakecomputing.com
• Port: leave as 443
• Database: your database name (e.g., YOUR_DATABASE)
• Warehouse: LOOKER_WH
• Username: LOOKER_USER
• Password: the password you set
• Schema: your schema name (e.g., YOUR_SCHEMA)
• PDT (Persistent Derived Tables): leave disabled for now
• Database Time Zone: usually UTC
• Query Time Zone: e.g., 'America/New_York'

Graphed

Still Building Reports Manually?

Watch how growth teams are getting answers in seconds — not days.

Watch the Demo

Step 5: Test and Finalize the Connection

At the bottom of the connection settings page in Looker, there is a crucial button: Test These Settings.

Click it. Looker will attempt to connect to Snowflake with the credentials and information you've provided. If all the steps above were completed correctly, you’ll see success messages confirming it can connect and access your data.

If you encounter an error, common issues include:

• Incorrect username, password, or host.
• Firewall rules blocking Looker's IP addresses.
• Inadequate permissions for the LOOKER_ROLE in Snowflake. Double-check your GRANT statements.

Once the tests pass, click Add Connection. That's it! You are now ready to start creating Looker projects and modeling your Snowflake data.

Bonus: Top Best Practices

Just connecting the tools is half the battle. Using them effectively together is what truly drives value.

• Monitor Costs: Use Snowflake’s account usage views or dashboards to monitor query history and warehouse credit consumption for your designated LOOKER_USER. This will give you clear visibility into your Looker-related spend.
• Optimize Your LookML: The quality of your LookML model directly affects the SQL that Looker generates. A well-designed model produces efficient queries, leading to faster dashboards and lower Snowflake costs. Avoid overly complex joins in a single Explore where possible.
• Right-Size Your Warehouse: Don't be afraid to experiment with your Looker warehouse size. If dashboards are slow, try scaling it up (e.g., from XSMALL to SMALL). If it seems over-provisioned, scale it down. A dedicated warehouse makes this an easy variable to adjust without impacting other Snowflake workloads.

Final Thoughts

So, does Snowflake connect to Looker? Absolutely. They integrate seamlessly to form a top-tier modern BI stack that offers unmatched performance, governance, and scalability. The setup process is very methodical: you prepare a secure, isolated environment for Looker within Snowflake, then simply plug those details into Looker's interface to establish the live connection.

We believe that accessing data insights should be even simpler. While tools like Looker and Snowflake are incredibly powerful, they still demand significant technical setup to create dedicated users, manage compute warehouses, and build out complex modeling layers like LookML. At Graphed, we connect directly to your sources - from data warehouses like Snowflake and BigQuery to SaaS platforms like Shopify and Salesforce - and handle all the complexity in the background. Our goal is to let you skip the hours of setup and get straight to asking questions in plain English to build real-time, interactive dashboards instantly.