> For the complete documentation index, see [llms.txt](https://docs.quper.co/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.quper.co/integrations/data-warehouses-and-platforms/snowflake.md).

# Connect to Snowflake

Quper connects to Snowflake using a dedicated read-only user and role. It tracks warehouse credit consumption, query costs, storage spend, and idle compute — giving you full visibility into where your Snowflake spend is going and which teams or workloads are driving it.

## What Quper tracks from Snowflake

* Warehouse credit consumption by warehouse, user, and query type
* Storage costs including data storage, failsafe, and stage storage
* Query-level cost attribution — spend per query, role, user, and database
* Idle warehouse time and auto-suspend performance
* Credit usage trends and anomalies in real time

## Before you begin

* You must have **ACCOUNTADMIN** access to create the Quper user and role
* Quper supports connecting **multiple Snowflake accounts** — repeat this process for each account

## Connection steps

{% stepper %}
{% step %}
**Go to Settings → Integrations**

In your Quper workspace, navigate to **Settings → Integrations** and click **Add Integration → Snowflake**.
{% endstep %}

{% step %}
**Run the setup SQL in Snowflake**

Quper provides a ready-to-run SQL script that creates a dedicated user and role with read-only access. Run the following in a Snowflake worksheet as ACCOUNTADMIN:

```sql
-- Create Quper role
CREATE ROLE IF NOT EXISTS QUPER_READER;

-- Grant access to billing and usage views
GRANT IMPORTED PRIVILEGES ON DATABASE SNOWFLAKE TO ROLE QUPER_READER;

-- Create Quper user
CREATE USER IF NOT EXISTS QUPER_SERVICE_USER
  PASSWORD = '<generate-a-strong-password>'
  DEFAULT_ROLE = QUPER_READER
  DEFAULT_WAREHOUSE = '<your-warehouse>'
  MUST_CHANGE_PASSWORD = FALSE;

-- Assign role to user
GRANT ROLE QUPER_READER TO USER QUPER_SERVICE_USER;
```

Replace `<generate-a-strong-password>` with a strong password and `<your-warehouse>` with any warehouse that can run lightweight queries.
{% endstep %}

{% step %}
**Enter credentials in Quper**

Back in Quper, enter:

* **Account Identifier** — your Snowflake account locator (e.g. `xy12345.us-east-1`)
* **Username** — `QUPER_SERVICE_USER`
* **Password** — the password you set in the SQL above
* **Warehouse** — the warehouse name you specified

Click **Connect**. Quper verifies the connection and begins ingesting usage and billing data.
{% endstep %}
{% endstepper %}

## What queries Quper runs

Quper queries the `SNOWFLAKE.ACCOUNT_USAGE` schema to pull billing and usage data. These queries are read-only and run once per hour on a small warehouse. Quper's query load on your Snowflake account is minimal — typically less than 1 credit per day.

## Troubleshooting

**"Insufficient privileges" error** Confirm that `IMPORTED PRIVILEGES` was granted on the `SNOWFLAKE` database to the `QUPER_READER` role. This is the specific grant that enables access to `ACCOUNT_USAGE` views.

**Data has a 45-minute delay** This is expected. Snowflake's `ACCOUNT_USAGE` views have a built-in latency of up to 45 minutes. Quper reflects this accurately in its ingestion timestamps.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.quper.co/integrations/data-warehouses-and-platforms/snowflake.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.