# How to export SourceLoop data to Snowflake, BigQuery, S3, or Google Cloud Storage

Keep every visit, lead, deal, payment, and attribution credit in your own warehouse or bucket, updated on a schedule, with no duplicates.

Source: https://sourceloop.ai/help/export-data-to-your-warehouse/
Updated: 2026-10-10

---

Your BI team wants SourceLoop's data next to everything else. **Warehouse export** keeps a full copy of your SourceLoop data in your own Snowflake, BigQuery, Amazon S3, or Google Cloud Storage, updated on a schedule.

## Step 1: Add a connection

1. Open **Setup** in the left sidebar, then **Data sync**, and choose the **Destinations** tab.
2. Under **Warehouses**, click **Snowflake**, **BigQuery**, **Amazon S3**, or **Google Cloud Storage**.
3. Click **Add connection** and fill in the details for your warehouse.

## Step 2: Give SourceLoop access

Each drawer explains what to set up on your side:

- **Snowflake**: SourceLoop gives you setup SQL. Run it once in Snowflake as ACCOUNTADMIN. It creates a service user with access to one schema.
- **BigQuery**: create a service account, give it **BigQuery Data Editor** on the dataset and **BigQuery Job User** on the project, and paste its JSON key.
- **Amazon S3**: an access key with write access to the bucket and prefix.
- **Google Cloud Storage**: an HMAC key of a service account with **Storage Object Creator** on the bucket.

Snowflake and BigQuery get the same tables, kept up to date. S3 and Google Cloud Storage get compressed JSON files, one folder per table and day, ready for Redshift, Athena, BigQuery, or Databricks.

## Step 3: Set the schedule and privacy

- **Frequency**: daily, every 6 hours, or hourly
- **Personal data**: send emails, phones, and names as they are, or as SHA-256 hashes

Click **Connect**. The first run starts within about 10 minutes and loads your history.

## Step 4: Keep an eye on it

Each connection shows **Healthy**, **Waiting for the first run**, **Paused**, or **Failing**, with its last run. Use **Send now** to run it straight away, or **Send everything again** to reload all tables from scratch.

## Reports and BI tools

To read SourceLoop numbers straight into a BI tool without a warehouse, use the [Data Studio connector](/help/connect-looker-studio/) on the same tab.

## What's next

- **Bring events in from your CDP:** [Segment, RudderStack, and CDPs](/help/send-events-from-segment-or-rudderstack/)
- **Read SourceLoop in Data Studio:** [Connect Data Studio](/help/connect-looker-studio/)

## Frequently Asked Questions

### Which data is exported?

Visits, events, leads, people, companies, deals, stage changes, revenue, subscriptions, ad spend, and the attribution credit of every touch.

### How often does it run?

Daily, every 6 hours, or hourly, set per connection. Each run sends only what changed since the last one.

### Will rows be duplicated?

No. Rows are merged by their ID, so re-sending a row updates it. Deleted records stay with a deleted flag, so your history is complete.

### Can I leave out personal data?

Yes. Turn off personal data on the connection and emails, phones, and names are sent as SHA-256 hashes instead.

### How far back does the first export go?

The first run loads history, up to 18 months of events and visits and 36 months of everything else, then each run after that sends only changes.