Connect BigQuery to Parabola

Run SQL against BigQuery from Parabola to combine warehouse data with Shopify, NetSuite, and other systems, then automate reports and ops alerts on a schedule.

BigQuery is Google Cloud’s serverless data warehouse, used to query large datasets with standard SQL without managing infrastructure. Connecting BigQuery to Parabola lets analysts and ops teams run queries on a schedule, combine warehouse data with operational systems, and route the results to dashboards, spreadsheets, or downstream tools, without writing custom ETL code.

Pull from BigQuery

The Pull from BigQuery step runs a SQL query against your BigQuery project and brings the results into your flow. Use it to query any dataset and table the connected Google account has access to.

How to authenticate

BigQuery uses Google authentication to authorize the connection.

1. Add a Pull from BigQuery step to your flow.

2. Open the step, click Authorize, and sign in to your Google account, granting Parabola the requested permissions.

3. Each Pull from BigQuery step in a flow can re-use the same authentication. Make sure each step is configured to use your credentials.

Parabola securely stores your credentials and uses them to authenticate every request to BigQuery.

Required IAM permissions in Google

projects.list:

• bigquery.projects.list (or resourcemanager.projects.list)
• Allows listing projects the user can access.

jobs.query:

• bigquery.jobs.create — create and run query jobs.
• bigquery.tables.getData — read data from the tables referenced in the query.
• Users need table-level access for each table they query.

jobs.getQueryResults:

• bigquery.jobs.get — retrieve query job results.

Recommended IAM roles

• roles/bigquery.user — includes bigquery.jobs.create and bigquery.jobs.get.
• roles/bigquery.jobUser — includes bigquery.jobs.create and bigquery.jobs.get (more limited than bigquery.user).
• roles/viewer or roles/browser — includes resourcemanager.projects.list (for listing projects).

Plus dataset/table access:

• roles/bigquery.dataViewer — read access to specific datasets and tables.
• Or grant bigquery.tables.getData on the specific tables and datasets they need to query.

Minimum setup: assign roles/bigquery.user (or roles/bigquery.jobUser) plus roles/viewer (for project listing), and ensure the user has read access to the datasets and tables they query (via roles/bigquery.dataViewer or dataset-level permissions).

Available data

The BigQuery integration runs custom SQL against any dataset and table the connected account has access to. Write standard SQL to filter, aggregate, join, and transform data before it reaches Parabola, including BigQuery-specific features like window functions, array unnesting, and STRUCT access. The chat interface on the canvas can help draft and refine queries.

How to use the available connections:

• List Projects step — find a BigQuery project ID. Increase the Total pages limit value if you need to see more projects.
• Send Query step (Run a SQL query) — submit a query for execution.
• Pull results from BigQuery step (Poll for query results) — retrieve and load the results into the flow as rows.

Common use cases

• Automate recurring reports: Run a SQL roll-up daily or weekly, then export the result to Google Drive, email it to stakeholders, or write it to Smartsheet for a shared review.
• Combine warehouse data with operational systems: Join BigQuery analytics tables with Shopify orders, NetSuite financials, or Fulfil ERP records to spot discrepancies and build unified views.
• Monitor data quality on a schedule: Query BigQuery for row counts, freshness markers, and anomaly checks, then alert in Slack or via email when something looks off.
• Power dashboards with blended data: Pull summarized BigQuery metrics, combine with real-time data from other tools, and write the result to Snowflake, Redshift, or a BI-ready table.
• Trigger downstream workflows: Run an analytical query in BigQuery, pull the results into Parabola, and use them to push updates into Klaviyo, HubSpot, or Slack for activation.
• Feed warehouse data to marketing tools: Pull a BigQuery audience query and push the segment into Mailchimp or Klaviyo so lifecycle campaigns target the same users your warehouse defines.

Tips for using Parabola with BigQuery

• Filter and aggregate in SQL. Push as much logic into the query as possible, smaller payloads run faster, cost less, and keep your Parabola flow simpler. BigQuery is fast; let it do the heavy lifting.
• Use parameterized queries for dates. When a query depends on “yesterday” or “last 7 days,” parameterize the date so you can rerun the same query for any window without editing the SQL.
• Match cadence to use case. Hourly for active ops dashboards, daily for standard reporting, weekly for executive summaries. BigQuery costs scale with bytes scanned, so don’t run expensive queries more often than the use case justifies.
• Handle large result sets with pagination. For queries that return many rows, use Parabola’s pagination so the pull doesn’t time out and full results land in the flow.
• Validate before downstream pushes. Add a Filter or Check step in Parabola that flags unexpected nulls, zero counts, or values out of range before the data lands in Snowflake, Redshift, or a Slack alert.
• Document your SQL in cards. Use Parabola cards to explain what each query returns and why specific filters or joins exist, future-you and the next analyst will thank you.

FAQ

Can Parabola write data back to BigQuery?

The native step pulls from BigQuery. To write data into a BigQuery table, use a Send to an API step against the BigQuery REST API with a service account, or stage the data in Google Drive and load it via a scheduled BigQuery transfer.

How do I query a specific project or dataset?

Fully qualify table names in your SQL using the project.dataset.table form. Make sure the connected Google account has at least roles/bigquery.dataViewer on that dataset, or table-level bigquery.tables.getData access.

Why is my BigQuery query timing out?

Long-running queries time out at the API layer. Pre-aggregate in a scheduled BigQuery transform, narrow the date range, or break the query into smaller windows that you union later in Parabola.

Does Parabola support Standard SQL or Legacy SQL?

Standard SQL. Use Standard SQL syntax in your queries; Legacy SQL is not supported by the integration.

Can I run multiple BigQuery queries in one flow?

Yes. Add multiple Pull from BigQuery steps and they re-use the same authentication. Each step runs its own query and brings its own result set into the flow, ready to join, union, or transform.

---

With BigQuery and Parabola connected, your warehouse stops being a place data goes to sit. Reports, reconciliations, and alerts run on a schedule with output landing exactly where the team works.