Google BigQuery Source
Use an authorized BigQuery dataset as a governed source for Vendo queries and models.
Last reviewed July 13, 2026
Direct source — Vendo can query tables and views in an authorized BigQuery dataset. The connection can use Vendo’s managed service account or a service account owned by your organization.
Use Cases
- explore an existing warehouse table;
- bind warehouse fields to Vendo’s data model;
- build transformations, metrics, audiences, or reports from governed data;
- query an existing GA4 export or another customer-managed dataset.
Access Model
The BigQuery app is shared by source and destination workflows. For a read-only source, grant only the permissions needed to inspect metadata and run queries. Select the specific dataset in the connection form and use dataset-level access where practical.
See BigQuery setup for the two supported credential modes and the current form workflow.
Connect the Source
- Open Data → Apps and add Google BigQuery.
- Choose the managed or customer-owned service-account mode.
- Enter the project, grant the current principal access, and load datasets.
- Select the intended dataset and verify read access.
- Save the app, then create or configure the source workflow that uses it.
- Discover the schema and classify the tables or fields needed downstream.
Verify
Run a limited query and compare its row count, columns, timezone, and sample identifiers with BigQuery. Confirm that query billing and audit logs appear in the expected GCP project.
Troubleshooting
| Symptom | Check |
|---|---|
| Dataset missing | Project ID and dataset IAM for the selected principal |
| Query denied | Data Viewer on the dataset and Job User on the execution project |
| Table missing | Dataset selection, view authorization, and schema discovery |
| Unexpected cost | Query project, partition filters, selected columns, and scan volume |