Stack Dyno reads from the native Google Cloud Billing export so you never ship CSVs again. Use this checklist to wire it up.

1. Enable the export

1. Open the Google Cloud console → Billing → Billing export.
2. Choose Detailed usage cost and select BigQuery as the destination.
3. Create (or reuse) a dataset such as stackdyno_billing.

# Example dataset creation
bq --location=US mk --dataset my-project:stackdyno_billing

2. Grant access

• Give your Stack Dyno service account BigQuery Data Viewer on the dataset.
• Grant BigQuery Job User on the project to allow ad-hoc queries.
• Store the JSON key securely; the marketing site demo form never uploads it.

3. Track freshness

Verify that new partitions arrive at least daily:

SELECT MAX(_PARTITIONTIME) AS latest_partition
FROM `my-project.stackdyno_billing.gcp_billing_export_*`;

Add this query to a scheduled monitor or Stack Dyno alert so you know if Google stops pushing data.

4. Point Stack Dyno to the table

• Update frontend-user/.env with BILLING_EXPORT_TABLE=project.dataset.table.
• If you rely on the FastAPI backend, set STACK_DYNO_BILLING_EXPORT_TABLE before launching.

Once complete, the spending flow, commitment coverage, and forecasting cards show real numbers instead of demo data.

---