Google Cloud cost management, keyless from BigQuery

What CloudQuell ingests

Your BigQuery Detailed usage cost export — the same invoice-grade record Google bills you from.

Detailed usage cost export

Read directly from BigQuery, down to project and SKU. This is Google's invoice-grade billing export, not an estimate.

Projects & labels

Each project lands as its own sub-account; resource and project labels merge into tags (resource labels win on conflicts) for allocation and tag coverage.

Committed Use Discounts

CUD-covered usage lands with its commitment identity, so GCP counts toward commitment coverage and Effective Coverage %.

FOCUS: on the roadmap

CloudQuell reads the Detailed export today; Google's FOCUS export is still preview, and FOCUS support here is on the roadmap.

What the data looks like normalized

A BigQuery export row lands on the same schema as an AWS line item — project as sub-account, labels as tags.

How it connects

Keyless. You grant a service account two read-only BigQuery roles — there is no key to store, rotate, or leak.

Credential

None to hand over. Access works through Google's Workload Identity Federation: you grant CloudQuell's service-account email BigQuery Job User on the export project and BigQuery Data Viewer on the export dataset. Revoking those grants revokes CloudQuell.

Access scope

Read-only, scoped to the billing export dataset — no access to any other dataset or resource. Nothing is written back; the connect step runs a BigQuery dry-run that scans no data.

Setup

Three steps: enable the Detailed usage cost export, grant the two IAM roles, then enter the billing-account, project, and dataset and verify. Costs typically appear within minutes.

Time to first data

Google backfills only the current and previous month on a new export, and a fresh export takes 24–48 hours to populate; history grows forward. Daily sync re-pulls the current and previous invoice months.

What it unlocks

GCP spend flows into the same views as the rest of your bill, with two roadmap gaps noted.

• One ledger

  GCP spend by project, service, SKU, and label in the same dashboard, reports, and cost centers as AWS, Snowflake, and your AI spend.

• Allocation

  Projects as sub-accounts and labels as tags feed the same allocation rules and cost-center hierarchy as everything else.

  How allocation works →

• Budgets & anomalies

  Budget thresholds, trend and anomaly rules cover GCP spend, routed to the channel your team already watches.

• Commitments

  CUD coverage shows on the Commitments page and counts toward Effective Coverage %. Savings recommendations are AWS-only for now, and CUD expiry alerts aren't covered yet.

  Effective Coverage % →

Common questions

Is the Google Cloud integration available now?

Yes. It is generally available to every organization on every plan, Free included (within the Free plan’s 3 connected integrations) — connect it from the Integrations page. Reporting, budgets, allocation, alert rules, anomaly detection, and committed-use-discount coverage all include GCP spend. Savings recommendations are AWS-only for now, and CUD expiry alerts aren’t covered yet.

How does CloudQuell connect to GCP?

Keyless, through Workload Identity Federation. You grant CloudQuell's service account two read-only BigQuery roles — Job User on the export project and Data Viewer on the export dataset — and nothing else. There is no key or secret to store or rotate; removing the two IAM grants revokes access.

What GCP data does it read?

The Detailed usage cost export in BigQuery — Google's invoice-grade billing record, down to project and SKU. Projects appear as sub-accounts and resource and project labels arrive as tags. Google's FOCUS export is still preview, so CloudQuell reads the Detailed export today; FOCUS support is on the roadmap.

How far back does GCP history go?

Google backfills only the current and previous month when an export is first enabled, and a new export takes 24–48 hours to populate — so history grows forward from there. If you have already been exporting billing to BigQuery, you get that full history.