Cloudflare launches Billable Usage API for self-serve accounts

Cloudflare launched a Billable Usage API for self-serve accounts on August 3. The announcement says one endpoint returns account usage and cost by product and charge period, covering usage-based services including Workers, R2, D1, Workers AI, Vectorize, Images, and Stream.

Cost data gets a programmatic shape

Each response row identifies a service and family, charge-period boundaries, consumed and pricing quantities, contracted cost, currency, cumulative billing-period totals, and zone attribution where applicable. Data was updated daily at launch. Access uses an API token with Billing Read permission, which supports a read-only integration rather than a general account credential.

Cloudflare deliberately mapped many fields to the FinOps Open Cost and Usage Specification, or FOCUS. The company also states that the response was not fully conformant because some required columns were absent. Enterprise contract coverage, finer time windows, and forecasting were listed as future work; the live endpoint initially applied to self-serve accounts.

Daily visibility enables tighter controls

Teams can ingest the rows into their existing cost pipeline and compare product changes with deployments or traffic shifts. A useful implementation preserves the raw charge period and unit instead of converting every service into one generic usage number. Cost and quantity should be reconciled against the invoice before the API drives formal financial reporting.

Automation needs a stable owner for the token, explicit handling for late or corrected rows, and alerts on missing daily data. A program that provisions Cloudflare resources can consult the same feed, but the API itself does not impose a spending limit.

The release replaces screenshots and manual exports with a machine-readable account view. Its clearest benefit is earlier attribution: teams can see which product and zone moved before the end-of-month invoice arrives.