Cloud spend
Multi-cloud cost in one place, with the trend that matters more than the total.
FinOps → Cloud Spend collects cost across providers.
Connectors
| Provider | Method |
|---|---|
| DigitalOcean | API |
| Hetzner | Estimated from inventory |
| AWS | SigV4-signed Cost Explorer |
| OVH | Signed API |
| Manual | Enter figures for anything not connected |
Hetzner is an estimate from the resources present, not a billed figure, and the page says so. AWS and OVH connectors need verification against a real account before the numbers are trusted.
What is shown
| Month to date | Against the same point last month |
| Trend | Where the month is heading |
| By provider | |
| By service or tag | Where the provider reports it |
| History | Monthly snapshots |
The trend is the number
Month-to-date against the same point last month is the reading that catches things while they can still be changed. The total at month end is a post-mortem.
What usually shows up
A forgotten environment. A staging cluster nobody has used since a project ended. It is the single most common finding.
Storage growth. Storage only ever goes up unless somebody deletes something, and nobody does. Pair this with data retention.
Egress. Frequently the line nobody has looked at, and frequently caused by something moving data between regions unnecessarily.
Manual entry
Anything without a connector can be entered by hand — a colocation invoice, a SaaS subscription, a leased line. Better a complete picture with some manual rows than an automatic one that omits a third of the spend.
Related
- Capacity — what you are buying, rather than what it costs
- Warehouse cost — what queries cost
Where this behaviour lives: backend/internal/integration/costproviders.go, backend/internal/worker/cost_sync_worker.go. If the code and this page disagree, the code is right — please fix the page.
Part of Analytics and cost — Dashboards, SLOs, reports and what the infrastructure costs.