Multi-cloud cost normalization
Challenge
Each cloud provider exports billing data in different formats:
| Provider | Primary export |
|---|---|
| AWS | Cost and Usage Report (CUR 2.0) in S3 |
| Azure | Cost Management exports / API |
| GCP | BigQuery billing export |
| Kubernetes | Metrics collector + parent cloud infra costs |
Costimizer approach
- Ingest provider-native billing exports on their natural cadence
- Normalize line items to a common schema (service, resource, region, cost, usage quantity)
- Enrich with discovered resource metadata and tags
- Aggregate into unified views: Cost Explorer, pools, owners, recommendations
Unified dimensions
After normalization, filter and group costs consistently by:
- Cloud / data source
- Service and resource type
- Region
- Pool and owner
- Cost type (compute, storage, network, etc.)
- Kubernetes namespace (when connected)
Currency and rounding
Costs display in the billing currency of each data source. Multi-currency orgs should note exchange rate handling in exported reports.