Know the bill, then cut it
Cost visibility, forecasting and rightsizing across every account, with the allocation model finance teams need to actually assign spend.
The bill arrives after the money is spent, split across accounts nobody owns, in a format no engineer can act on.
Cost explorer with breakdowns by service, account, region and tag.
A cost explorer that breaks spend down by service, account, region and tag across all three providers, so a number can always be traced back to the thing that generated it. This is the base layer the rest of FinOps depends on: you cannot rightsize or forecast what you cannot attribute.
- Breakdowns by service, account, region and tag
- AWS, GCP and Azure in the same explorer
- Trace any total back to the underlying resources
FinOps reports
Scheduled, shareable cost reports delivered by email.
Cost conversations happen on a monthly cadence with people who will not log into a cloud console. Reports are scheduled and delivered by email, so budget owners get their numbers without being given platform access.
- Scheduled on a recurring cadence
- Shareable with stakeholders outside the platform
- Covers AWS, GCP and Azure spend
Cost forecasting
ML-based cost and usage forecasts with accuracy tracking against actuals.
Forecasts are only useful if you know how wrong the last one was. Cost and usage projections are generated from your own history, and accuracy is tracked against actuals so the forecast earns trust over time instead of being quietly ignored.
- ML-based cost and usage projections
- Accuracy tracked against actual spend
- Supports budget planning rather than guesswork
Waste management
Finds idle and orphaned resources: unattached volumes, unused IPs, stopped-but-billing instances.
A meaningful share of most cloud bills is resources nobody is using: volumes detached from a deleted instance, reserved IPs pointing at nothing, instances stopped but still billing for attached storage. These get found and listed so they can be cleaned up as routine work.
- Unattached volumes and orphaned snapshots
- Unused reserved IP addresses
- Stopped instances still generating charges
Compute rightsizing
Instance-level resize recommendations based on observed utilization.
Instance sizes are usually chosen once, defensively, and then never revisited. Recommendations are generated per instance from observed utilization, so the resize decision is based on how the workload actually behaves rather than what someone assumed at launch.
- Per-instance recommendations from real utilization
- Projected saving attached to each recommendation
- Available across AWS, GCP and Azure
Shrink or expand EBS, Persistent Disk and Azure Managed Disk with guided remediation.
Storage is the quietest form of cloud waste because volumes only ever grow. XamOps recommends the right size for EBS, Persistent Disk and Azure Managed Disk volumes and walks through the remediation, including the filesystem work that makes a resize safe.
- Shrink or expand based on observed usage
- EBS, Persistent Disk and Azure Managed Disk
- Guided remediation including filesystem expansion
Reservations and commitments
Coverage, utilization and purchase recommendations for RIs, Savings Plans and CUDs.
Commitment discounts are the largest single lever on a cloud bill and the easiest to get wrong in both directions. Coverage and utilization are tracked for existing Reserved Instances, Savings Plans and CUDs, with purchase recommendations for the gap.
- Coverage and utilization for existing commitments
- Purchase recommendations sized to real usage
- RIs, Savings Plans and CUDs in one view
Flex Compute
Capacity and KPI view for elastic compute pools.
Elastic compute pools need to be judged on capacity and efficiency rather than a raw instance count. Flex Compute gives a capacity and KPI view of those pools so scaling behaviour is visible alongside its cost.
- Capacity view across elastic compute pools
- KPIs for pool efficiency
- Currently AWS
Pricing calculator
Build and price an architecture before you deploy, with a saveable estimate cart. AWS live, GCP and Azure in progress.
The cheapest time to fix an architecture cost is before it is deployed. Build a proposed architecture, price it, and keep the result in a saveable estimate cart so options can be compared and shared instead of rebuilt from scratch each time.
- Price an architecture before you deploy it
- Saveable estimate cart for comparing options
- AWS available now, GCP and Azure in progress
Resource labels
Internal team, department and project tagging for cost allocation without writing tags back to the cloud provider.
Cost allocation usually stalls on tagging, because writing tags back into a production account needs change approval. XamOps keeps team, department and project labels internally, so allocation can start immediately without touching provider tags.
- Team, department and project labels
- No writes back to the cloud provider
- Allocate cost without a tagging project first
FinOps, answered.
Native tools stop at reporting one provider. XamOps puts AWS, GCP and Azure in one explorer, adds forecasting with accuracy tracked against actuals, and connects each finding to a remediation, including rightsizing and commitment purchases.
Yes. Resource labels let you assign team, department and project ownership inside XamOps without writing tags back to the cloud provider, so allocation can start before a tagging standard is agreed and rolled out.
Idle and orphaned resources such as unattached volumes, unused reserved IP addresses and instances that are stopped but still generating charges, alongside oversized compute and storage identified by rightsizing.
Yes. Coverage and utilization are tracked for existing Reserved Instances, Savings Plans and CUDs, with purchase recommendations sized against real usage.
The pricing calculator lets you build and price a proposed architecture and keep it in a saveable estimate cart. AWS is live today, with GCP and Azure in progress.
FinOps is one module of 8.
Ready to automate finops?
30-minute walkthrough. We connect to a sandbox and show this module running against real infrastructure.