Monitor
See every dollar, attributed.
- Cost attribution — bill + usage + network
- Allocation groups, per-team cost
- Budgets with live status
- Anomaly detection — three detectors
- Webhooks, API, personal access tokens
pricing
The plans differ by what the platform is allowed to write — not by what it can see. Every plan starts with a free trial, and every price conversation can start with a scoped POC.
See every dollar, attributed.
See it, then fix it safely.
Scale it across the company.
| Feature | Monitor | Optimize | Enterprise |
|---|---|---|---|
| Everything in every plan | |||
| Cost attribution — bill + usage + network | ✓ | ✓ | ✓ |
| Allocation groups, per-team cost | ✓ | ✓ | ✓ |
| Budgets with live status | ✓ | ✓ | ✓ |
| Anomaly detection — three detectors | ✓ | ✓ | ✓ |
| Webhooks — HMAC, retry, dedupe | ✓ | ✓ | ✓ |
| REST API + personal access tokens | ✓ | ✓ | ✓ |
| Automation | |||
| Workload — rightsizing, repack, commitments | — | ✓ | ✓ |
| Infrastructure — Spot, Karpenter, idle nodes | — | ✓ | ✓ |
| Approval-gated execution + verification | — | ✓ | ✓ |
| Rollback — always available | — | ✓ | ✓ |
| Enterprise | |||
| SAML + OIDC SSO | — | — | ✓ |
| 365-day retention | — | — | ✓ |
14–30 days, with success criteria agreed before we start. You connect one cluster; we set up attribution, budgets, and a first set of recommendations. Automation stays off unless you turn it on.
The agent reads metrics, events, and manifests; it never mutates your cluster. Writes happen only as approved actions on the Optimize and Enterprise plans, and each one goes through the same cluster-safety gates and leaves an audit entry.
Cost visibility is never revoked — read access keeps working so you can still see your numbers. Only writes and automation stop. Rollbacks for actions already taken remain available.
Yes. That’s the intended path: one non-prod cluster first, approvals by hand, then widen per policy as the audit trail earns trust.
14–30 days with success criteria agreed up front. You connect a cluster; we build attribution, budgets, and an initial recommendation set. You keep the configuration either way.
AWS is the v1 cloud. The others are on the roadmap with no committed dates — we turn a cloud on when its billing and attribution are correct, not when it’s present.
14 days of Monitor: full cost visibility, budgets, and the API — no automation, no card. When the trial ends, read-only visibility continues; only paid features stop.