When to run it
A finance review of cloud spend, a budget overrun, or a FinOps tool full of recommendations nobody has acted on.
Cut the waste.Keep the headroom.
Rightsizing cloud spend without breaking SLOs
The playbook in one paragraph
The cloud bill keeps climbing and nobody is sure which resources are safe to shrink. Ana finds idle and oversized resources, checks them against real utilization and SLOs, and ships the changes as reviewed IaC.
Works with
- Cloud billing and cost explorer data
- Metrics from your observability stack
- IaC repositories
- Kubernetes requests and limits
What Ana needs from you
- SLOs and minimum headroom per service
- Commitments (reserved instances, savings plans) in place
- Which environments are in scope
How Ana runs it
Step by step.
5 steps from trigger to done - each one visible to your team as it happens.
- 01
Attribute spend
Ana maps cost to services, teams and environments, including untagged resources, so savings have an owner.
- 02
Find candidates
Idle resources, oversized instances and node pools, over-provisioned Kubernetes requests, unattached volumes and old snapshots are listed with their cost.
- 03
Check against reality
Each candidate is checked against utilization history and the service’s SLO headroom; anything that would put an SLO at risk is dropped.
- 04
Change in code
Ana ships the changes as IaC and manifest PRs with the expected saving and the rollback, rolling out in stages.
- 05
Verify
After each change Ana watches latency, errors and saturation, and reports realized savings against the estimate.
Who approves what
Approval gates
- Production resizing
- Service owner
- Deleting resources
- Resource owner, with backup confirmation
- Commitment purchases
- Finance - Ana only recommends
What you get
Deliverables
- 01Cost attribution by service and team
- 02Rightsizing PRs with rollback
- 03Post-change SLO verification
- 04Realized-savings report
What usually goes wrong
And how Ana handles it.
Risk
Saving money by causing an outage
SLO headroom is a hard constraint, and changes roll out in stages with automatic verification.
Risk
Deleting something still needed
Deletions require owner approval and a snapshot or backup first.
FAQ
Common questions.
No. Ana recommends commitments; purchasing stays with finance.
Keep going
Related playbooks.
SRE / DevOps
Reconciling infrastructure drift
Hot-fixes in the console and manual changes pile up until Terraform plans become scary. Ana finds every drifted resource, decides with you whether code or cloud is right, and brings them back in line.
Read the playbook →SRE / DevOps
From a 3 a.m. page to root cause and fix
An alert pages the on-call engineer. By the time they open the laptop, Ana has correlated metrics, logs, traces and recent deploys, named the likely cause, and drafted the remediation for approval.
Read the playbook →Software
Upgrading a framework across dozens of services
A major framework or runtime version is going end-of-life and it touches every service. Ana plans the migration, upgrades each service in its own branch, fixes what breaks, and hands your team small, tested pull requests instead of one terrifying diff.
Read the playbook →See it in your stack
Run this playbook on your systems.
Thirty minutes, your repositories and cloud, one real instance of this job.