Cut your AWS bill 18-30% in the first year.
Without touching application code.
AWS gives you the tools to do FinOps well. Most mid-market shops never use them. The first-year savings come from operational discipline (tagging, lifecycle policies, Savings Plans rebalancing), not from architectural rewrites. Below is the six-layer pattern we apply.
Each layer compounds on the one before it.
We work top-down. Tagging first (because nothing else works without it), then quick wins, then commitment optimization, then right-sizing, then storage tiers, then architecture. The first three layers usually deliver 80% of the year-one savings.
- 01
Foundation: tagging discipline
Without consistent resource tags you cannot allocate costs to teams, products, or environments — which means you cannot make informed cost decisions. Most FinOps engagements start with a 2-week tagging cleanup driven by SCPs that prevent untagged resources from being created.
Typical find: Roughly 30-50% of resources have inconsistent or missing tags at engagement start. - 02
Quick wins: idle resources
EBS volumes detached from terminated instances, snapshots from 2019, idle ELBs, dormant RDS instances, EC2 instances stopped for months but still billing for storage. Trusted Advisor + Compute Optimizer surface most of this in the first hour.
Typical find: 5-15% of bill recovered in week one from killing things nothing depends on. - 03
Compute Savings Plans rebalancing
Most mid-market shops have legacy Reserved Instances locked to specific instance families that have since changed. Compute Savings Plans (compute-flexibility) cover any instance family in the region. Quarterly portfolio review keeps coverage at 60-80% of baseline without over-committing.
Typical find: 8-15% additional savings from converting RIs to Savings Plans + rightsizing coverage. - 04
EBS gp2 → gp3 + S3 lifecycle
gp3 is faster AND cheaper than gp2 for nearly every workload. Most clients running on gp2 do so because nobody migrated. S3 lifecycle policies (Standard → Intelligent-Tiering → Glacier) on long-retention buckets are pure cost savings with no performance impact.
Typical find: 3-8% saved on storage with zero application changes. - 05
Right-sizing compute
Compute Optimizer identifies over-provisioned instances based on actual utilization. Engineers default to the next-larger size "to be safe" — Compute Optimizer typically finds 20-30% of instances over-sized.
Typical find: 5-12% saved by matching instance size to actual usage. - 06
Workload-level architecture
Spot instances for batch / dev / non-critical workloads. Aurora Serverless v2 for variable-load databases. Lambda for event-driven workloads instead of always-on EC2. These are larger architectural changes with longer payback but compounding returns.
Typical find: Where the second-year FinOps gains come from.
FinOps is operational discipline, not a tool.
The companies that maintain low AWS spend over years run quarterly FinOps reviews with named accountability. The companies that drift do not. We embed the cadence as part of the engagement so the first-year savings hold.