Free PDF Download · Updated for 2026
The 70-Point AWS Cost Optimization Checklist
The exact audit we run before every AWS cost engagement. Catches IPv4 hourly fees, NAT leaks, CloudWatch bloat, container over-allocation, and SageMaker waste.
Average savings: 30-40%. No sales calls. Instant PDF delivery.
Get the Full 70-Point Checklist
Instant PDF. No sales calls. No spam.
Fire-Drill Quick Wins
What You'll Find in 60 Seconds
These are the most common cost leaks. Run this quick scan before diving into the full checklist.
Full Checklist Preview
70 Cost Controls Across 9 Categories
Updated for 2026 AWS pricing changes including IPv4 hourly fees, NAT Gateway costs, and new savings opportunities.
Baseline, Visibility & Observability Guardrails
- Enable Cost Explorer daily views for fast spend attribution
- Enable Cost & Usage Report (CUR) to S3 for deep billing analytics
- Enable Cost Anomaly Detection for unexpected spikes
- Set AWS Budgets + alerts for top services and environments
- + 5 more in full PDF
EC2 Optimization
- Terminate idle EC2 instances (no CPU/network/disk for 7-14 days)
- Right-size instances using CPU, memory, network, and EBS metrics
- Replace static headroom with Auto Scaling for variable traffic
- + 5 more in full PDF
Containers (ECS, EKS, Fargate)
- Right-size Task CPU/Memory and K8s requests/limits to real usage
- Use Fargate Spot for tolerant stateless workloads and background jobs
- Split baseline vs burst capacity (standard for baseline, Spot for async)
- + 5 more in full PDF
Storage & S3
- Enable S3 Intelligent-Tiering for variable access patterns
- Apply lifecycle rules to transition cold data to Glacier classes
- + 6 more in full PDF
Databases (RDS, Aurora, DynamoDB, ElastiCache)
- Right-size RDS instance class based on real utilization
- Stop non-prod databases on schedules where possible
- + 6 more in full PDF
Lambda & Serverless
- Tune memory vs duration using Lambda Power Tuning
- Reduce timeouts using P95/P99 durations and fail fast
- + 6 more in full PDF
Networking & Data Transfer
- Audit Public IPv4 usage and remove public IPs from private subnet instances
- Migrate to IPv6 where supported to reduce public IPv4 hourly charges
- + 6 more in full PDF
Special Workloads (AI/ML, SageMaker)
- Stop idle SageMaker Notebooks (closing the tab does not stop billing)
- Right-size inference endpoints for real traffic volume
- + 3 more in full PDF
Purchasing & Governance
- Commit with Savings Plans only after usage stabilizes
- Split coverage: Savings Plans for baseline, Spot for burst, On-Demand for tail
- + 6 more in full PDF
Operationalize It
Run It As A 3-Week Loop
A rotation that prevents "fix once, leak again" AWS spending. Repeat monthly or quarterly for continuous optimization.
- Tags, budgets, anomaly alerts
- IPv4 tax, NAT, VPC endpoints
- Data transfer and routing paths
- EC2 right-sizing, Spot, Graviton
- Fargate Spot, requests/limits
- Lambda memory tuning, ARM64
- S3 lifecycle, Intelligent-Tiering
- Snapshot cleanup, gp3 migration
- RDS/DDB/Redis right-sizing
This checklist covers the what. For help with the how, see our AWS consulting services or AWS security audit for teams that need compliance and cost work done together. Already using managed services? Our managed services include ongoing cost monitoring as part of the package.
Want Us To Find The Savings For You?
The checklist shows you what to look for. Our free 48-hour diagnostic shows you exactly how much you can save.
Same audit. Done by senior AWS engineers. Delivered in 48 hours.