The Definitive AWS Cost Optimisation Guide for Growing Startups
We cut one client's AWS bill by 60% in three weeks. Here's the exact playbook.
AWS bills grow faster than engineering teams expect. The default settings, the convenience choices made at 2am, and the test resources nobody deleted — they compound. Here is the playbook we use to systematically cut costs without touching production stability.
Step 1: Enable AWS Cost Explorer and set up billing alerts immediately. You cannot optimise what you cannot see. Tag every resource by environment, team, and feature from day one. Untagged resources are the black holes of cloud spending.
Step 2: Right-size your EC2 and RDS instances. The single biggest quick win is almost always an oversized database. Use AWS Compute Optimizer and CloudWatch metrics to identify instances running at under 20% utilisation and downsize them. We typically save 30-40% on compute alone.
Step 3: Move to Graviton. ARM-based Graviton instances offer 20-40% better price-performance than x86 equivalents for most workloads. The migration is usually a single Terraform change.
Step 4: Audit your data transfer costs. S3 to CloudFront is free. EC2 to S3 in the same region is free. Cross-region and internet egress is expensive. Rerouting traffic through CloudFront and keeping data in the same region eliminates most surprise bills.
Step 5: Use Savings Plans for predictable workloads. Committing to Compute Savings Plans for your baseline load saves 30-40% with no operational risk. Leave spot instances for batch workloads and dev environments.
Ready to work with us?
Let's build something extraordinary together.