15 AWS Cost KillersEvery Startup Should Check

Instances running 24/7 that are only used during business hours. Use auto-scaling groups or scheduled scaling to shut them down outside working hours.

Right-size your instances by analysing CPU and memory utilisation. Most startups over-provision by 2-3x because they guessed instead of measured.

If you have stable baseline workloads, Reserved Instances or Savings Plans can cut compute costs by up to 72% compared to on-demand pricing.

Orphaned EBS volumes from terminated instances sit there costing money. Audit and delete volumes that are not attached to any running instance.

Snapshots accumulate over time and are easy to forget. Implement a lifecycle policy to automatically delete snapshots older than 30-90 days.

Data accessed infrequently should be moved to S3 Infrequent Access or Glacier. Use S3 Intelligent-Tiering for unpredictable access patterns.

Without lifecycle policies, objects stay in Standard tier forever. Automate transitions to cheaper tiers and set expiration rules for temporary data.

NAT Gateways charge per GB of data processed. Move high-bandwidth services to public subnets or use VPC endpoints for AWS services to avoid NAT costs.

AWS charges for Elastic IPs that are not attached to a running instance. Release any unused Elastic IPs to stop unnecessary charges.

Database instances are commonly oversized. Analyse performance metrics and downsize to a smaller instance class if CPU and memory are underutilised.

Production databases run 24/7, making them ideal candidates for Reserved Instances. A 1-year commitment can save up to 40%.

CloudWatch Logs default to never expire. Set appropriate retention periods (7-30 days for most logs) and export older logs to S3 if needed.

Load balancers with zero or minimal traffic still incur hourly charges. Consolidate or remove ALBs and NLBs that serve no active targets.

Data transfer between regions is expensive. Keep services that communicate frequently in the same region and use CloudFront for global distribution.

Without proper tagging, you cannot attribute costs to teams or projects. Implement a tagging strategy and use AWS Cost Explorer to identify waste.