When we review AWS environments, S3 is often where unnecessary storage shows up first.

Old object versions are still sitting in buckets. Multipart uploads were started months ago and never finished. Lifecycle policies were never configured or were added inconsistently.

None of this causes outages. But it does increase storage month after month.

For many teams, S3 becomes one of the simplest places to recover costs.

The FinOps community regularly sees organizations reduce cloud spend by 25 to 40 % once optimization practices are applied. On S3, lifecycle policies are usually the easiest starting point. They can automatically remove noncurrent object versions and clean up incomplete uploads without affecting production data.

The real improvement comes when those policies are defined in code instead of the console. With AWS CDK, lifecycle policies live in your infrastructure repository. They are reviewed through pull requests and applied the same way across environments.

That approach prevents the most common problem with S3 cost optimization. Someone fixes a bucket once, but the next ten buckets are created without the same policy.

If you want to see exactly how this works, including the Storage Lens metrics to look at and the CDK code used to implement the policies, we explain it in detail in our technical guide.

Read the full article on Stable: https://www.stableapp.cloud/blog/s3-cost-optimization-automating-savings-with-aws-cdk-and-lifecycle-policies/