How I Built a System to Cut AWS Bills by 20–40% (And Packed It Into a Kit)

Every startup founder I know has had the same moment.

You open your AWS billing dashboard on the first of the month and your stomach drops.

“Why is this $800? It was $200 last month.”

You dig in. EC2 instances running 24/7 that nobody uses. S3 buckets storing logs from 2021. RDS snapshots piling up silently. NAT Gateway charges you forgot existed.

It’s not that you’re bad at engineering. It’s that AWS wants you to forget. Every forgotten resource is revenue for them.

The Real Problem With AWS Cost Optimization

Most guides tell you the same things:

• “Use Reserved Instances”
• “Enable Cost Explorer”
• “Right-size your EC2s”

That’s fine advice. But it assumes you already know where to look and what questions to ask.

Most teams don’t. So costs creep up month after month.

What Actually Works

After going through this pain myself, I built a repeatable system — a checklist-based approach that covers every major AWS cost leak:

• Compute — idle EC2s, oversized instance types, missing auto-scaling
• Storage — forgotten S3 buckets, old EBS volumes, unattached snapshots
• Database — RDS idle time, over-provisioned Aurora clusters
• Networking — NAT Gateway abuse, unused Elastic IPs, data transfer costs
• Monitoring — CloudWatch log retention, unused dashboards

Going through this systematically, most teams find 20–40% waste they didn’t know existed.

The Kit

I packaged everything into the AWS Bill Shock Fix Kit — checklists, reference sheets, and a step-by-step audit process you can run in an afternoon.

If your AWS bill is giving you anxiety, grab it here.

One month’s savings will pay for it many times over.

---

Originally published on Medium.