A worklist, not a wall of charts
Most cost tools show you that you spent money. CloudMonitor shows you where the recoverable money is and what to do about it. Every recommendation names the resource, estimates the monthly saving, rates the severity and effort, and lists the console steps to fix it.
What CloudMonitor looks for on AWS
- Unattached EBS volumes — storage you pay for that no instance uses. Snapshot if you need the data, then delete.
- gp2 volumes — gp3 is roughly 20% cheaper per GB and includes a 3,000 IOPS baseline. The gp2 vs gp3 guide covers the one gotcha to check.
- Large RDS storage that can move to gp3.
- Stopped instances that keep paying for their attached volumes and Elastic IPs.
- Older instance families — t2, m4, c4, r4 and earlier — where a current-generation type is cheaper for the same capacity.
- Elastic IPs attached to stopped instances, and load balancers worth a review.
Advisory ideas — S3 lifecycle tiering and version expiry, Lambda memory tuning, RDS sizing and Savings Plans — appear as savings ranges, because the real number depends on access patterns and usage you need to confirm. Hard savings and advisory ranges are kept apart, so the headline figure isn’t padded with guesses.
Waste Score and realized savings
The Waste Score expresses the share of resources that are idle or unattached as a 0–100 number. When a recommendation disappears between scans — because someone fixed it — CloudMonitor marks it resolved and books its saving as realized, which gives finance a running total of what the work actually returned.
Coverage
Recommendations cover AWS today. Azure and Google Cloud resources are in the inventory and the security checks. To weigh commitments and Spot, see pricing analysis; for the full playbook, read how to reduce AWS costs.