Cloud bills rarely grow because of one big decision. They grow because of hundreds of small ones nobody revisits. The good news: the first savings are usually the safest.
Where the money goes
In most environments I review, the biggest line items are the same:
- Compute that was sized for a peak that never came.
- Non-production environments running 24/7.
- Storage — old snapshots, unattached volumes and logs kept forever.
Low-risk wins
| Change | Typical effort | Risk |
|---|---|---|
| Stop dev/test outside working hours | Low | Low |
| Delete unattached volumes and old snapshots | Low | Low (after review) |
| Right-size over-provisioned VMs | Medium | Low–Medium |
| Move infrequently accessed data to cheaper tiers | Medium | Low |
| Commit to reserved capacity for steady workloads | Low | Financial only |
Make cost visible
Tag every resource with an owner and environment. When each team can see its own spend, most of the waste disappears without a central clean-up project.
