A 10-point checklist to cut your cloud bill

Cloud bills rarely jump overnight. They creep — a test server left running here, an oversized database there. Here's the checklist we run through with every new client.

Quick wins

1. Switch off what nobody uses. Old test environments, forgotten virtual machines and unattached disks are the most common waste we find.
2. Right-size your servers. If a machine averages 10% CPU, it's probably two sizes too big.
3. Schedule non-production. Development and staging environments rarely need to run nights and weekends.
4. Clean up storage. Move old backups and logs to cheaper storage tiers, and set them to expire automatically.
5. Check data transfer. Traffic between regions and out to the internet adds up — keep chatty services close together.

Structural savings

1. Commit to what you'll definitely use. Reserved instances or savings plans cut the price of steady workloads significantly.
2. Use managed services wisely. A managed database often costs less than the engineering time to run your own.
3. Autoscale. Let capacity follow demand instead of paying for peak traffic all day.
4. Containerise where it fits. Packing several small services onto shared infrastructure improves utilisation.

Keep it under control

1. Make costs visible. Tag resources by project and team, set budgets and alerts, and review spend monthly.

The cheapest server is the one you remembered to switch off.

How much can you save?

In our experience, a first review typically finds 20–40% savings without any impact on performance — more if nobody has looked closely for a year or two.

Want a second pair of eyes on your cloud setup? Talk to our cloud team.