Managed cloud
Someone has to own the infrastructure at 3am
Cloud is not a service you buy once. It is an ongoing set of decisions about cost, access, capacity and recovery — and it degrades quietly when nobody is making them.
In short
Managed cloud means someone is accountable for the infrastructure a business runs on: monitoring and alerting, patching and dependency upgrades, backups with restores that are actually tested, access control and secrets, capacity and cost, and incident response when something breaks. The distinguishing question is not which provider you are on — it is whether a specific person is responsible when a restore is needed, and whether anyone has proved the restore works.
When it applies
Signals this is the work
If more than one of these is true today, the problem is usually further along than it looks from the outside.
The bill grows and nobody can explain it
Spend rises steadily. Nobody can point at which workload is responsible or what could safely be switched off.
Backups exist but have never been restored
The backup job reports success. No one has proved a restore produces a working system.
Patching happens after an incident
Updates are applied reactively, usually in a hurry, usually at the worst time.
One person holds the credentials
Access is informal. Offboarding has never been tested and would be uncomfortable.
Alerts go to a channel nobody reads
Monitoring exists but has been muted into irrelevance by noise.
How we do it
The sequence
- 01
Inventory what exists
Every running resource, what it costs, what depends on it and whether anything still needs it. This alone usually pays for itself.
- 02
Fix access before anything else
Named accounts, least privilege, secrets out of code and config, and an offboarding path that has been walked through once.
- 03
Monitor what a customer would notice
Alerting on the symptoms that matter, tuned to the point where an alert firing means something. A noisy monitor is worse than none.
- 04
Prove the restore
Backups are verified by restoring them, on a schedule, into a working environment. Until that has happened the backup is an assumption.
- 05
Patch on a rhythm
Predictable maintenance rather than emergency response, with a rollback path for each change.
- 06
Review cost as an operational habit
Right-sizing, commitments, unused resources and storage tiers reviewed regularly, with the savings shown rather than claimed.
What we will not do
The refusals
- Migrate to a different provider to fix a problem that is about how the system is built.
- Run Kubernetes because it is expected, on a workload that does not need it.
- Report uptime while backups have never been restore-tested.
- Mark up cloud spend. Your accounts, your billing relationship, our fee separately.
Questions
Before you ask
Tell us what isn't working.
One process, one system, one decision you are stuck on. We will come back with how we would approach it, what it would take, and whether it needs building at all.