- 01Managed hosting means TLS, DNS, backups, monitoring, patching, and on-call all run without anyone having to remember to do them.
- 02When something needs a judgment call — an unusual spike, a patch window, a restore — a person makes the call with you, not a ticket queue.
- 03Product decisions and application code stay yours; the infrastructure underneath stops being on your list.
"Managed" is doing a lot of work in that sentence
Every hosting provider says "managed" somewhere on its pricing page. In practice it can mean anything from "we'll email you when a certificate is about to expire" to "we own every piece of this stack end to end and you never think about it again." The difference is the hours it actually gives back — not the word itself.
The maintenance list nobody budgets time for
None of this is difficult work. It's just constant, easy to postpone, and expensive the one time it's missed:
- TLS certificates — issued and renewed automatically, before they expire
- DNS — records kept correct as services and subdomains change
- Backups — taken daily, without a script someone has to remember to run
- Monitoring — uptime and error-rate watched continuously, not "when someone reports it"
- Patching — the operating system and runtime kept current against known vulnerabilities
- On-call — someone accountable when it breaks at an inconvenient hour
Each one is small on its own. Together, they're a part-time job that most teams building a product would rather not staff — and the reason "managed" exists as a category at all.
What "a human answers" actually means
A lot of managed-hosting marketing quietly means "there's a dashboard and a status page." Here it means what it says: when something needs a decision — a spike that looks unusual, a patch that needs a maintenance window, a restore from backup — a person on the other end makes the call with you, not a ticket queue that routes you to documentation.
The hours it gives back
- Without managed hosting
- Someone on the team tracks certificate expiry, writes and tests a backup job, sets up monitoring, schedules patch windows, and carries the pager — on top of their actual job.
- With it
- All six items above run without anyone remembering to do them, and the one time something needs a judgment call, a person is already looking at it.
- What's still yours
- Product decisions, application code, and anything specific to how your app behaves — the infrastructure underneath it stops being on your list.
Where to see the specifics
The security page covers exactly what's patched and how backups and TLS are handled; support covers who answers and how fast; platform covers how the edge network, compute, and DNS fit together underneath all of it. If any of that is already true of the way you host today, that's most of what "managed" is supposed to buy you.