Service level agreement

We target 99.9% monthly availability for managed hosting, measured from external monitoring nodes at one-minute intervals.

Planned maintenance is announced at least forty-eight hours in advance, scheduled outside business hours, and excluded from the calculation.

Severity 1 — the service is down, or data is at risk. Nothing else is worked on until it is resolved.

Severity 2 — a major function is broken but the service is usable. Worked continuously during covered hours.

Severity 3 — a minor fault, cosmetic issue or change request. Scheduled into the normal queue.

Standard — next business day, Sunday to Thursday, 09:00–18:00.

Priority — four hours during covered hours, Sunday to Thursday, 08:00–20:00.

Dedicated — one hour, twenty-four hours a day, every day, with an on-call engineer paged automatically for Severity 1.

Response time means a human acknowledging and beginning work, not an automated receipt.

Backups run nightly as standard and hourly on Dedicated, retained for thirty days, with off-site replication on Dedicated.

Restores are tested monthly rather than assumed. An untested backup is not a backup.

Recovery point and recovery time objectives are stated per tier in your contract.

Where availability falls below the committed figure in a calendar month, service credits are applied to the following invoice on the scale set out in your contract.

Credits are applied automatically. You should not have to notice and ask.

Outages caused by third-party platforms, by changes made by you or another vendor without our involvement, or by force majeure, are outside this commitment. We will still help you resolve them.