Legal
Service Level Agreement
What we commit to on availability and support response, how it's measured, and what you get if we miss it.
This document is in preparation. It is not yet in force.
Below is the structure the final text will cover, published so you can see what will be in it. We have deliberately not filled it with generated legal wording — text that reads like a contract and binds nobody is worse than an honest gap. If you need the current draft for a review, email sales@securive.net.
What the commitment covers
The control plane, the backup service and the restore path are three different things with three different availability profiles. Conflating them produces a number that means nothing.
Availability target
The uptime figure we will actually contract to, by plan. This must be a number the operations team has signed off on, not a marketing round number — see docs/CONTENT-TODO.md.
How availability is measured
Measurement window, what counts as downtime, and where the measurement comes from. An SLA without a stated measurement method is unenforceable in practice.
Exclusions
Scheduled maintenance with notice, customer-side network failures, and force majeure. Kept narrow — an SLA that excludes everything is theatre.
Support response times
Severity definitions and first-response targets by plan, matching what /pricing states.
Service credits
The remedy, how it's calculated, how it's claimed, and the deadline for claiming.
Maintenance windows
Notice period, typical duration, and whether backups continue during maintenance.
Recovery commitments
What we commit to on restore initiation, and what remains bounded by the customer's own bandwidth and workload size — because it is, and the document should say so.