Service-Level Agreement SLA

Service-Level Agreement is the contractual promise built into a managed-services relationship that defines what good looks like, what bad looks like, and what happens when bad happens. An SLA without consequences is marketing copy; an SLA with credits or penalties when missed is a contractual obligation.

What an SLA typically specifies

Response time — how quickly the provider acknowledges a reported issue, usually broken out by severity:

• Critical (everything down): minutes to a couple of hours
• High (major function impaired): hours
• Medium (workaround exists): same business day
• Low (cosmetic or convenience): next business day

Resolution time — how quickly the issue is resolved, often softer because resolution depends on vendors outside the provider’s control.

Uptime guarantees — typically expressed as nines (99.9% = ~8.77 hours of allowed downtime per year). Important to know what’s included:

• Does scheduled maintenance count against uptime?
• Does third-party outage count?
• What’s the measurement window — monthly, quarterly, annually?

Remedies when SLA is missed — credits on the next invoice, escalation paths, in some cases termination rights.

Common SLA mistakes to watch for

• Vague language — “best effort”, “commercially reasonable”, “we will try to”. These mean nothing in a dispute.
• Severity definitions that put everything in “medium” — if there’s no objective definition of what’s critical, you’ll never get critical-severity response.
• Uptime that excludes everything important — “uptime excludes scheduled maintenance, force majeure, third-party outage, customer-caused issues…” after the exclusions, the SLA covers very little.
• No remedies — without a credit or penalty, a missed SLA has no consequence and isn’t really an SLA.
• No measurement methodology — who decides whether the SLA was met? A vendor’s own dashboard? Independent monitoring? Your stopwatch?

Bytes Unlimited’s SLA shape

Our managed-IT clients get a 1-2 hour response SLA during normal coverage hours, with severity tiers in the contract. We document RTO/RPO targets per client based on their actual business needs (see RTO and RPO). For hosting plans, uptime is explicitly defined and SiteGround’s underlying infrastructure SLA flows through. We don’t run premium 24/7 emergency tiers because most clients don’t pay for what they wouldn’t use — but emergencies for managed clients are handled on an exception basis.