What a 15-Minute SLA Actually Means

Every managed IT provider advertises a response-time SLA, and almost nobody explains what the number actually measures. "15-minute response" can mean a guaranteed engineer on the problem, or it can mean an automated email that says "we got your ticket." Those are very different products at very similar prices, so it's worth knowing what you're buying.

A response-time SLA measures the gap between when an issue is reported (or detected by monitoring) and when a human being begins working it. It does not measure resolution — that depends on the problem. A dead switch and a forgotten password both start the clock, but one is a five-minute fix and the other is a truck roll. A serious SLA therefore tiers its commitments by severity: a site-down, business-halting incident gets a tighter response guarantee than a single-user inconvenience.

The details that matter are the ones providers bury. Is the SLA measured 24/7/365, or only during business hours? Is "response" a real engineer, or an auto-acknowledgment? What are the severity definitions, and who decides the severity — you or the vendor? What happens when the SLA is missed: a credit, an escalation, or nothing? If a provider can't answer those in one sentence each, the number on the brochure is marketing, not a commitment.

Here's how we structure ours, as a reference point. Critical incidents — anything stopping the business — get a 15-minute response, day or night, from a live engineer, not a bot. High-priority issues affecting a team get a one-hour response during business hours. Standard requests are same-business-day. Every response target is measured from our monitoring or your ticket, whichever comes first, because catching a failing drive before you notice it is the entire point of monitoring.

The right SLA for you isn't automatically the tightest one. A 15-minute critical response costs more to staff than a one-hour one, and if your business can tolerate an hour of downtime on a bad day, you shouldn't pay for the faster tier. Match the guarantee to what an outage actually costs you — then hold the provider to it in writing.