A service level agreement (SLA) is meant to answer a simple question: what can I expect when something goes wrong? In practice, many SLAs are written in a way that makes the answer sound better than it is. Understanding a few terms makes it much easier to compare providers on substance rather than headline numbers.
Response Time Is Not Fix Time
Response time is how quickly the provider acknowledges your request — often an automated email, or a technician opening the ticket. It tells you the queue is being watched. It does not tell you that anyone has started solving your problem.
Resolution time is how long it takes to actually fix the issue. This is what your staff care about, and it is the figure many providers are reluctant to commit to, because it is harder to control. A provider that offers only a response time is giving you a promise that is easy to keep.
A useful test: Ask "what happens at the end of the response window?" If the honest answer is "you get an email saying we've received it", the number is about acknowledgement, not help.
Priority Levels Matter More Than the Headline
A sensible SLA doesn't treat every ticket the same. A single user who can't change their desktop background and a whole office with no internet should not sit in the same queue. Look for a clear priority structure, with defined criteria for each level.
- Critical: Business-stopping issues affecting many people or a core system.
- High: Serious disruption to an individual or small group.
- Normal: Standard faults and requests with a workaround available.
- Low: Minor issues, questions and planned work.
The important details are who decides the priority, and whether it can be escalated if the impact turns out to be bigger than first thought.
The Small Print Worth Reading
- Support hours. Does the clock run 24/7, or only during working hours? A "1-hour response" that starts at 9am the next morning is not what it sounds like.
- How the clock is measured. Many SLAs pause the clock while waiting for the customer. That's reasonable, but check how generously "waiting" is defined.
- What counts as in scope. Support for some applications, personal devices or third-party systems may be excluded or charged separately.
- Remote versus onsite. A fast remote response is only part of the picture if the problem needs someone to physically attend. See the difference in onsite support.
- Remedies. What actually happens if the provider misses the target? Service credits, if offered at all, are usually modest, but they show whether the commitment is taken seriously.
Measures That Say More Than an SLA
An SLA is a promise. Reporting shows whether it's kept. A good provider should be able to show you, without being asked twice:
- Average and median time to resolution, not just response
- The proportion of tickets fixed at first contact
- Ticket volumes and recurring problems, so root causes get fixed rather than patched repeatedly
- How often SLA targets are met, and what happened when they weren't
- User satisfaction, where it is measured
Questions to Ask Before You Sign
- Is the commitment about response, resolution, or both?
- What are the priority levels, and who assigns them?
- What are the support hours, and what happens outside them?
- What is excluded from the agreement?
- What reporting will we receive, and how often?
- What happens when a target is missed?
If you're weighing up providers, our guide on choosing a managed IT provider covers the wider checklist, and the signs it's time to switch are worth reading if your current SLA reads better than it performs.
Questions worth asking your current provider: What was our average time to resolution last quarter? How many of our tickets were fixed first time? Can we see the figures, rather than a summary?
The Bottom Line
A fast response is nice to have. A reliable fix is what you're paying for. Read the SLA for what it commits to, check it against real reporting, and treat any provider unwilling to share resolution figures as telling you something.