What is an SLA in support?
A service-level agreement, or SLA, is a written promise about how quickly and reliably a support team will answer and resolve customer enquiries. It sets clear targets, the hours they apply during, and a compliance figure to measure against, so "we reply quickly" becomes something you can actually check.
Definition
SLA, in one paragraph
A service-level agreement (SLA) is a documented commitment to a standard of support: how fast a customer will hear back, how fast their issue will be resolved, the hours those promises apply during, and the percentage of tickets that must hit the targets. It can be a formal contract clause with a paying customer, or an internal standard a team holds itself to. Either way, it makes good service measurable instead of a matter of opinion.
The word "agreement" matters. An SLA is most useful when both sides understand it: the customer knows what to expect, and the team knows what it has signed up to deliver. The clearest agreements spell out four things, which the rest of this page walks through: the targets, the priority levels, the hours, and the compliance percentage.
| SLA | Service-level agreement: the overall promise about response and resolution standards. |
|---|---|
| First-response time | How long a customer waits for the first human reply after getting in touch. |
| Resolution time | How long the whole enquiry takes to close from first contact to done. |
| Priority (P1 to P4) | How urgent a ticket is, which decides how tight its SLA targets are. |
| Compliance | The percentage of tickets that met their target in a given period, for example 95 per cent. |
| Breach | A ticket that missed its SLA target, so it counts against your compliance figure. |
The two clocks
First response vs resolution
Almost every support SLA tracks two separate timers, because a fast first reply and a fast fix are not the same thing.
First-response time
First-response time is the gap between a customer reaching out and the first genuine human reply landing in their inbox. It is the figure customers feel most, because it answers the anxious question "did anyone even see my message?" An automated "thanks, we have got this" acknowledgement is not usually counted as a first response, because it does not move the issue forward. A first-response target of one business hour, for instance, says a real person will be in touch within an hour of the team being open.
Resolution time
Resolution time is the gap between first contact and the moment the enquiry is genuinely closed. It is the figure that reflects whether problems actually get sorted, not just acknowledged. Resolution is slower and lumpier than first response, because some issues need a refund to clear, a supplier to reply, or an engineer to ship a fix. That is exactly why the two are measured apart: a team can be brilliant at fast first replies and still have a backlog of half-finished tickets, and you only see that if you watch both clocks.
A common refinement is to pause the resolution clock while you are waiting on the customer. If you have asked for an order number and the customer goes quiet for two days, most teams do not want those two days counted against their resolution target, because the delay was not theirs. Whether your SLA pauses on "waiting for customer" is a choice worth making explicitly, since it changes the numbers a lot.
Priority levels
What P1, P2, P3 and P4 mean
Priority bands let one SLA promise faster targets for urgent issues and relaxed ones for routine requests. The exact targets are yours to set; the bands give them structure.
P1, critical
A business-stopping or widespread failure: the storefront is down, payments are failing, no one can log in. P1 carries the tightest targets, often measured in minutes for first response, and usually triggers an alert rather than a quiet queue.
P2, high
A serious problem that is not total: an important feature is broken, or one key customer is badly affected. Targets are tight but a notch behind P1, with a fast first response and a same-day resolution aim.
P3, normal
The everyday request: a question, a small change, a how-do-I. P3 is where most tickets live, so its targets set the tone for your service, typically a few business hours to respond and a day or two to resolve.
P4, low
Informational or non-urgent: a feature idea, a "no rush" query, a nice-to-know. P4 carries the most relaxed targets so it never crowds out work that genuinely matters more.
When the clock runs
Business hours change every number
Two teams with the identical "one hour first response" target can report wildly different figures, purely because of which hours their clock runs.
Most small support teams run their SLAs against business hours: the clock counts only during working hours, pauses overnight, and pauses at weekends and on public holidays. So a message that arrives at 17:30 on a Friday, against a 09:00 to 17:30 Monday-to-Friday SLA, is not "sixty-three hours late" by Monday morning. The clock parks at close of business on Friday and restarts at nine on Monday, which is the fair way to measure a team that is not staffed at the weekend.
A 24/7 clock, by contrast, counts every minute of every day. It only makes sense if you actually have round-the-clock cover, which most small businesses do not. Promising 24/7 targets you cannot staff is the fastest way to breach your own SLA, so be honest about the hours you can keep.
Public holidays deserve a special mention. A UK-focused SLA should pause on England and Wales bank holidays (and the relevant dates for Scotland and Northern Ireland if you serve those nations), or your compliance figure will quietly sag every time a bank holiday lands midweek. The rule of thumb: the hours in the agreement should match the hours your people are genuinely at their desks.
Measuring success
Why 95 per cent, not 100
An SLA needs a target percentage, the share of tickets that must meet their goal. Most teams land near 90 to 95 per cent, and there are good reasons not to chase a perfect score.
95 per cent compliance means at least nineteen of every twenty tickets hit their target in the period you are reporting on. The remaining five per cent is a deliberate allowance for genuine exceptions: the message that arrived during a system outage, the case that needed a third party, the odd ticket that slipped through on a chaotic day. That allowance keeps the figure honest.
Aiming for 100 per cent sounds admirable but tends to backfire. When every single ticket must be perfect, people start gaming the metric: closing tickets early to stop the clock, or reclassifying a breach as something that does not count. A realistic target, paired with a habit of reviewing the misses, drives better behaviour than an impossible one. The point of the percentage is to spot a slipping trend before customers feel it, not to produce a flawless scoreboard.
A worked example
Say a team sets a P3 first-response target of two business hours and a resolution target of one business day, against 09:00 to 17:30 hours, Monday to Friday, with a 95 per cent compliance goal. In a month they handle 400 tickets. If 388 of them got a first reply within two working hours, that is 97 per cent compliance, comfortably above target. If only 360 did, that is 90 per cent, below target, and the dozen worst breaches are worth a look to find out what went wrong, perhaps a Monday morning pile-up or a holiday gap in cover.
SLA versus SLO and KPI
An SLA is the promise you make to others. An SLO (service-level objective) is the internal target you set yourself, often a little stricter so you have headroom before you breach the promise. A KPI is any key performance indicator you track, of which SLA compliance is one. Useful to keep straight, but for a small support team the SLA is the figure that matters most.
Where Solves fits
How Solves measures SLAs
Solves is an email-first support tool, and SLA timers are one of its core features. You set first-response and resolution targets per brand or for the organisation, and a breaching view surfaces overdue tickets so the work due soonest is easy to find. Tickets carry a priority you can use to apply tighter or looser targets, and the reports show how you are tracking against your compliance goal.
The aim is to make a sensible SLA practical for a small team, without a contact-centre suite behind it. Solves does not run phone lines or live chat, so the targets it tracks are the ones a support inbox can genuinely keep. If you want to set targets that match the hours you actually work, the SLA timers feature and the guide below are the place to start.
FAQ
Common questions about SLAs
What is an SLA in customer support?
An SLA, or service-level agreement, is a written promise about how quickly and reliably a support team will respond to and resolve customer enquiries. It usually sets a first-response target, a resolution target, the hours those targets apply during, and a compliance percentage such as 95 per cent. It turns a vague intention to "reply quickly" into a measurable commitment everyone can check against.
What is the difference between first response and resolution time?
First-response time measures how long a customer waits for the first human reply after they get in touch. Resolution time measures how long the whole enquiry takes to close from start to finish. A first reply can be fast even when the underlying fix is slow, so most teams set separate targets for each and report on both.
What do P1, P2, P3 and P4 priority levels mean?
They are priority bands that decide how urgently a ticket is handled. P1 is a critical, widespread or business-stopping issue with the tightest targets; P2 is high impact but not total; P3 is a normal request with standard targets; and P4 is low priority or informational, with the most relaxed targets. Tighter priorities get faster SLA clocks.
Does an SLA clock run overnight and at weekends?
Only if you say so. Most small teams run SLAs against business hours, so the clock pauses outside working hours, at weekends and on public holidays, and resumes when the team is back. A few teams run 24/7 clocks for critical issues. The agreement should state clearly which hours apply, because it changes every figure you report.
What does 95 per cent SLA compliance mean?
It means at least 95 per cent of tickets in a given period met their target, leaving a small allowance for genuine exceptions. Most teams pick a realistic percentage like 90 or 95 per cent rather than 100 per cent, because aiming for a perfect score tends to push people to game the figures rather than help customers.
Turn an SLA into something you can keep
See how first-response and resolution SLA timers work, with a breaching view for overdue tickets, or read the step-by-step guide to setting your targets.