What service level (SLA) means and how VICIdial reports it

Service level is the promise you make to inbound callers about how long they will wait. It is the share of calls you answer inside a target number of seconds, and it is the single most quoted number in any inbound operation. VICIdial reports it from the queue data it already records, so you do not have to reconstruct it by hand.

What service level measures

Service level is usually written as a pair, like 80/20, meaning 80 percent of calls answered within 20 seconds. The formula is the number of calls answered inside your target wait time, divided by the calls offered, expressed as a percentage. The target is yours to set; the math is the same whatever you pick. This commitment is what people mean when they talk about an Service level agreement (SLA) for a queue.

The part that trips people up is the denominator. A call is offered the moment it lands in the queue. If the caller hangs up before an agent picks up, it still counts as offered, which is why your Abandonment rate and your service level move together. You cannot fix service level by hoping abandoned calls disappear; they are baked into the same pool.

sequenceDiagram
  participant C as Caller
  participant Q as Queue
  participant A as Agent
  C->>Q: Call offered
  Q->>Q: Wait timer starts
  Q->>A: Answered within target
  A-->>C: Counts toward SLA
  Q-->>C: Hangup before answer = abandoned

Where VICIdial shows it

The Inbound Service Level Report is the screen built for this. It breaks the basic call statistics for a selected in-group down per day, and the two bottom sections show hold times across 15-minute increments, so you can see exactly how quickly calls were answered and where the long waits clustered. That increment view is what turns a single floor-wide percentage into something you can actually act on.

If you want the underlying counts, the Inbound Daily Report spells them out: total calls offered, total calls answered, total calls abandoned, the abandon percent as abandoned over offered, and the average answer speed, which is how long callers waited before an agent picked up. There is also an Inclusive SLA view that buckets answered calls by wait-second intervals, starting at 5 seconds and widening as the wait grows, so you can read service level at several thresholds at once. That widening grid is handy because it lets you test more than one promise on the same data. You might find you hit 90 percent within 30 seconds even when you miss your 20-second target, which tells you the queue is close rather than broken.

One habit worth building: always read service level per day, not just as a single rolled-up number for the month. A monthly average can look fine while hiding a string of bad afternoons, and the daily and 15-minute breakdowns are exactly where those bad stretches surface. The fix is almost always staffing the right people at the right hour rather than asking everyone to work faster.

What counts as good

The classic industry target is 80/20: answer 80 percent of calls within 20 seconds. It is a default, not a law. A premium support line might aim higher, and a low-stakes queue might happily run looser to save staffing. Watch Average speed of answer (ASA) beside service level, because the two tell a fuller story than either alone, and treat service level as one KPI you balance against cost rather than a number to maximize at any price.

Where it fits

Service level only makes sense alongside your other queue numbers, which is why our guide to VICIdial reports lays out how the inbound screens connect, and the inbound service level report walkthrough goes line by line.

VICIfast provisions a tuned inbound-ready VICIdial box in under 40 seconds. See pricing for plans.