Average speed of answer (ASA) is a contact center and help desk metric for how long, on average, answered calls wait in queue before a live agent picks up. It is usually shown in seconds and calculated per queue and time interval. ASA is one of the simplest measures of responsiveness, which is why it often appears in staffing plans, dashboards and outsourcing contracts, but an average can hide the callers who waited longest, so it is best read alongside service level and abandonment rate.
At a glance
- ASA is total queue wait time for answered calls divided by the number of answered calls.
- Abandoned calls are normally excluded, so ASA should be read with abandonment rate.
- Definitions vary, such as whether IVR time or ringing time counts; check how your platform measures it.
- ASA is driven mainly by staffing against demand, plus routing and how long calls take.
- Targets differ by queue and business; there is no single “good” number.
What problem it solves
Customers judge a contact center first by how long they wait. Long waits lead to hang-ups, repeat calls and complaints, and agents start many conversations with a frustrated caller. Managers need a simple, consistent way to see whether queues are keeping up with demand.
ASA gives that view. A rising ASA is often the first sign that staffing has fallen behind volume, that calls are taking longer, or that a routing change has sent calls to the wrong place. Because it is easy to understand, ASA is also a common target in contact center service level agreements with outsourcers and in internal service commitments.
How it works
Measurement. The platform timestamps when a call enters a queue and when an agent answers. ASA is the sum of those waits for answered calls divided by the number of answered calls. For example, if 100 calls were answered and together they waited 3,000 seconds, ASA is 30 seconds. Some platforms start the clock when the call reaches the system rather than the queue, and some count the agent’s ringing time; those choices change the number.
What is excluded. Calls abandoned before an agent answers don’t contribute to ASA, so a queue where many callers give up can show a deceptively good average. That is why ASA is normally reported next to abandonment rate.
What drives it. The main lever is the number of agents available compared with incoming volume. Average handle time (AHT) matters too: longer calls keep agents busy and lengthen waits for everyone else. Workforce management (WFM) tools forecast volume and handle time by interval and schedule staff to hit a target ASA or service level. Routing design, the call queue set-up and self-service options also shape it.
Reporting. Real-time dashboards show current wait times so supervisors can move agents between queues. Historical reports should break ASA down by queue and interval, since a good daily average can hide a peak hour where waits were long.
When it matters for buyers
- Contracting with an outsourcer. If ASA or service level is a contract target, define the formula, the measurement interval, which calls are excluded and what happens when the target is missed.
- Choosing a contact center platform. Check how the platform calculates ASA, whether you can change the definition, and whether callbacks and transfers are reported clearly.
- Planning staffing and budget. Lower ASA targets usually need more staff and lower agent occupancy; decide what the wait is worth to the business.
- Reducing wait without adding staff. Callbacks through virtual hold, better routing and self-service can help, but they change what ASA measures, so redefine reporting when you add them.
Our contact center as a service overview covers platforms with built-in queue reporting and workforce tools.
Questions to ask vendors
- Exactly how do you calculate ASA: when does the clock start and stop, and which calls are excluded?
- Can we report ASA by queue, skill, channel and 15- or 30-minute interval?
- How are callbacks, transfers and overflow calls counted?
- Do you report service level and abandonment rate with the same definitions alongside ASA?
- For outsourcers: what ASA or service level do you commit to, how is it measured, and what credits apply if you miss it?
- What workforce management tools are included to forecast and staff to the target?
How it differs from service level
ASA and contact center service level both measure answer speed but answer different questions. ASA is the average wait for answered calls. Service level is the share of contacts answered within a set threshold, such as 80% within 20 seconds. A queue can have a reasonable ASA while a minority of callers wait far too long, which service level reveals and ASA hides. Service level, ASA and abandonment rate together give a fuller picture than any one alone. ASA also differs from first contact resolution (FCR), which measures whether the issue was solved, not how quickly the call was answered.
Problems this helps with
- Customers complain about long hold times
- Don't know whether our contact center is staffed for the call volume
- Outsourcer reports a good average but customers still wait too long
- Need a wait-time target to put in a contact center contract
