What is average speed of answer?
Average speed of answer (ASA) measures how long, on average, callers wait in the queue before an agent picks up. It’s one of the simplest and most widely reported contact centre metrics, and it maps directly onto something customers notice: how long they were on hold before speaking to someone.
The formula is ASA = total queue time for answered calls / number of answered calls. Most systems start the clock when the call is offered to the agent queue by the automatic call distribution system, after any IVR menu, and stop it when an agent answers.
Why ASA matters
ASA is a direct measure of customer wait time, and wait time shapes the rest of the experience. Customers who wait longer are more likely to hang up, more likely to arrive frustrated, and more likely to rate the interaction poorly even if the agent does a good job.
Long waits also create extra work. Callers who give up often ring back, so a high ASA tends to push up abandonment rate and inflate call volume at the same time. In a call avalanche, ASA can climb from seconds to an hour or more in a very short time.
The limits of an average
ASA is useful, but it is an average, and averages hide the tails. Consider two scenarios with the same 40 second ASA:
- Almost every caller waits between 30 and 50 seconds.
- Most callers are answered instantly, but a few wait 15 minutes.
The second is a much worse experience for the people stuck in the queue. That’s why ASA is normally tracked alongside:
- Service level - the share of calls answered within a threshold.
- Longest wait - the worst experience in the interval.
- Abandonment rate - how many callers gave up, since abandoned calls are usually excluded from ASA.
Reporting ASA by half-hour or hour, rather than daily, also matters. A daily average can look fine while customers face long waits at the busiest times.
How to improve average speed of answer
ASA improves when available capacity better matches demand, or when demand drops. The main levers are:
- Better forecasting and scheduling. Workforce management and Erlang C calculations help match agents to expected volume in each interval.
- Reduce handle time without rushing. Lower average handle time frees agents sooner for the next caller.
- Offer a callback. Queue callback lets customers keep their place without waiting on the line, which smooths peaks and reduces abandonment.
- Automate routine calls. An AI voice assistant can answer simple requests instantly at any volume, leaving agents free for calls that need a person.
- Add flexible capacity. Overflow call handling and surge capacity cover peaks that your own team can’t reasonably staff for.
- Prevent the call. Clear communication, such as proactive outage notifications, answers the question before customers need to ring.