A Guide to Average Speed of Answer (ASA) in the Call Center
Every second a caller spends in your queue shapes how they feel about the conversation before it starts. Average speed of answer (ASA) is the metric that tells you how long that wait really is.
Key takeaways
ASA is one of the most-watched metrics in any contact center. By reading this guide you can:
Define ASA precisely and avoid the most common mistake: confusing it with hold time.
Calculate ASA correctly, by interval as well as by day.
Benchmark your performance against the ~28-second global average and the 80/20 service level standard.
Lower ASA through better forecasting, scheduling, adherence, routing, self-service, AI assistance, and coaching.
What is ASA in a call center?
ASA stands for average speed of answer. It is a call center metric that measures the average time callers wait in the queue before a live agent answers, over a set period. ASA covers answered calls only and does not include the time a caller spends navigating the IVR. It is one of the clearest signals of whether a contact center has the right number of agents, with the right skills, available at the right time.
Because it is expressed in seconds and updated in real time, ASA is often the first number supervisors check on a wallboard. A rising ASA is an early warning that demand has outpaced staffing, and it usually shows up before abandonment rates and satisfaction scores start to move.
What does ASA include and exclude?
Different systems treat wait time slightly differently, so it helps to be explicit about scope. Standard ASA measures the time from the moment a call enters the agent queue to the moment an agent picks up.
HTML
| Included in ASA | Not included in ASA | |
|---|---|---|
| Time | Time waiting in the queue after the IVR; ring time until the agent answers | Time spent in the IVR or with a virtual agent before transfer to the queue |
| Calls | Answered calls only | Abandoned calls (tracked as abandon rate) |
| Hold | — | Hold time after the agent answers (tracked as hold time) |
How do you calculate average speed of answer?
To calculate ASA, divide the total time callers waited for answered calls by the number of calls answered in the same period.
Example: Your contact center answered 1,200 calls yesterday. Together, those callers waited 36,000 seconds in the queue before reaching an agent. 36,000 ÷ 1,200 = an ASA of 30 seconds.
How to get an accurate ASA reading
- Measure by interval, not just by day. A healthy daily ASA can hide a Monday-morning or lunchtime spike where callers wait several minutes. Track ASA in 15- or 30-minute intervals, the same intervals your staffing model uses. Workforce planners use Erlang formulas to translate interval ASA targets into required headcount.
- Always read ASA next to abandon rate. Abandoned calls drop out of the ASA calculation, so a center that loses many long-waiting callers can report a deceptively good ASA.
- Keep the definition consistent. Decide whether ring time is included and where the clock starts after the IVR, then use the same rules in every report.
What is the industry standard for average speed of answer?
There is no single rule, but two benchmarks are widely used. The global average ASA is roughly 28 seconds, and the most common service level target is 80/20: 80% of calls answered within 20 seconds. For more, see Verint’s roundup of popular call center benchmarks.
| Benchmark | Figure | What it means |
|---|---|---|
| Global average ASA | ~28 seconds | A typical reference point for voice queues |
| Standard service level | 80/20 | 80% of calls answered within 20 seconds |
| Alternative service level | 95/120 | 95% of calls answered within 2 minutes (proposed by SQM Group) |
| Average time to abandon | 2 min 36 sec | When callers typically hang up (SQM Group) |
Meeting 80/20 consistently is harder than it sounds. SQM Group research found only 16% of call centers achieve it consistently, and that calls answered within two minutes of leaving the IVR have no negative impact on customer satisfaction. That is why SQM suggests 95/120 as a more realistic standard for many operations.
The right target for your team depends on your industry, your customers’ expectations, and how complex your calls are. Lines where speed is critical, such as urgent healthcare or fraud reporting, typically set tighter targets, while complex technical support may accept a longer wait if first contact resolution is high. Whatever you choose, set it as a service level goal as well as an ASA goal, because the percentage shows how many callers had a long wait, which an average can hide.
Is a lower ASA always better?
Not necessarily. Driving ASA toward zero means staffing for the peak of every interval, which raises cost per contact and leaves agents idle between calls. The goal is the ASA that keeps abandonment and customer satisfaction in range at the lowest sustainable cost. Finding that balance, interval by interval, is exactly what call center forecasting and scheduling software is designed to do.
Why is ASA important?
ASA matters because it sits at the start of every phone interaction. When it drifts upward, the effects spread quickly across the operation.
Higher abandonment
The longer callers wait, the more of them hang up. Each abandoned call is a customer who didn’t get help, and many of them call back later, adding to tomorrow’s queue.
Lower customer satisfaction
Customers who have already waited arrive frustrated. Agents spend the first minute of the call repairing the experience rather than solving the problem, and CSAT and NPS scores reflect it.
Agent stress and attrition
A constantly full queue puts pressure on agents to rush, skip breaks, and cut conversations short. Over time, that pressure drives burnout and turnover, which in turn makes the staffing gap worse.
Higher operating costs
Repeat calls from customers who abandoned, escalations from frustrated callers, and the overtime needed to catch up all add cost. Keeping ASA in range is usually cheaper than recovering from a backlog.
How does ASA relate to other call center metrics?
ASA is most useful when read alongside the metrics that cause it and the metrics it affects.
| Metric | What it measures | How it relates to ASA |
|---|---|---|
| Service level | % of calls answered within a target time, such as 80/20 | Shows the spread of wait times that ASA averages into one number |
| Abandon rate | % of callers who hang up before an agent answers | Rises as ASA rises; abandoned calls are excluded from ASA |
| Average handle time | Talk time + hold time + after-call work per call | Longer handle times keep agents busy and push ASA up |
| Hold time | Time a caller is placed on hold after the agent answers | A separate metric that is often confused with ASA |
| First call resolution | % of issues resolved on the first contact | Low FCR creates repeat calls that inflate queue volume |
| Schedule adherence | Whether agents are available when scheduled | Poor adherence leaves queues short-staffed at peak times |
7 ways to lower ASA in your contact center
Most ASA problems come down to one of two things: not enough agents available when calls arrive, or calls taking longer than they need to. These seven tactics address both.
1. Forecast call volume more accurately
Every staffing decision starts with a forecast. Build yours from interval-level history, and account for seasonality, marketing campaigns, billing cycles, and outages. Even small forecasting errors at the interval level translate directly into queue time. Our guide to call center forecasting methods covers the main techniques.
2. Schedule to demand, not to averages
A schedule built around daily averages will be overstaffed in quiet hours and short in busy ones. Use workforce management software to match shifts, breaks, and training to the forecast interval by interval, and to re-optimize quickly when volume shifts.
3. Track schedule adherence in real time
A perfect schedule does not help if agents are not where it says they will be. Real-time adherence tools alert supervisors the moment staffing drifts from plan, so they can adjust breaks or pull in help before the queue builds.
4. Route calls to the right agent the first time
Skills-based and intelligent call routing send each caller to an agent who can resolve the issue, which cuts transfers and repeat calls that would otherwise re-enter the queue.
5. Offer self-service and callbacks for simple requests
Not every contact needs a live agent. An intelligent virtual assistant can resolve routine requests such as order status, balance checks, and password resets, keeping agents free for complex calls. For calls that do need an agent, a queue callback option lets customers hang up without losing their place, which lowers abandonment and smooths demand across the interval.
6. Reduce handle time with AI assistance
Every second removed from average handle time returns capacity to the queue. AI agent assist surfaces answers and next steps during the call, and automation such as the Interaction Wrap-Up Bot reduces after-call work so agents become available for the next caller sooner.
7. Coach continuously
Well-trained agents resolve issues faster and more completely. Rather than relying only on periodic classroom sessions, use real-time coaching to give agents targeted guidance based on their own interactions, so improvements show up in handle time and first call resolution quickly.
Customer example: After moving to Verint Workforce Management, Catalina Express cut its average speed of answer from one minute to 30 seconds and raised its service level from 69% to 83%. Read the Catalina Express case study.
How does Verint help contact centers improve ASA?
Lowering ASA sustainably takes more than adding headcount. Verint connects the forecasting, scheduling, and real-time management that puts the right agents in place with the AI that shortens each interaction. Verint Workforce Management forecasts demand and builds optimized schedules, real-time adherence keeps the plan on track, and AI agents across the Verint CX Automation Platform handle routine requests and cut after-call work, all to help you hit your ASA and service level targets without overstaffing.
Want to see how Verint can help you answer customers faster? Get a demo today.
Frequently asked questions about ASA
ASA stands for average speed of answer. It measures the average time callers wait in the queue before a live agent picks up, excluding time spent in the IVR.
Divide the total wait time for answered calls by the total number of answered calls. For example, 36,000 seconds of wait time across 1,200 answered calls equals an ASA of 30 seconds.
Around 28 seconds is a commonly cited global benchmark, and many contact centers target a service level of 80% of calls answered within 20 seconds. The right goal depends on your industry, your customers, and how quickly they tend to abandon.
ASA is the average wait in seconds. Service level is the percentage of calls answered within a target threshold, such as 80% in 20 seconds. Service level shows how many callers had a long wait, which an average can hide.
No. Standard ASA counts only answered calls and starts timing when the caller enters the agent queue, so IVR time and abandoned calls are excluded. Track abandon rate alongside ASA to get the full picture.
AI virtual agents resolve simple requests without a queue, AI-driven forecasting improves staffing accuracy, and agent-assist tools shorten handle time so agents become available sooner. All three reduce the time callers wait.