Call Center KPIs

The core call center KPIs by category

Most operations track 15 to 25 metrics but run on fewer than ten. Grouping them makes trade-offs visible.

Accessibility: can customers reach you?

• [Service level](/glossary/service-level/): percentage of contacts answered within a target time, for example 80% in 20 seconds.
• Average speed of answer (ASA): average wait before an agent answers.
• Abandonment rate: percentage of callers who hang up before reaching an agent.

Efficiency: how well do you use agent time?

• [Average handle time](/glossary/average-handle-time/): talk plus hold plus after-call work, per contact.
• [Occupancy](/glossary/occupancy/): share of logged-in time agents spend handling contacts.
• Schedule adherence: whether agents are doing the right activity at the scheduled time.
• Shrinkage: share of paid time not available for contacts, from breaks, training, meetings and absence.
• Cost per contact.

Quality and resolution: did you get it right?

• [First call resolution](/glossary/first-call-resolution/): share of issues resolved on the first contact.
• QA score: evaluation against the quality scorecard.
• Transfer rate and repeat contact rate.

Customer outcomes: how did it feel?

• Customer satisfaction (CSAT), Net Promoter Score (NPS) and customer effort score (CES).

Workforce: can you sustain it?

• Agent attrition, absenteeism, training attrition, speed to proficiency and agent engagement or eNPS.

Formulas and typical targets

Targets depend on industry, channel and what the business is willing to pay for. Commonly used figures:

• Service level = contacts answered within threshold ÷ total contacts offered (definitions vary on abandons). Voice targets of 80/20 or 80/30 are common; chat is often 80% in 60 seconds; email is usually measured in hours.
• ASA = total wait time of answered calls ÷ answered calls. Often under 30 seconds for customer service lines.
• Abandonment rate = abandoned calls ÷ calls offered. Many centers aim for under 5% to 8%, often excluding calls that hang up in the first few seconds.
• AHT = (talk + hold + after-call work) ÷ contacts handled. Ranges from about 4 minutes for simple service to 10 or more for technical support.
• Occupancy = handle time ÷ available time. 75% to 85% is common; sustained levels near 90% or above tend to burn agents out.
• FCR = contacts resolved on first contact ÷ total contacts. Many centers land between 65% and 80%.
• Schedule adherence = time in adherence ÷ scheduled time. Targets of 90% to 95% are typical.
• Agent attrition = leavers ÷ average headcount, annualized. 30% to 45% is common for in-house frontline teams.

Worked example: reading KPIs together

A 180-seat center at a regional bank reviews a bad week:

• Service level fell from 80/30 to 64/30.
• AHT rose from 360 to 410 seconds.
• Occupancy hit 92%.
• FCR dropped from 74% to 68%.
• A new class of 22 agents was released from nesting the week before.

Looking at service level alone, the obvious response is "answer faster", which usually means pushing AHT down. The full picture tells a different story. The new class has higher handle times, lower resolution and generates repeat calls. Repeat calls inflate volume, which drives occupancy up and service level down.

The right fixes are more floor support for the new agents, routing simpler call types to them for another two weeks and a workforce plan that counts them as partial capacity. Pressuring the whole floor on AHT would have pushed FCR down further and made the volume problem worse.

Metrics that pull against each other

Almost every call center KPI has a counterweight, and managing one in isolation damages another.

• AHT vs FCR and quality. Rushing calls cuts handle time and creates repeat calls.
• Service level vs cost. Each extra point of service level needs more staff, and the cost climbs steeply near the top.
• Occupancy vs attrition. Very high occupancy looks efficient and wears agents out.
• Adherence vs customer outcomes. Strict adherence policing can make agents rush the last call before a break.

A balanced scorecard pairs each efficiency metric with a quality or customer metric. Agent-level scorecards should do the same: an agent with low AHT and a weak FCR is not your best performer.

The people metrics most centers underweight

Operational dashboards are usually crowded with service and efficiency metrics, while people metrics sit in a monthly HR report. That is backwards for a high-volume operation, because workforce health drives every other number a few weeks later.

At minimum, put these on the weekly operations review: attrition by tenure band, absenteeism, training and nesting pass rates, speed to proficiency per class, and open seats against the hiring plan. A center that misses its hiring plan by 15 agents in March will miss service level in April no matter how well the floor is run. Tie these to the call center attrition and recruiting reports so operations and HR look at the same numbers.

Common mistakes with call center KPIs

• Too many metrics on the agent scorecard. Agents can focus on three to five. Ten metrics means none of them get attention.
• Averages hiding intervals. A daily service level of 80% can mean 95% in the morning and 50% at lunch. Review by interval.
• Inconsistent definitions. Abandonment and service level formulas vary. Document yours, especially when a BPO reports to a client under an SLA.
• Measuring agents on things they cannot control. Service level is a staffing and scheduling outcome. Holding individual agents accountable for it breeds frustration.
• Reporting without acting. Every KPI on the dashboard should have an owner and a defined response when it moves out of range.