Skip to main content
TurnellaBeta
WFM reference

Contact centre metrics — the 10 KPIs every WFM team must track

Every contact centre runs on a handful of metrics. Some predict tomorrow's staffing. Some diagnose today's failure. Some justify next year's budget. Here are the ten that matter, explained plainly, with benchmarks and direct links to the tools that compute them.

01

Service Level (SL)

Contacts answered within threshold ÷ total contacts offered × 100

The percentage of inbound contacts answered within a target time threshold (e.g. 80% in 20 seconds). The primary planning metric: every staffing calculation starts here.

Benchmark

Voice: 80% in 20s · Chat: 80% in 30s · Email: 80% in 4h

WFM impact

Sets the Erlang C agent requirement. Raising SL from 80% to 90% typically needs 10–20% more agents.

02

Average Handle Time (AHT)

(Talk time + Hold time + After-Call Work) ÷ total contacts handled

The average time an agent spends on a single contact, including talk, hold, and post-contact wrap-up work. AHT is the most powerful driver of staffing requirement: a 10% reduction in AHT is roughly equivalent to adding 10% more agents.

Benchmark

Voice: 3–8 min · Chat: 8–14 min · Email: varies by complexity

WFM impact

Direct input to Erlang C (traffic intensity = volume × AHT ÷ interval). See exactly how AHT changes affect agent count.

03

Occupancy Rate

Traffic intensity (Erlangs) ÷ seated agents × 100

The percentage of logged-in agent time actually spent handling contacts. Occupancy is the flip side of service level: the same agent count that achieves 80% SL also determines occupancy. High occupancy (>90%) causes queue build-up and erodes service level during any volume spike.

Benchmark

Voice: 70–85% · Chat: 80–88% · Email: 88–93%

WFM impact

Occupancy above 88–90% leaves no buffer for variance. Even minor volume peaks tip into failure. Use it to diagnose over-stretched teams.

04

Average Speed of Answer (ASA)

Total wait time for queued contacts ÷ total contacts handled

The mean time a customer waits before being connected to an agent. ASA is a useful secondary metric but inferior to service level for planning: a good ASA can hide a long tail of very long waits.

Benchmark

Voice: 20–40s · Chat: 15–30s

WFM impact

Use as a diagnostic alongside service level, not as the primary KPI. A rising ASA during a period of stable SL can indicate an AHT change or adherence problem.

05

Abandonment Rate

Contacts abandoned ÷ total contacts offered × 100

The percentage of inbound contacts where the customer hangs up or disconnects before being answered. Rising abandonment is a leading indicator of service level failure: customers are leaving the queue rather than waiting.

Benchmark

Voice: <5% · Chat: <3%

WFM impact

Abandonment reduces the apparent queue length (reducing the visible problem) while losing potential revenue and CSAT. Exclude immediate abandons (<5s) to get a meaningful rate.

06

First Contact Resolution (FCR)

Contacts fully resolved on first contact ÷ total contacts × 100

The percentage of contacts fully resolved without a repeat contact. FCR is the strongest predictor of customer satisfaction and cost: every repeat contact is a second full AHT you didn't need.

Benchmark

Voice: 70–80% · Chat: 60–70%

WFM impact

A 5% improvement in FCR reduces total contact volume by ~5%, directly reducing your staffing requirement. FCR is measured via CRM case tracking, follow-up calls within 7 days, or post-contact surveys.

07

Customer Satisfaction (CSAT)

Positive responses ÷ total survey responses × 100

A post-contact survey measure of customer satisfaction, typically scored on a 1–5 or 1–10 scale. CSAT correlates strongly with wait time, FCR, and agent knowledge, all of which are within WFM influence.

Benchmark

Voice: 75–85% · Industry-dependent

WFM impact

CSAT drops sharply when service level falls below 70%. Overstaffing rarely improves CSAT beyond a threshold; FCR and agent quality matter more than speed after a certain point.

08

Schedule Adherence

Time in scheduled state ÷ total scheduled time × 100

The percentage of time agents are doing what the schedule requires: logged in and ready when rostered, on break when scheduled, and so on. Schedule adherence is the link between your Erlang C plan and what actually happens on the floor.

Benchmark

Target: 90–95% · Warning below 85%

WFM impact

At 30% shrinkage, a 5% adherence shortfall is equivalent to losing an additional 5% of your seated agents, pushing occupancy up and service level down. Real-time adherence tracking is the core of intraday WFM.

09

Cost Per Contact (CPC)

Total operating cost ÷ total contacts handled

The total cost of operating the contact centre divided by the number of contacts handled. CPC is the primary financial efficiency metric: it links staffing decisions to budget impact.

Benchmark

Voice: £3–8 · Chat: £2–5 · Email: £2–4 (UK estimates, 2024)

WFM impact

CPC rises when occupancy is too low (overstaffed) or when AHT is too high. A 10% reduction in AHT or a 10% improvement in adherence typically reduces CPC by 5–10%. Use CPC to justify WFM investment.

10

FTE (Full-Time Equivalent)

Total rostered hours ÷ standard hours per agent per period

A normalisation unit that converts rostered hours to an equivalent number of full-time agents. FTE is the standard unit for WFM budget modelling, headcount reporting, and capacity planning.

Benchmark

Planning unit: 1 FTE = 37.5–40h/week in most markets

WFM impact

FTE = (Erlang C seated requirement ÷ occupancy ÷ (1 − shrinkage)) expressed per interval, then rolled up to weekly. The chain from traffic to FTE is: volume → Erlang C → seated → shrinkage → scheduled → FTE.

How the metrics connect

These metrics aren't independent. They form a chain, and changes upstream ripple downstream:

Volume × AHT → Traffic intensity (Erlangs)↓ Erlang CTraffic intensity + Agents → Service Level + Occupancy↓ Shrinkage formulaSeated agents ÷ (1 − Shrinkage) → Scheduled headcount → FTE↓ AdherenceScheduled headcount × Adherence % → Effective seat count↓ ImpactsEffective seats → actual SL → Abandonment rate → CSAT

Frequently asked questions

What is the most important metric in a contact centre?

Service level is the primary operational metric for inbound contact centres: it directly measures whether customers are being served quickly enough. All workforce planning decisions flow from the service level target. However, the most important metric depends on the operation: for outbound, it might be conversion rate or contact rate; for back-office, it might be throughput and backlog age.

What is a good AHT for a contact centre?

Average Handle Time (AHT) depends heavily on the type of contact. Simple transactional calls (account balance, address change) average 2–4 minutes. Customer service calls average 4–8 minutes. Technical support averages 8–15 minutes. Longer is not always bad: forcing agents to rush increases repeat contacts. AHT is most useful compared against your own operation's baseline, not an external benchmark.

What is a good schedule adherence rate?

Most contact centres target 90–95% schedule adherence. Below 85% typically signals a systemic problem: agents arriving late, extended breaks, or WFM tools not being used effectively. Above 97% can mean the schedule is too rigid and agents feel micromanaged. Adherence should be measured by the minute, not just by whether agents showed up for their shift.

What is the difference between occupancy and utilisation?

Occupancy is the percentage of seated (logged-in) agent time spent actively handling contacts. Utilisation is the percentage of all rostered agent time spent on productive work, including training, meetings, coaching, and back-office tasks. Occupancy measures on-phone efficiency; utilisation measures total workforce productivity. A well-run contact centre targets 70–85% occupancy and 85–92% utilisation.

Model your metrics in Turnella

Set your volume, AHT, and service level target. Get seated agents, occupancy, and scheduled headcount in one step.

Related guides