Workforce Planning Metrics for Call Centers — What to Track, How to Calculate It, and What to Do With the Results

Workforce planning in a call center answers one question: how many people do you need, when do you need them, and what does it cost? Every other metric exists to inform that answer.

Most call centers track operational metrics — AHT, service level, occupancy — as performance indicators. Workforce planning metrics are different. They are input metrics: the numbers you need before you build the schedule, before you approve a hiring request, before you set a budget. If your planning inputs are wrong, your staffing will be wrong regardless of how well you manage day-to-day operations.

This post covers the metrics that feed workforce planning decisions — what each one measures, how to calculate it, what a healthy range looks like, and what planning action to take when the number is off.

Forecast accuracy

Forecast accuracy measures how close your predicted contact volume was to what actually happened. It is the single most important workforce planning metric because every downstream staffing decision is built on the forecast. If the forecast is wrong, the staffing plan is wrong.

Calculation:

Forecast accuracy (%) = 100 − |((Actual volume − Forecasted volume) / Forecasted volume) × 100|

Example: You forecast 5,000 calls for the week. Actual volume was 5,400.

Forecast accuracy = 100 − |((5,400 − 5,000) / 5,000) × 100| = 100 − 8 = 92%

What to check when accuracy is low:

• Bias direction. Are you consistently forecasting too high (overstaffing, idle agents) or too low (understaffing, overtime)? Consistent over-forecasting wastes money. Consistent under-forecasting burns out agents and misses service levels.
• Interval vs. daily accuracy. Your daily forecast may be 95% accurate, but if you predicted the volume would come in the morning and it actually came in the afternoon, interval-level accuracy is poor — and the schedule will be wrong even though the daily total was right.
• Event impact. Marketing campaigns, outages, billing cycles, and seasonal patterns create volume spikes. If these are not in your forecast model, accuracy will suffer on those days.

Required staff calculation

The required staff number converts your volume forecast into the number of agents who need to be on the phones during each interval. This is the bridge between "how many calls will come" and "how many people do I schedule."

Inputs needed:

Step 1 — Calculate workload:

Workload (in hours) = Call volume × AHT (in seconds) / 3,600

Example: 120 calls × 360 seconds / 3,600 = 12 hours of work in a 30-minute interval

Step 2 — Calculate base staff needed:

For a simple estimation: Base staff = Workload / Interval length (in hours) = 12 / 0.5 = 24 agents as a minimum

This is the minimum to handle the work if agents were at 100% occupancy. In practice, you need more agents to meet the service level target — the exact number depends on the Erlang C formula or your WFM tool's calculation. For an 80/20 service level with the workload above, you would typically need 27–29 agents.

Step 3 — Adjust for shrinkage:

Scheduled staff = Base staff / (1 − Shrinkage rate)

Example: 28 / (1 − 0.30) = 28 / 0.70 = 40 agents scheduled

This means you need 40 agents on the schedule to have 28 actually on the phones after accounting for breaks, absences, training, and other non-productive time.

Shrinkage

Shrinkage is the percentage of scheduled time that agents are paid for but not available to handle contacts. It is the gap between how many agents you schedule and how many are actually on the phones at any given moment.

Calculation:

Shrinkage (%) = (Total unavailable hours / Total scheduled hours) × 100

Shrinkage components:

The most common planning mistake: Using a shrinkage estimate that is too low. Many operations estimate 15–20% shrinkage because they only count breaks and PTO. They miss training, meetings, unplanned absences, and late arrivals — and they are understaffed on every shift as a result.

How to fix it: Calculate actual shrinkage from time tracking data over a 4–8 week period. Do not estimate — measure. If your actual shrinkage is 32% and you have been planning at 20%, your schedule has been short by roughly 15% of required staff on every interval.

Schedule efficiency

Schedule efficiency measures how well the schedule matches the staff-to-requirement ratio across intervals. A perfectly efficient schedule would have exactly the required number of agents in every interval — no overstaffing, no understaffing.

Calculation:

Schedule efficiency (%) = (Required staff hours / Scheduled staff hours) × 100

Example: Over a week, your intervals require a total of 1,600 agent-hours. Your schedule provides 1,840 agent-hours. Schedule efficiency = 1,600 / 1,840 = 87%. The remaining 13% is excess coverage — agents scheduled during intervals where they are not needed.

What causes low schedule efficiency:

• All agents start at the same time instead of staggered starts
• Equal numbers assigned to each shift regardless of volume patterns
• No mid-day or split shifts to cover the lunch period or afternoon peak
• Days-off patterns that pull too many agents off on the same day

Attrition rate and its planning impact

Attrition is a workforce planning metric because every departure creates a future staffing gap that must be filled by hiring, training, and ramping a replacement. The planning question is not just "what is our attrition rate" but "how many people do we need to hire per month to maintain headcount?"

Calculation:

Monthly attrition rate = Departures in month / Average headcount during month × 100

Annual attrition rate = Monthly rate × 12 (or sum of 12 monthly rates)

Planning impact — replacement hiring calculation:

If you wait until agents leave to start recruiting, you will be 7 weeks behind — running short-staffed and paying overtime to cover the gap during the entire recruiting-training-ramp cycle. Workforce planning uses the attrition rate to build a continuous hiring pipeline that starts before the departures happen.

Attrition by tenure — where to focus:

Occupancy

Occupancy measures the percentage of time agents spend handling contacts or in after-call work versus waiting for the next contact. It indicates whether agents are appropriately loaded.

Calculation:

Occupancy (%) = (Total handle time / Total logged-in time) × 100

Planning implication: If occupancy is chronically above 85%, you do not have a performance problem — you have a staffing problem. No amount of coaching or process improvement will fix occupancy that is driven by insufficient headcount.

Schedule adherence

Schedule adherence measures whether agents are doing what the schedule says they should be doing at any given moment — on the phone when scheduled to be on the phone, on break when scheduled for break, in training when scheduled for training.

Calculation:

Adherence (%) = (Time in correct state / Total scheduled time) × 100

Why adherence matters for planning: If you plan for 95% adherence but actual adherence is 88%, you have 7% fewer agents on the phones than planned — which has the same effect as 7% additional shrinkage. Low adherence must either be improved or built into the shrinkage assumption so the schedule accounts for it.

Connecting metrics to planning decisions

Each metric above feeds into a specific planning decision. The table below shows which metrics matter for which decisions:

BPO-specific planning metrics

BPOs need additional planning metrics that single-client operations do not:

Review cadence

Workforce planning metrics should be reviewed on a regular cadence — not just when something goes wrong.

The weekly review is the most actionable — it catches problems early enough to adjust next week's schedule but is frequent enough to identify trends before they become crises.