Turn Contact Demand Into Interval Workload

Suppose 120 calls arrive in one hour and average handling time, including after-call work, is five minutes. That creates 600 minutes of work in a 60-minute interval, or 10 Erlangs of offered load. Ten Erlangs describes workload, not the number of agents needed to meet a waiting-time target. Arrival and handling-time variation also affect service.

Calculate Service Coverage Before Applying Shrinkage

Assume one interchangeable agent group, random arrivals, exponential handling times and no abandonment. Under Erlang C, 13 agents are estimated to answer about 76.6% of calls within 20 seconds; 14 agents, about 86.7%. For an illustrative target of 80% within 20 seconds, 14 agents are the minimum under these assumptions. These are model estimates, not a guarantee of actual queue performance.

Keep Occupancy and Shrinkage Separate

With 10 Erlangs of offered load and 14 agents available, modeled occupancy is 10 ÷ 14, or 71.4%. If shrinkage is assumed to be 30%, schedule 14 ÷ 0.70 = 20 people to provide 14 available agents for that interval. Do not apply shrinkage again to breaks or other unavailable time already included in the schedule.

Convert Coverage Into a Workforce Plan

Concurrent staffing is the number of agents available to serve the queue at the same time. Scheduled headcount is the number of people assigned to cover an interval after accounting for shrinkage. Total FTE is paid staffing hours divided by the organization’s full-time hours standard. To estimate it, build feasible schedules across opening hours, skills and contract limits, then sum paid hours and divide by that standard.

Start with interval demand, then calculate service coverage, apply shrinkage once and build schedules that meet opening hours, skill needs and contract limits.