If you need to plan staffing for a call center, help desk, or support queue, this Erlang C calculator gives quick and practical queueing results. Enter your demand and staffing assumptions to estimate service level, average speed of answer (ASA), occupancy, and more.
What this Erlang C calculator does
Erlang C is a queueing model used for workforce management. It answers one key question: given your incoming workload and staffing level, what are customers likely to experience?
- Probability that a contact waits in queue
- Expected service level at a target answer time
- Average speed of answer (ASA)
- Agent occupancy (utilization)
- Estimated minimum agents needed to hit a target service level
How to use the calculator
1) Enter demand
Use forecasted contacts for a time block (for example, 300 calls in 60 minutes). Add your average handle time in seconds. This includes talk, hold, and after-call work if your WFM definition includes all three.
2) Enter staffing
Enter how many agents are actually available to take contacts in that interval. If your team has meetings, breaks, or offline activities, account for that before entering the number.
3) Enter your service objective
Set your target answer time and service level goal (for example, 80% answered in 20 seconds). The calculator then compares your current staffing to that objective.
Erlang C formula summary
First, workload in erlangs is calculated as:
Traffic (A) = Contacts × AHT / Interval seconds
Then, for a given number of agents N and traffic A:
- Occupancy =
A / N - Probability of wait from Erlang C
- Service level at time t =
1 - P(wait) × e^{-((N-A)×t/AHT)} - ASA =
P(wait) × AHT / (N-A)
Important condition: you need N > A for a stable queue. If staffing is below offered load, queue delay grows without bound.
Worked example
Suppose you receive 300 contacts in an hour, AHT is 240 seconds, and you schedule 25 agents. Workload is:
300 × 240 / 3600 = 20 erlangs
That gives 80% occupancy before random variation is considered. The model will estimate wait probability and expected service level at your chosen threshold, such as 20 seconds.
How to interpret results
If occupancy is very high
Occupancy above about 85-90% often feels stressful for teams and is sensitive to volatility. A small spike in arrivals can sharply increase waits.
If service level misses target
Use the “required agents” output as a first-pass staffing estimate, then add shrinkage and schedule constraints in your workforce planning process.
If service level seems too good
Re-check AHT and true productive staffing. Overly optimistic assumptions are the most common reason for unrealistic projections.
Limitations of Erlang C
- Assumes no customer abandonment
- Assumes stable, random arrivals and exponential service behavior
- Does not include multi-skill routing complexity
- Does not directly include shrinkage, absenteeism, or schedule inefficiency
Use Erlang C as a baseline model, then adjust with operational knowledge and historical performance.
Final note
This calculator is designed for quick scenario analysis: “what happens if volume rises 10%?” or “how many agents are needed for 85/30?”. It is ideal for planning conversations, budgeting, and daily staffing checks.