Erlang A inputs
One interval at a time, per queue or skill.
Fill this in to see what your current team would achieve.
To hit 80% in 20s with an average patience of 120s.
Agents required plus shrinkage, the number to roster.
| Model | Agents | Service level | Occupancy |
|---|---|---|---|
| Erlang A | 87 | 81.1% | 92.6% |
| Erlang C | 91 | 81.4% | 91.6% |
Erlang C asks for 4 more agents because it staffs for callers who would have hung up anyway. Erlang A counts only the work that stays.
Save or share these results
Take a snapshot of the exact inputs and outputs on screen.
Get the full Excel/VBA version
The full Excel/VBA version with live formulas, RTM tracking, and dashboards is free for now.
Results are estimates for planning purposes only and should not replace professional workforce management judgment.
Erlang A models the same queue as Erlang C with one honest addition: callers have limited patience. Each waiting caller hangs up at a steady rate, so the queue drains itself even when the team is behind. That is why the model stays stable, and useful, even when the team is smaller than the workload.
Step by step
- Traffic intensity = (calls per interval × AHT) ÷ interval seconds. The workload in Erlangs, exactly as in Erlang C.
- Patience sets the abandonment rate. A patience of 120 seconds means a waiting caller hangs up at a rate of 1 in 120 per second.
- Steady-state probabilities come from the birth-death balance of the queue: arrivals add callers, service and abandonment remove them.
- Abandonment % = waiting callers ÷ (patience × arrival rate). Every caller who leaves is one the team never handles.
- Agents required is the smallest team that still hits your service level target once abandonment is counted.
Why Erlang C over-staffs when callers abandon
Erlang C prices every call as if it waits as long as it takes. When real callers hang up, part of the workload disappears on its own, and Erlang C still buys agents to cover it. The busier the queue and the shorter the patience, the bigger the gap between the two models.
Using it well
- Estimate patience from abandoned-call wait times in your phone system, not from a guess.
- Run both models: treat Erlang C as the conservative ceiling and Erlang A as the realistic plan.
- Be careful with very short patience figures: they make staffing look cheap by accepting high abandonment.
- Compare against the Erlang C Calculator and sense-check the roster with the Capacity Planning Calculator.
Frequently asked questions
What is the difference between Erlang A and Erlang C?
Erlang C assumes every caller waits forever. Erlang A models callers hanging up once their patience runs out, so it usually needs fewer agents for the same service level.
What is average patience?
The average time a caller will wait in queue before hanging up. Estimate it from your phone system as the average time abandoned calls spent waiting, then add a margin for the callers who would have waited longer.
When should I use Erlang A instead of Erlang C?
Whenever abandonment is material, roughly above 1 to 2 percent, or patience is short. When almost nobody hangs up, the two models give nearly the same answer.
Why does Erlang C over-staff when callers abandon?
Because it staffs for callers who in reality leave the queue on their own. Every abandoned call is work the team never has to do, and Erlang C prices that work in anyway.
Related reading: why standard Erlang C overstaffs digital channels and how callbacks move workload rather than remove it.
More tools
See all toolsService Level Recovery Calculator
Behind on service level mid-month? Get the volume-weighted SL you need for the rest of the period, and see instantly if the target is already unreachable.
Open calculatorErlang C Staffing Calculator
Turn call volume, AHT and service level targets into required agents, occupancy, ASA and shrinkage-adjusted headcount.
Open calculatorBack Office FTE Calculator
Turn email, claims and processing volumes into the FTE you need, per work type, with shrinkage and utilisation built in.
Open calculator