The Complete Guide to Erlang C Staffing: From Formula to Headcount

Erlang C is the foundation of every WFM staffing model — but most teams use it wrong. They plug in call volume and AHT, get a number, and hand it to finance. The problem is that Erlang C gives you minimum agents at the interval level. It doesn’t account for shrinkage, schedule inefficiency, or the reality that you can’t hire 0.4 of a person.

This guide walks through the complete Erlang C workflow: how to calculate traffic intensity (erlangs), why occupancy above 85% destroys agent experience, how to layer shrinkage correctly (most teams double-count breaks), and how to convert interval staffing into FTE headcount requests that finance actually approves. We cover the common mistakes — using daily averages instead of peak intervals, ignoring AHT variance across skill groups, and the dangerous assumption that 80/20 service level means 80% of calls are answered in 20 seconds (it means 80% are answered within 20 seconds of entering queue, which is different).

Every calculation includes benchmarks from real operations. We show what good looks like at 200-seat, 500-seat, and 1000+ seat centres — because the math scales differently than you expect. At smaller volumes, a single agent makes a massive service level difference. At larger volumes, the efficiency gains are logarithmic. If you’re running Erlang C without understanding these dynamics, you’re either overstaffing (wasting budget) or understaffing (burning agents and missing SLAs).

More from the Knowledge Base