Help Desk Staffing Calculator
Calculate the number of help desk agents required to meet your service level targets using the industry-standard Erlang C formula. Enter your call volume, average handle time, and desired service level to get staffing recommendations.
Formulas Used
1. Traffic Intensity (Erlangs):
A = (Calls per Hour × Average Handle Time in Hours)
2. Erlang C — Probability a Call Waits P(W):
P(W) = [Aᴺ/N! × N/(N−A)] / [Σₖ₌₀ᴺ⁻¹(Aᵏ/k!) + Aᴺ/N! × N/(N−A)]
Where N = number of agents, A = traffic intensity in Erlangs. The system is only stable when N > A.
3. Service Level:
SL = 1 − P(W) × e^[−(N−A) × (Target Time / AHT)]
4. Average Speed of Answer (ASA):
ASA = P(W) × AHT / (N − A)
5. Agent Occupancy:
Occupancy = A / N × 100%
6. Staffed Agents (with Shrinkage):
Staffed Agents = ⌈Raw Agents / (1 − Shrinkage%)⌉
Assumptions & References
- The Erlang C model assumes calls arrive following a Poisson process and service times follow an exponential distribution (M/M/N queue).
- All callers wait indefinitely (no abandonment). For abandonment modeling, Erlang A (ERTA) would be more accurate.
- Shrinkage accounts for time agents are scheduled but unavailable: breaks, training, meetings, coaching, and unplanned absence. Industry standard is 25–35%.
- Occupancy above 85–90% is generally unsustainable and leads to agent burnout and degraded service quality.
- The 80/20 rule (80% of calls answered within 20 seconds) is the most common industry service level benchmark (ICMI, HDI).
- Log-space arithmetic is used internally to prevent numerical overflow for large call volumes or agent counts.
- References: Erlang (1917); ICMI Call Center Management Handbook; HDI Technical Support Practices & Salary Report; Koole & Mandelbaum (2002) Queueing Models of Call Centers.