Free contact centre staffing calculator. Powered by the Erlang C formula, the same maths behind every enterprise WFM system. No login. No spreadsheet. Just answers.
How many agents do you need to hit your service level? Adjust inputs and results update instantly.
See how many agents you need across the day. Settings are pulled from the Calculator tab.
| Time | Calls | Agents | Occ % |
|---|
This tool uses the Erlang C formula, the global standard for contact centre staffing. It was created in 1917 by Danish mathematician A.K. Erlang and is built into every major workforce management system in the world, including NICE, Verint, Genesys and Calabrio.
When you use this tool, you're using the same maths that powers enterprise WFM software costing tens of thousands of dollars. The formula itself is open, proven and has been validated for over 100 years.
Calls arrive randomly. Even if you average 100 calls per hour, you might get 12 in one five-minute window and 3 in the next. This randomness means you need more agents than a simple "calls divided by capacity" calculation suggests.
The Erlang C formula models this randomness. Given your call volume, handle time and service level target, it calculates the probability that a call will wait in queue, then finds the minimum agents needed to keep that probability below your target.
On top of the raw calculation, this tool factors in shrinkage (agents unavailable due to breaks, meetings and leave) and maximum occupancy (how busy agents should be before burnout becomes a risk).
Traffic Intensity (Erlangs): A = (Calls x AHT) / Interval Length Example: (100 calls x 180s) / 1800s = 10 Erlangs Erlang C - Probability of Waiting: P(wait) = [A^N/N! x N/(N-A)] / [Sum(k=0..N-1) A^k/k! + A^N/N! x N/(N-A)] Where: N = agents, A = traffic intensity Service Level: SL = 1 - P(wait) x e^(-(N-A) x TargetTime/AHT) Average Speed of Answer: ASA = P(wait) x AHT / (N - A) Occupancy: Occ = A / N After shrinkage: Adjusted = Raw Agents / (1 - Shrinkage%)
No model is perfect. Being upfront about limitations builds trust in the results:
Assumes random arrival. If your calls are scheduled (callbacks, outbound), the formula overestimates. It works best for inbound, unscheduled contacts.
Single-skill only. Erlang C models one queue. Multi-skill or skill-based routing needs more complex modelling.
Not ideal for chat or email. Agents handle concurrent chats, which breaks the one-at-a-time assumption. Use as a rough guide only for non-voice.
Point-in-time. It tells you how many agents you need for a given call rate. It doesn't tell you how to schedule shifts across a week. That's where the Day Planner and Roster Builder come in.
Practical guides on the concepts behind this calculator, written for team leaders who are figuring this out for the first time.
Staffing a contact centre is a balancing act. Too few agents and callers wait in long queues, abandon calls and leave frustrated. Too many agents and you're paying for people to sit idle. The challenge is finding the number that keeps callers happy without burning through your budget.
Most team leaders start by guessing. They look at yesterday's numbers, add a buffer and hope for the best. That works until it doesn't. A spike in volume, a few people calling in sick, or a team meeting scheduled at the wrong time can blow the whole day open.
The better approach is to calculate your staffing needs using a proven formula. The Erlang C formula takes your call volume, average handle time and service level target, then works out the minimum number of agents needed to keep wait times below your threshold. It accounts for the randomness of call arrivals, which is why simple "calls divided by capacity" maths consistently underestimates the agents you need.
To calculate staffing, you need four numbers from your phone system or ACD reports:
The Erlang C formula was created in 1917 by Danish mathematician A.K. Erlang while working at the Copenhagen Telephone Company. He needed to calculate how many telephone operators were required to handle a given volume of calls without making callers wait too long.
Over 100 years later, the same formula is built into every major workforce management system in the world, including NICE, Verint, Genesys and Calabrio. When you use this calculator, you're using the same maths that powers enterprise WFM software costing tens of thousands of dollars per year.
The formula works by modelling the random arrival of calls. Even if you average 100 calls per hour, the actual pattern is uneven. You might get 15 calls in one five-minute window and 3 in the next. This randomness means you always need more agents than a simple average would suggest. Erlang C calculates exactly how many more.
The calculator on this page adds two practical adjustments on top of the raw Erlang C result: shrinkage (to account for agents who are rostered but not on the phones) and maximum occupancy (to prevent burnout from back-to-back calls with no breathing room).
Terms you'll see in this calculator and across workforce management.