Contact Centre Headcount Against a 90% Service Target
A regional contact centre runs two queues, A and B. In week 12, queue A received 4,800 calls and queue B received 3,600. 92% of A calls and 85% of B calls were answered within 60 seconds. Each answered call takes 6.0 minutes of agent time on average. Each agent is scheduled for 36 hours per week, and 80% of scheduled time is available for call handling; the rest is breaks, coaching and team meetings. The centre currently employs 22 agents. For week 13, volume is forecast to rise 15% in queue A and fall 10% in queue B. The service target is 90% of calls answered within 60 seconds, applied separately to each queue. Assume every answered call takes the same 6.0 minutes regardless of queue, and that calls that miss 60 seconds still consume 6.0 minutes.
1. Using week 12 data, how many agents are needed to handle the actual answered-call workload? Show the calculation and comment on the 22-agent establishment.
Read the worked answer
Answered calls: A 4,800 × 0.92 = 4,416; B 3,600 × 0.85 = 3,060; total 7,476. Agent minutes needed: 7,476 × 6.0 = 44,856. Productive minutes per agent: 36 hours × 60 × 0.80 = 1,728. Agents required: 44,856 ÷ 1,728 = 25.96, so 26 agents.
22 agents provide 22 × 1,728 = 38,016 minutes, a shortfall of 6,840 minutes, equivalent to about four agents. The 92%/85% outcomes were therefore produced with less capacity than the workload required; either the 6.0-minute average is overstated, or the centre absorbed the gap through overtime, short calls or unpaid breaks. The headcount and the handle-time assumption cannot both be right.
2. Will 22 agents meet the week 13 target? What headcount is required?
Read the worked answer
Forecast answered calls at the 90% target: A 4,800 × 1.15 = 5,520 calls; 90% = 4,968. B 3,600 × 0.90 = 3,240 calls; 90% = 2,916. Total 7,884. Minutes: 7,884 × 6.0 = 47,304. Agents: 47,304 ÷ 1,728 = 27.37, so 28 agents.
22 agents give 38,016 minutes, a shortfall of 9,288 minutes, about 5.4 agents. Even 26 agents, enough for week 12, fall short. The target is not achievable with current headcount at a 6.0-minute handle time. Note the trap: queue B's volume falls, but because B must now hit 90% rather than 85%, its answered volume rises from 3,060 to 2,916? No - it falls slightly, but queue A's 15% growth dominates the total.
3. If only 26 agents are available in week 13, what average handling time would be needed to process the forecast answered calls within available time? Is that plausible?
Read the worked answer
26 agents × 1,728 = 44,928 minutes available. Required average handle time = 44,928 ÷ 7,884 = 5.70 minutes, i.e. 5 minutes 42 seconds, a reduction of 0.30 minutes or about 5.0% from 6.0.
That is a small but real stretch: it must apply across both queues and all call types, and it assumes absence, breaks and shrinkage stay at 20%. If the reduction is not achievable, the options are 28 agents, overtime of about 2,376 minutes (about 1.4 agent-weeks), or relaxing the 90% target. Reducing the target is a business decision, not a scheduling fix.