Turns your volume, handle time, service level target and shrinkage into the FTE (full-time equivalent agents) you need to schedule, using Erlang C, the standard queueing model for staffing. Erlang C assumes one contact per agent at a time, so it fits voice queues.
Base Agents
88
On the phones, before shrinkage
Scheduled FTE
126
With 30% shrinkage
Occupancy
90.9%
Above 90%, recovery time between contacts collapses. Burnout and attrition risk is high. Add capacity or reduce load.
Service Level
81.8%
Meets target: 80% in 20s
Avg Speed of Answer
13s
Average wait before an agent answers
Probability of wait
28.4%
Chance a caller waits
What it means
Staffing to your service level alone puts occupancy (the share of paid queue time agents spend handling contacts) at 90.9%, above the 83 to 87% band. Erlang C often lands there at this volume; it does not point to an error in your inputs. Occupancy is the constraint to manage here, since the service level target is already met. Holding an 87% ceiling instead would take 132 FTE against 126 in the busiest interval. The 6 FTE difference is the price of agent recovery time; held across your open hours it comes to about $523K a year on benchmark wages.
Occupancy risk · target 83 to 87%
0%<85% healthy85 to 90% caution>90% critical
Occupancy steps down each time another agent is required, so it rises then drops as volume grows. Small teams swing more than large ones.
Staffed to service level
126 FTE
90.9% occupancy
Staffed to a 87% ceiling
132 FTE
87.0% occupancy
Difference
+6 FTE
$523K a year across your open hours, the price of recovery time
What-if scenarios
+20% volume spike
150 FTE
+24 agents, 91% occupancy
+10% AHT increase
139 FTE
+13 agents, 91% occupancy
+5pt shrinkage
136 FTE
+10 agents, 91% occupancy
Raise SL to 85%
128 FTE
+2 agents, 90% occupancy
Methodology
Erlang C is the standard queueing model for staffing. We solve it through the Erlang B recursion, which stays accurate from a handful of agents to several thousand. It assumes calls arrive at random (Poisson arrivals), handle times vary exponentially and callers never hang up, so it tends to over-staff. Enter an average patience to see the estimate adjusted for abandonment. The optional occupancy cap staffs to whichever is higher: the agents that meet service level, or the agents that hold occupancy at or below your ceiling. Shrinkage is applied after the agent calculation to turn base agents into scheduled FTE. Erlang C assumes one contact per agent at a time. Chat, messaging and email agents run several sessions at once, so applying these numbers to a digital queue overstates headcount, often by half or more. Every formula, constant and a worked example are in the published method.
Why Occupancy Risk gives a different cost for this ceiling
Holding the 87% occupancy ceiling, counted and priced each way:
$523K a yearStaffingThis page6 more scheduled FTE after 30% shrinkage across your open hours, at the fully loaded benchmark rate
$233K a yearOccupancy RiskThe same ceiling counted its way: 4 more agents on the phone, at $21.53 an hour plus benefits
Why they differ
Staffing counts scheduled FTE: the agents on the phone grossed up for shrinkage, the paid time agents spend off the phones (30% here). Occupancy Risk counts only the agents who must be on the phone.
Staffing prices an FTE fully loaded at 1.95 times wage, or at your own TCO figure: benefits, supervision, seats and technology. Occupancy Risk prices an agent at wage plus benefits, 1.3 times wage.
Staffing sizes agents with Erlang C against your service level and then holds the occupancy ceiling; Occupancy Risk divides the workload by the target and compares it with the agents you have today, so the starting point can differ too.
Use Staffing's figure for a budget, since it carries the whole cost of an added seat. Use Occupancy Risk's to see the agent time the target needs.
Published figures to hold your result against. Each measures something slightly different, so read what it counts before you compare. None of them changes your result or its grade.
Service level targetsPractice range
The common target is a convention. A few regulators set their own standard for the centers they oversee.
SQM Group, 19 Jan 2023. What Are the Industry Standards for the Top Call Center KPIs?
80% in 20 sTraditional service level standardSQM adds that it finds no satisfaction penalty for calls answered within 120 seconds.
80% in 30 sCalifornia water utilities, Class A and BRegulatory standardShare of callers reaching a representative in business hours; abandoned calls at or below 5%.California Public Utilities Commission, 2009. General Order 103-A, Appendix E, customer service standards for Class A and B water utilities (effective 10 Sep 2009).
2:00Medicare Advantage lines, CMS standardRegulatory standardAverage hold after the phone menu. CMS test calls measured 0:32 in January to June 2025.CMS, 2025. 2026 Part C and D Display Measures Technical Notes, Attachment A (in the 2026 display measures download).
74 sAverage speed of answer, USBenchmark studyOne third above the pre-pandemic average.ContactBabel, 2026. The 2026 US Contact Center Decision-Makers' Guide, key findings.
6%Abandon rate, industry standardBenchmark studySQM counts below 5% as favourable.
Voice handle timeBenchmark study
All industries. Publishers define handle time differently, so the figures are not averaged. No free source publishes handle time by industry.
11:37Average handle time, 2024697 seconds of talk and wrap-up, 18% more than the year before.SQM Group, 6 Feb 2025. Call Center FCR Benchmark 2024 Results by Industry.
7+ minTypical service call, US, 2024"Over seven minutes", 38% longer than in 2012.ContactBabel, read on Contact Center Pipeline, 14 Aug 2025. The 2025 US Contact Center Decision-Makers' Guide, as reported in "The Great Contact Center Standoff".
Agent pay where your agents workOfficial statistic
The US Bureau of Labor Statistics median for customer service representatives, May 2025. The tool opens at the national median; if you know your own pay, enter it in the tool.
$21.53Median hourly wage, customer service representatives, United States$44,770 a year, before benefits. Half earn more, half less.US Bureau of Labor Statistics, read on O*NET OnLine, May 2025. Occupational Employment and Wage Statistics, May 2025, Customer Service Representatives (43-4051), by state.
This scenario is too detailed to fit in a link. Download instead.
A person on our team sends it, usually within one business day. Your address is used to send this report and to reply if you ask a question. Nothing you entered leaves your browser unless you send it here, and the download never asks for it.
Run this next
Each result raises a sharper question. This is the diagnostic that answers it.
Send this analysis for an independent review. You get a written read on what the numbers support, what they do not, and which diagnostic is worth running next. No sales call. No vendor introduction.
88 on the phones, 90.9% occupancy, 81.8% answered in 20s
How sureDirectionalHeld by evidenceShows the direction and rough size. Something behind it is still a default, an estimate or an open choice.
Annual cost of this plan
$11.0M
126 FTE on payroll: 126 scheduled in an average open interval, held for 40 hours a week, at $7K per agent per month. Based on a benchmark median of $21.53 per hour, loaded at 1.95x; these are not your own figures. Run the TCO Calculator (total cost of ownership) to price this on your own cost base.
Still at our default hours open and average interval: this year assumes 40 hours open and every open interval as busy as the one above. Enter yours under Question 3.