TLDR
Workforce scheduling in a call center means having the right number of agents on the phones at the right time. Forecast calls for each hour, work out the agents you need, add shrinkage, build shifts that match, and check schedule adherence every week. LiveAgent gives you the call data, but you build the roster yourself.
What is workforce scheduling in a call center?
Workforce scheduling is the process of matching agent hours to expected call volume. Done well, you have enough people for the busy hours and no one waiting idle in the quiet ones. It is the scheduling side of workforce management (WFM), which also covers hiring, training and performance, so you may also see it called “workforce management scheduling”.
In a call center the work follows one loop: forecast the calls, calculate the staff, add shrinkage, build shifts, then measure adherence and adjust. The sections below walk through each step with one example queue and one busy hour. For the wider picture, see our guide to call center management .
Step 1: How do you forecast call volume?
Start with your own history. Count the calls in each interval (30 or 60 minutes) for the same weekday over several weeks, and average them. This is the core of call center forecasting, and it works better than a gut feeling.
Then adjust the average for what you already know: a campaign, a product launch, a holiday or a seasonal peak. Calls on a Monday morning look different from calls on a Friday afternoon, so keep each weekday separate. Our page on call volume explains the basics.
LiveAgent’s Call load report lists every hour of the last 30 days, with the minimum, average and maximum of slots, active calls and customers waiting in the queue. The report keeps only 30 days. Copy the table into a spreadsheet every month, and you build a longer history than LiveAgent stores.
For example, the last four Tuesdays had 110, 125, 118 and 127 calls between 10:00 and 11:00. The average is 120 calls, and that is the forecast for the next Tuesday at 10:00. We use this hour in the next steps.
Step 2: How many agents do you need per interval?
First turn calls into workload. Offered load is calls per hour multiplied by average handle time (AHT), measured in Erlangs. One Erlang is one agent busy for a full hour. Our page on average handle time shows how to measure it.
In the example, 120 calls with a 6-minute AHT make 720 minutes of work, which is 12 Erlangs. Twelve agents are not enough. Calls arrive at random times, so you need extra agents to keep callers from waiting.
Erlang C is the standard formula for this. It uses the offered load and the number of agents to estimate the chance that a caller has to wait. The formula goes back to A. K. Erlang, a Danish engineer who joined the Copenhagen Telephone Company in 1908 and published his work on telephone exchanges in 1917, according to the MacTutor History of Mathematics .
Erlang C has one known limit. It assumes callers never hang up, so it can overstate the staff you need, as TechTarget explains . Treat the result as a safe upper estimate and check it against real data.
You also need a service level target. It is the share of calls answered within a set time, and teams often write it as 80/20, meaning 80% of calls in 20 seconds.
In the example (12 Erlangs, 6-minute AHT, 20-second target), Erlang C gives 16 agents: with 16 agents, about 84% of calls are answered within 20 seconds. With 15 agents, only about 73% are, which misses the 80/20 target.
This step is also where call center capacity planning starts. Repeat the calculation for every interval of the day, and you get the number of agents you need on the phones for each hour.
Join our newsletter
The latest tips and deals delivered straight to your inbox.
What is shrinkage in a call center, and how do you calculate it?
Shrinkage is the share of paid time when agents are not available to take contacts. Breaks, training, meetings, coaching, holidays and sick days all count. Call center shrinkage is the reason a team of 16 on the schedule cannot cover a 16-agent requirement.
Calculate it as paid hours not available for contacts, divided by total paid hours, multiplied by 100. Some of it is planned (training, meetings) and some is not (sick days). Track both parts, because you can move the planned part and only predict the unplanned part.
For example, a team has 400 paid hours per week, and 120 of them go to breaks, training and absence. Shrinkage is 120 ÷ 400 × 100 = 30%. This 30% is an example, not a benchmark. Use your own numbers.
Then raise the staff count: scheduled staff equals required staff divided by (1 minus shrinkage). In the example, 16 ÷ 0.70 = 22.9, so you schedule 23 agents for the 10:00 hour.
Step 3: How does call center scheduling work with shifts?
Call center scheduling turns the staffing curve into shifts that people can work. Your goal is to cover every interval of the curve with as few extra hours as possible. Build the shifts around the curve, not the other way around.
These tactics help with call center shift scheduling:
- Stagger start times. Starts at 8:00, 9:00 and 10:00 follow a rising morning curve better than one big shift.
- Put breaks off the peak. Move breaks and lunches to the quiet intervals, and keep them off the busiest hour.
- Use part-time shifts for peaks. A four-hour shift over the lunch peak is cheaper than a full day.
- Plan for evenings and weekends. Customer service reps often work during busy times, including evenings, weekends and holidays, according to the U.S. Bureau of Labor Statistics (checked October 2026). Rotate these shifts fairly.
- Include remote and hybrid agents. Nearly 70% of contact centers use a hybrid model, according to the ICMI State of the Contact Center 2024 , so your schedule often covers home and office agents together.
- Publish early and keep it stable. Agents plan their lives around the schedule, and a stable one reduces swaps and absence.
Check the finished schedule against the curve, interval by interval. A gap of one or two agents in the peak hour hurts more than an extra agent at 14:00. The same method applies to contact center scheduling when you also have chat and email agents, but each channel needs its own forecast.
What is schedule adherence, and how do you measure it?
Schedule adherence shows how closely agents follow the schedule. A perfect schedule fails if agents are on a break when the peak arrives, so call center adherence matters as much as the forecast.
The schedule adherence formula is the minutes an agent spent in the scheduled state, divided by total scheduled minutes, multiplied by 100. For example, an agent scheduled for 480 minutes who spent 432 of them in the planned state has 432 ÷ 480 × 100 = 90% adherence.
Some teams count the minutes out of adherence instead. Both ways work if you pick one and keep it.
To improve adherence in a call center, try these steps:
- Share the schedule and the reason for it. Agents follow a plan they understand.
- Keep breaks short and fixed, and allow swaps through a simple process.
- Review adherence together every week, not only when numbers drop.
- Look for the cause of out-of-adherence time, such as long wrap-up or system problems, before blaming agents.
- Fix the schedule when it keeps failing the same interval.
Our guide to call center best practices covers adherence to schedule in the wider context of team standards. The related agent utilization page explains how busy agents are while they are available.
Which formulas do you need?
This table collects every formula from the example in one place.
| Metric | Formula | Example value |
|---|---|---|
| Offered load (Erlangs) | Calls per hour × AHT in hours | 120 calls × 6 min ÷ 60 = 12 Erlangs |
| Service level | % of calls answered within the target time | 80/20 target: 80% in 20 seconds |
| Required agents | Smallest number of agents (N) where service level = 1 − Pw × e^(−(N − Erlangs) × target seconds ÷ AHT seconds) reaches the target; Pw is the Erlang C chance that a caller waits | 16 agents (about 84% in 20 seconds; 15 agents give about 73%) |
| Shrinkage % | Paid hours not available ÷ total paid hours × 100 | 120 ÷ 400 × 100 = 30% |
| Scheduled staff | Required agents ÷ (1 − shrinkage) | 16 ÷ 0.70 = 22.9, so 23 |
| Schedule adherence % | Minutes in the scheduled state ÷ total scheduled minutes × 100 | 432 ÷ 480 × 100 = 90% |
All values are examples. Replace them with your own data.
How does LiveAgent support call center scheduling?
LiveAgent is our product. As our call center software comparison says, scheduling, shift management and predictive staffing are not built in and require third-party tools. What LiveAgent gives you is the data to feed a schedule and the controls to protect it.
The call center is available on the Team, Business and Enterprise plans. The reports are on Team and up. You can see plan details on the pricing page .
| Need | Where in LiveAgent |
|---|---|
| Hourly call demand for the forecast | Call load report in the Reports section: 30 days of hourly slots, active calls and queue |
| Times the call center was open | Call availability report: the exact times of each day when the call center service was available |
| Which agents were online, and when | Agent availability reports: agents, departments and hours online, kept for 30 days |
| Talk time, missed calls and pickup time | Performance report: calls, missed calls, call pickup average time (includes time in IVR and queue) and call minutes (rounded up per call); call minutes ÷ calls gives a rough average talk time for your AHT |
| Worktime and calls per agent | Agent report, or the API agents report for an outside tool |
| Who gets the call | Call routing in Configuration > Call > Settings: Random Assignment or Priority Assignment, plus a routing time of 15 to 120 seconds (default 30) |
| Limit on waiting | Maximum queue time in the same settings: after it, the call continues in the IVR offline section (SIP numbers only) |
| Calls outside open hours | A rule on “incoming call started” that changes the IVR when the time is outside your business hours |
| Callers who do not want to wait | Automatic callback |
Agents control their own availability with the “Available for calls” checkbox and by choosing the departments they take calls for. Teams that use an outside workforce management tool can pull worktime and call counts from the LiveAgent API.
Two limits matter for planning. The availability and load reports keep data for only 30 days, so copy what you need each month. LiveAgent also does not show real-time adherence, so adherence is something you check after the fact, by comparing availability reports with your schedule.
If your call center works under response targets, our guide to SLA management shows how to set them. When you compare a remote and an on-site team, our page on on-site vs remote customer service helps you plan both.
Sources
This page follows LiveAgent’s support documentation, checked on October 21, 2026: Reports section overview , Load reports , Availability reports , Performance report , Agent report , Call center settings , Availability for call center service and Changing IVR automatically .
Outside sources: TechTarget, Erlang C definition , MacTutor, Erlang biography , U.S. Bureau of Labor Statistics, Customer Service Representatives and ICMI, State of the Contact Center 2024 .
Next step
Open the Call load report today and copy last month’s hourly data into a spreadsheet. Use it as the base for your next forecast, then work through the six formulas above. Not a LiveAgent user yet? Start your free trial and see the call data on your own queue.