Erlang C Staffing Model
The first thing you need to have in your toolkit when launching a call center is the Erlang C Calculator. This mathematical calculation is the starting point in call center staffing because it allows you to roughly estimate the number of agents you need.
The Erlang C Calculator is based on a formula, developed by Danish mathematician A.K.Erlang in 1917. It was the time when nobody could even imagine some call center software but people had already used phones. Most telephone exchanges were completely unautomated so callers had to wait in the queue much longer than now having no opportunity to request a callback. Erlang worked in Copenhagen Telephone Company and tried to calculate how many telephone operators were needed to handle a given volume of calls. The result was the Erlang C formula that remained relevant up till now.
How can you use this call center staffing calculator? There are two main formats of it. The first is an Excel-based worksheet that uses macros, automated input sequences, or add-ins. Using them, you can easily build a mini workforce management system and compare whole days or weeks of calls. But you also should be aware of their cons. Many of these calculators do not include shrinkage. Also, they typically work up to around 200 agents. And finally, handmade Erlang Calculators may contain between 20% and 40% errors.
The second format is an online call center staffing calculator. Such tools are freely available and easy to test. They can easily perform some what-if type staffing calculation in the call center. Their limit of agents is 700 (you can exceed this number but the results are most likely to be wrong). Many of these calculators do not include shrinkage as well. And you don’t need to install any additional software to use them.
Regardless of what type the Erlang Calculator you apply, the inputs are basically the same: the number of calls, time period (for example, per one hour), Average Handling Time (AHT), and Service Level (percentage of calls answered within a period of time, e.g. 80% of calls in 20 seconds). The output will be the number of agents you need.
Using the Erlang C call center staffing model is a good idea when setting up a call center. At this stage, you are aware of the approximate number of calls, business hours, and service level. But a lack of historical data does not allow you to make more precise call center forecasting. With this calculator, you can estimate how many agents you need and later adjust it to your current situation.
Additional Factors You Need to Consider
Erlang C formula gives you a basic understanding when creating a staffing model for your call center but it is not enough. The formula was designed for telephone exchanges, not for call centers. It is therefore not surprising that you need to take into account something else. So how to calculate staffing needs in your call center more precisely? Here are some additional factors you need to consider.
1. Attrition rate
Call center attrition is inevitable. Some agents do their job for years while others quickly understand that it is not their cup of tea and quit. Attrition can also be involuntary when employees are fired because of different reasons (due to attendance policy or failing to meet their KPIs). So you should anticipate some attrition rate and take it into account when designing a staffing model. Call center turnover is a reality you have to accept.
Shrinkage is a measurement of anything that takes your agents away from doing their job. It can be external (sickness, public holidays, absenteeism, lateness) or internal (such call center activities as team meetings, coaching, training, system downtime, etc). It is calculated by the following formula:
As a rule, shrinkage is expressed as a percentage of time across a complete 12-month period.
3. Abandoned calls
The Erlang C formula was designed when abandoned calls simply did not exist. But now it is a pervasive reality for call centers. So you have to take it into account. Experts recommend deducting the number of calls that abandon after a short delay (e.g. 10 seconds). It will allow you to remove invalid calls from staffing calculations.
To learn more about how to reduce the number of abandoned calls, read our article.
4. Multiskilled agents
Erlang created his formula for inbound calls only. But what if your agents handle outbound campaigns as well? We can recommend using that formula from the HR manager Dina Ghazaly:
If your call center works with several projects you can add to the numerator the number of them and multiply by the frequency of calls and length of a call.
However, this formula is suitable for call calculations only. If your agents work with other channels - webchat, email, social media - you need to look for a workforce management system (WFM). It is the only efficient way of planning the work in a multichannel call center.
If you want to learn more on how to calculate your agents' productivity, check out our article on this topic.
Full-Time Equivalent (FTE): What Is It and How to Calculate It
Full-time equivalent (FTE) is another metric you have to take into account. It is a unit that shows the workload of every employed person. You can calculate it by dividing an employee's scheduled hours divided by the employer's hours for a full-time workweek. This metric is useful when you have many part-time employees because it helps you to determine the agents’ workload. For example, if your employee works 40 hours a week (or less, depending on your workweek duration), they are 1.0 FTE. If they work 22 hours a week, they are 0.55 FTE. When the percentage of part-timers is small, experts recommend counting all of them as 0.5 FTE. It is less accurate but it will help you not to overstate your workforce.
FTE calculation may be used for different purposes. Employers may apply it for calculating the company’s expenses when paying its workers and for benchmarking revenues or profits per employee. It is also required by some laws, especially The Patient Protection and Affordable Care Act (PPACA) in the US.
Staffing calculation in a call center may be quite a tough task, especially when launching it. But with accumulating market knowledge and historical data, it becomes easier and more accurate to forecast your workload and the number of staff you need.