While different call centers have different priorities and different functions, the challenge of staffing remains relatively consistent regardless of size or specialty. These six steps can help a call center manager successfully traverse the staffing minefield.
1. Gather and Analyze Data
The most accurate and reliable guide to staffing, as anyone who studies workforce management can tell you, is to look back at past performance and call center history. Review the reports generated by the automatic call distributor for data on average handle time, number of incoming calls and other key performance indicators.
To create a staffing schedule for, say, the first week of April, the obvious place to start is with the data for the first week of April of the previous year, and the year before if that information is available. The more historical data used, the better the chance of an accurate forecast. Variations should also be considered – where does Easter fall this year? Will that impact call volume? Will more students be on spring break?
When consulting previous weeks/months/years of information, the two numbers that will most strongly impact forecasts are call volume and average handle time, either calculated per hour or per half-hour.
2. Crunch the Numbers for a Workload Calculation
There are three methods typically employed by call centers to translate historical data into a staffing forecast:
With this system the call center relies on a basic apples-to-apples comparison of a future point in time with that same point in the past. For example, forecasting next June 15 based on traffic numbers from June 15 of last year. While this is a good starting point, it will not be precise as it does not account for more recent calling trends or new products or promotions.
With this method a manager would average relevant past numbers to predict call volume, preferably while relying more heavily on recent data (by creating a formula that uses these numbers more prominently). However, this may still not take into account some changes or events that would have figured into older data.
Time Series Analysis
With time series analysis, historical data is calculated alongside monthly or seasonal changes, as well as more recent events and other variables. It is a more comprehensive approach that typically results in better forecasts.
3. Staffing Calculations
Steps #1 and #2 are used to create the forecast. Now it’s time to formulate a schedule. The call volume forecast numbers are factored into workload predictions, workload being the number derived from multiplying the amount of forecasted calls and the average call handle time.
Most managers will add additional staff to whatever number of agents is deemed appropriate, both to compensate for unexpected absences and to maintain customer service levels should call volume be higher than anticipated. The unproductive hours designated as “shrinkage” – breaks, training time, tardiness, meetings – must also be considered. At most call centers, shrinkage rates fall somewhere between 20% and 35%, depending on the size of the business. In general, larger call centers can absorb these variables more easily because of a more favorable staff-to-workload ratio.
Another factor is how busy each agent will be during a shift, referred to as agent occupancy. The goal is to achieve an optimum balance between not sitting around for extended periods of time between calls, and not having a long queue of calls waiting that might result in rushing a customer call, to the detriment of that engagement. As a percentage, 85-90% is considered an acceptable occupancy range.
4. Create Assignments
Creating a staff schedule is all about getting the right number of the right people in position to handle the customer service needs of the call center. Once the calculations from the previous step have been completed, the manager should know how many agents would be needed for the shift in question.
As some call centers operate with full-time agents and others use part-time and telecommuting employees, this is when shift lengths and resources must be defined, days off specified and personnel scheduled. Depending on the size of the call center, there may be dozens, if not hundreds, of scheduling possibilities. If skill-based routing is also a priority, this will also affect staffing decisions. Once personnel have been selected, the manager also has the option of staggering start times by 15 or 30 minutes, which can reduce instances of too many agents taking lunch breaks or other diversions at the same time during a shift.
5. Management and Adjustment
There is no way to know if a plan is going to work until it is executed. Even with the preparations and calculations already described, staff schedules will likely still have to be adjusted every day. This ongoing management of staff and schedule is referred to as performance tracking.
The main components of performance tracking are call volume, AHT and staffing levels. Deviations from forecasting predictions may require staffing adjustments, assuming enough flexibility has been built into the schedule to make the necessary changes. If not, call center service goals may be in jeopardy. Tracking the number of a calls in queue may also require some “instant forecasting” to adjust the remainder of the shift accordingly. However, over-reaction should also be avoided, lest a random surge be mistaken for a full-day trend.
6. Review, Analyze, and Adjust
The end of a shift is the beginning of preparation for the next one. The challenge of staffing is ongoing, but each day’s results deliver data to analyze that may result in ways to improve performance, both for each individual agent and the entire team.
Many of the most persistent challenges of staffing can be mitigated when call center managers know what to look for, when they have the information they need, when they need it, and when they can act upon it quickly.
No one every said predicting the future was easy. But an effective, automated workforce management solution can make the necessary calculations, remove much of the guesswork and improve the accuracy of schedules and forecasts. Through real-time measurement of call center metrics, agents and managers gain the data visibility necessary to deliver the service that customers expect, and can react more quickly to issues and resolve them before they impact operations.