By Keith Johnson and Heather Rafferty
Call center managers must meet certain requirements as efficiently as possible and they only have a few sources of spend. How do you determine if you are over or under staffed? If you are under, how can you justify more staffing? Fortunately, there tools available to help you deliver on customer expectations either in support of, or as a prelude to a transition to an Automated Call Distribution Platform.
Modeling the Data
Many call center managers have invested in technology to manage incoming calls and track key performance indicators. Modeling can be used to transform this data into actionable information and lead to a 10-30% increase in efficiency.
To illustrate, let’s consider an inbound call center at a bank. Last year, they upgraded their phone system and have collected several months of performance data.
Key assumptions needed to create our model are: calls per hour, average handle time, service level, hours of operation, shift schedules. Efficiency is measured by the total cost to operate the call center, the number of seated agents required, and percent of calls answered within an amount of time.
Calls per hour has a large impact on the number of seated agents required. For most organizations, a month’s worth of call data will tell you what you need to know. Some calls may take a half hour, most measure handle times in seconds. Different companies will also use varying operational definitions to measure handle time – you must include time spent on the line with an agent plus any wrap up time afterward i.e. updating records, making notes etc.
Run the Model
In this example, we’re using a tool called an Erlang calculator, named after Agner Karup Erlang, a Danish mathematician credited with creating the field of telephone network analysis. Figure 1 below was created in Minitab showing the average handle time in seconds.
Consider our bank, which receives an average of 390 total calls per day with an average handle time of 180 seconds. Basic math shows that the total time supporting our customers is 390 calls x 180 seconds per call x 3600 seconds per hour = 19.5 hours per day
Next, we incorporate the Service Level Agreement (SLA), setting the standard of service between your organization and its customers. This is measured as an acceptable average percent of calls answered within a time limit. A common goal is 80/30, where 80% of calls are answered by an agent within 30 seconds. This has the single greatest impact on cost. An SLA of 90/20 is more expensive than an 80/30. Overdoing this, for example 60/100, can cause problems with callers getting tired of waiting and hanging up. You won’t need many agents but you could lose customers – it’s a fine balance.
On the staffing tab it looks like you’re overstaffed. The day’s shift has 10 seated agents, but the required number of seated agents varies between two and seven depending on variation in call volume.
The scenario below describes understaffing. In this instance, an average percent of calls is answered in 146 seconds by 5 employees. The impact will show up as less satisfied customers and low morale amongst agents fielding their calls.
Now that you have a clear understanding of what’s happening, you can make strategic decisions.
When your data describes sufficient staffing for the majority of the day, but you’re still falling short at key peak moments, reconsider scheduling. Shown below, a call center needs 9 seated agents from 11AM-2PM, but only 5 for the rest of the day. The scheduling solution can come in the form of working with contract, overtime, part-time, or temporary labor. Are there current employees that have extra availability to step in when needed? Are you using lead agents, whose primary job is coaching and assisting other agents, to sit the phones during peak hours? Rather than adding a full-time employee, investigate the possibility of supplementing your team’s capacity as needed.
Call center management requires both taking care of the customer and the business. Just as you don’t want to waste the financial resources you have, you also have to meet the needs of your customer. An overstaffed floor often leads to a bored and under-stimulated staff with low morale. On the opposite end, a talent pool that is stretched too tightly across a barrel will be ineffective and hard to keep. Optimizing scheduling is one of many ways to get started with improving your call center’s efficiency.
Keith Johnson is a principal with Firefly Consulting (www.firefly-consulting.com).
Heather Rafferty is a professional marketing and communications specialist with experience in Lean Six Sigma, continuous improvement, and operational efficiency.