A Simple Approach to Staffing for Outbound Calls
By Penny Reynolds
Workforce management is all about getting the right number of staff in place each interval of the day to handle contacts to meet service goals while minimizing cost. While getting the resources in place for incoming calls is the most common workforce management problem, staffing for outbound calling also happens for most centers.
In terms of all the steps of workforce management, staffing for outbound calls can be a simpler problem than inbound staffing. First, the center gets to decide how many calls to place and when they will happen. There is a science to determining best times to call, but an intricate analysis for trend, seasonality, and other calling patterns is generally not needed. Usually the management team decides that a number of calls will be placed over a defined period of time and staffing is calculated accordingly.
There are a few things to keep in mind, however, when addressing outbound calling needs. It is important to consider workload calculations, the right staffing model to use, and productivity considerations.
Workload for inbound calling is calculated by multiplying forecast calling volume by average handle time (AHT). Outbound call workload uses the same calculation. Determining call volume is pretty simple, but the main difference in calculating workload for outbound calling involves identifying total handle time. While inbound calling AHT is made up of talk time and after-call work, outbound call workload has an additional component – call set-up time.
How much set-up time depends on how outbound calls will be placed. If an automated dialer is being used, then there may be little to no set-up time as part of agent handle time. In a predictive mode, the dialer may look up the number, dial the number, and wait for customer connect before placing the call with an agent. In this case there is no pre-call work for the agent and no additional set-up time is part of handle time.
However, in a manual dialing scenario, the agent’s time for each call may include the following:
- Looking up number. This can be a manual look-up process or an automated one.
- Dialing number. This can be a manual dialing process, or one done by the ACD or dialer.
- Waiting for connect. There is typically a small amount of time for the call to be processed out of the center and through the telephone network.
- Waiting for customer connect. Once the call rings on the other end, there may be several rings before answer and conversation.
- Time on non-connects. Not all calls result in a live customer contact. Some may ring and not answer, get a busy signal, or go to voice mail. Each of these takes time and has to be factored in to the overall set-up time.
Consider an example where it takes an agent 2 seconds to activate the handset or headset, and another 5 seconds to look up a telephone number. This is followed by 4 seconds to dial the number, 3 seconds of network connect, and waiting three rings (at 6 seconds each) for answer. With these numbers, that is an additional 32 seconds per call to factor into handle time for every call.
Compare that set-up time to the rest of handle time. If an agent spends 135 seconds talking to the customer and there is no wrap-up time after the call, the total workload (or handle time) per call is 167 seconds. The 32 seconds of set-up time compared to the 167 total seconds of the call is about 20% of the call. And this is a conservative number, since it assumes all calls are answered.
Now consider what happens when we factor in the other types of calls: the no-answers, busy signals, voice mail, etc to the calculation. The table here provides some averages for these numbers assuming a sample of 100 calls.
|Call Volume||Type of Call||Set-up Time||Talk Time|
|40||Answered call||32 sec||135 sec|
|40||Voice mail||34 sec||0 sec|
|15||Busy signal||22 sec||0 sec|
|5||No answer||46 sec||0 sec|
|100 calls||All Calls||3200 sec||5400 sec|
If your center has an automated dialer, you should have reports that show what percent of calls in any campaign have a customer connect versus getting a busy signal, going to voice mail, or ringing with no answer. If you don’t have a dialer, simply note on a representative sample of calls what percentage go into each of these categories and how much time is associated with each scenario.
In the example here, assume 40% will be answered, another 40% will go to voice mail, 15% will get a busy signal, and the remaining 5% will ring and ring with no answer. We assume set-up time of 32 seconds for connected calls. Assume 34 seconds for voice mail with no messages being left, 22 seconds for a busy signal (fewer rings), and 46 seconds for a ring no answer (more rings).
As shown in the table, that’s a total of 3200 seconds of set-up time compared to only 5400 seconds of conversation time. That means in this example, 37% or over a third of agent time, is spent in call set-up time. This makes it easy to see why automated dialers have such a quick return on investment.
Now that set-up time has been calculated, it can be incorporated into a workload calculation. In this example, let’s assume the center wants to place 400 calls in an hour period.
Workload = 400 calls x (135 + 32) / 3600 = 18.55 hours
Once workload has been calculated, it’s time to apply a staffing model to determine the number of agents needed to place the calls. Contact centers use an Erlang C formula to predict what happens in an inbound calling situation with random workload. However, in an outbound call center, the calls do not follow a random pattern, so Erlang C is not appropriate. Outbound calling generally represents sequential work, where calls can be placed in a back-to-back fashion. With this sequential work, each person can handle a full hour of work in an hour timeframe, so the number of staff needed would simply be the hours of work to do.
The simple model outlined here assumes there are multiple calls that can be made in blocks of time. Obviously, if these outbound calls are each generated by an inbound call, then the model would be closer to a random-calling Erlang C model.
While a simple 1:1 ratio can be used to determine staffing in a sequential work scenario, it does assume that the staff will work at 100% productivity. You’ll want to factor in some non-productive time that will happen throughout the hour. This is not done for inbound staffing as the random arrival of calls naturally provides some idle time between calls as staff wait for the next call to arrive. In back-to-back outbound calling, however, you must factor in this time.
Assuming 90% productivity, divide 18.55 by .90 to get the number of staff needed to handle the outbound calls. (18.55/.9 = 20.6 or 21 staff). The productivity factor will vary depending on the type of work done. If staff are placing outbound sales calls with the chance to earn a commission, they may not want much of a break between calls. On the other hand, if they are placing outbound collections calls, more idle time will be needed between calls to strategize or de-stress.
One of the ways that contact centers adjust to the peaks and valleys of inbound calling is to staff to the peaks and then find other work for the staff to do during the valleys. If outbound calling can be done anytime throughout the day, this is a great fill-in activity and can serve to even out resource utilization.
The ultimate in efficiency is to link a dialer to the ACD and when the dialer notes periods of inactivity and staff availability, it can place an outbound call and match up to an available agent. This is the most efficient use of staff, but it’s not always a practical approach. Some agents can’t deal with the true call by call mixture of inbound and outbound calling. Just because the technology makes it possible doesn’t mean you should do it.
Most centers choose to assign periods of time for outbound calls to be made, so each type of work is done in a block of time. In this arrangement, agents know what type of call they’ll be handling each hour. It also allows for specific staff to be scheduled and assigned to this type of work. Not all staff are interested or capable of doing the outbound skill, whether it’s sales, or collections, or simple proactive outbound service calls. Scheduling by skill ensures the right staff are assigned to the outbound calling.
While true inbound/outbound call blending can be a complicated staffing model, separating the two by hours or blocks of time makes each a simpler, more straightforward staffing problem.