Setting a Reasonable Adherence Goal

By Maggie Klenke

One of the more challenging parts of a workforce management strategy is setting a goal for schedule adherence that balances the needs of the customers, management, and the agents. A high expectation may serve both customers and management well but be perceived as unfair by agents. A goal with low expectations might give agents more breathing room but be expensive and result in unacceptable fluctuations in customer service. Finding the best balance must be based on the unique characteristics of each operation.

Definitions

Before setting the goal, a definition of “adherence” is needed. There are two primary elements of focus on whether agents are following their schedules and they go by different names by vendors and contact centers. In this article, the word “conformance” is defined as measuring whether the agent worked the total amount of time each day that was scheduled. For example, if the agent has a schedule that begins at 9AM and ends at 5PM, that is an 8-hour shift span. However, during that 8 hours, the agent will be scheduled for a half-hour lunch period and two 15-minute breaks. That leaves a total of 7 hours of time that the agent should be available for work. Even if the agent logs in 10 minutes late in the morning but goes home 10 minutes late in the afternoon, if the total work time is 7 hours, then conformance has been achieved.

The other element measures whether the agent worked the exact details of that schedule. For purposes of this article, this second element is defined as “adherence.” In this case, the agent who came in late and went home late is out of compliance with the details of the schedule. She may have left the center understaffed in the morning resulting in longer than acceptable queues for customers and heavier workload for her peers. In the afternoon, there may not have been enough work to keep her and the other staff busy when she was working during unscheduled time. This can result in poor productivity during this period and wasted payroll dollars.

Basis for the Schedule

The process of developing the schedule begins with a forecast of the amount of work that will be arriving in the contact center during each period of each day. This normally fluctuates with peaks of workload during some days and hours while others will have lighter loads. Determining the number of agents that need to be available to handle the work is also done on a period-by-period basis (typically 15-minutes) so that the speed at which contacts are handled meets management’s goal for acceptable customer service.

The next step is to determine the potential losses of productive time that will be experienced out of each agent’s paid time. This is referred to as “shrinkage.” Even though a full-time agent may be paid for 40 hours per week and 52 weeks per year, he cannot be expected to be at the desk handling work for that entire time. Some of the productive losses include:

• Sick time
• Vacation
• Training
• Meetings
• Coaching
• Breaks
• Non-call work (such as research, call-backs, emails, etc.)

When all this time is added up, the total shrinkage is typically 25-40% of the payroll time in North America but higher where the workday is shorter or more time-off is common (such as in Europe). It is higher in centers that do more off-phone activities, have tenured employees with more vacation allotments, deal with many changes and the associated training, or experience high levels of absenteeism (such as FMLA excused time). This means that the agents cannot be scheduled for the full 40 hours and 52 weeks. Some agents will not be on the schedule at all for a specific week, others will work only part of the time and be in training some periods, and some will be scheduled to be on the desk the entire week with only scheduled breaks.

Even those agents who are scheduled may be lost due to unplanned illness, tardiness, or just not logged in as planned. This loss can be 5 to 10% of the workforce on any given day. This loss must be built into the planned staffing so that when it does occur, the customers will not feel the impact of understaffing and wait longer for service than is acceptable. In addition, the impact on the staff who are working can be very little breathing time between contacts as fewer agents share the workload planned for more people.

Building the Schedule

With a reasonably accurate forecast of the workload and the expected losses, the scheduling process begins. This process starts with the hours of operation that must be covered, the mix of full-time and part-time staff available, and any work rules that must be applied. For example, these work rules might include the limitations on the length of a part-time schedule (such as no more than 20 hours per week), and minimum or maximum intervals between breaks.

The other major factor in the schedule building is the availability of the agents for specific time periods. Some may be willing and able to work anytime the center is open, but more commonly, agents may have limitations such as college class schedules, daycare availability, transportation issues, and other reasons that define availability. In addition, the center must consider the agents’ desires in terms of schedule to minimize dissatisfaction and turnover.

The scheduling process takes the workload for each period and the agents available under the limitations defined and makes the best match to ensure that the customers receive reasonable service and the agents are assigned to a reasonable schedule. Assignment of schedules to agents can follow a ranking system that defines the order of selection by preference. It is often done by seniority, performance results, or some combination of the two.

Setting the Conformance/Adherence Goals

Once the workload is defined and agent requirements are determined, shrinkage must be deducted from the available payroll time before the schedules are determined. Extra agents will need to be on the payroll to cover for these losses since every person cannot be scheduled for every day.

One of the elements of shrinkage is time that the agent is scheduled to be available but is simply not logged in as planned. It is not sick time, meetings, training, or other planned activities, but unplanned unavailability during the scheduled day. It can be tardy in the beginning of the shift, late returning from lunch, longer than planned break, etc. This unplanned loss during the day is the focus of both the conformance and the adherence goals. Remember that missing agents affect the speed of answer for customers and the level of occupancy experienced by the other agents. In addition, extra agents available when not scheduled impact payroll waste and can lower productivity for all agents.

This is one of the things that makes a contact center different from other kinds of work activity. It is a bit like a production line in that if one person is not doing their job along the line, the other workers need to pick up the extra work or the line productivity will be slowed. The goal to answer calls is typically set in seconds of delay so there is no opportunity to pick up the slack after hours or during low volume periods. It must be done when the customers demand it.

One of the elements of the workday that is uncontrollable is how long an individual contact might take to handle once it has begun. For example, if the average time to handle contacts is 5 minutes, a single contact might require 1 minute or even 10 minutes. The challenge comes when the agent has been scheduled for a break, lunch, or other off-phone time and gets stuck on one of the longer calls. If the call started at 3 minutes prior to the scheduled break time but lasts 8 minutes, the agent will be 5 minutes late going to break. This situation creates the biggest complaint of agents regarding adherence goals since they cannot cut the calls short just to log out on time. Therefore, the adherence goal must take this potential into account.

Assuming a 7.5-hour work schedule including two 15-minute breaks, the agent is expected to work 7 hours or 420 minutes during the shift. Assuming the agent is caught on a long contact at the beginning of every break and lunch and is delayed leaving at the end of the day, the total out of adherence time might be significant. Even though the likelihood of the long contact happening every one of these times is very low, it probably will happen a few times per week. Assuming 5 minutes late leaving for each incident, the total could be 20 minutes per day. If calls are longer or shorter, the losses would change, of course. In general, assuming half of an average handle time is a reasonable estimate of the impact of a long call delay.

The adherence goal is generally set as a percentage of the time scheduled. Using the 420-minute schedule, a goal of 95% adherence would allow for a loss of 5% of the day equal to 21 minutes or 1.75 hours per 5-day week without any penalty to the agents for non-adherence. Of course, this adherence goal should also consider “human needs” of agents that might require logging off during the day as well. The percentage calculation can be done for any situation in which the total scheduled time and the adherence forgiveness time has been determined. In general, the adherence goal is set for a period of a week or longer so that potentially bad days get averaged in with good days. The chart below shows some possible goals and their impacts based on a 7.5-hour (450 minutes) work schedule:

Communication of the goal and the reasons for it is critical to the process. Involving agent representatives in the discussions of goal setting and ensuring that all agents understand what the goal is, how it is measured, and why it is used will go a long way in ensuring that agents feel fairly treated.

Managing Schedule Exceptions

The purpose of setting a reasonable adherence goal is to build in some forgiveness for time that agents are not available unexpectedly. Some of that time will be serving customers while some may be just to give agents time to get a drink or visit the restroom between scheduled breaks.

Schedule exceptions are data entries into the schedule tracking system to account for (and forgive) time when the agent is out of adherence. For example, if a supervisor keeps the team longer than planned at a team meeting, schedule exceptions might be entered into each agent’s schedule to indicate that their schedule was changed beyond their control. That would prevent that time from counting against their adherence goals.

Since long call lost time has already been accounted for in the percentage, there should be no need to also enter a schedule exception into the agent’s daily results to account for any time a long call takes her past her scheduled time. In addition, when the agent takes the time to ask for the exception, the supervisor approves it and the workforce management analyst does the data entry, the impact of the few minutes of non-adherence can be significantly increased. Many centers have found that setting the adherence goal with this in mind can reduce workload for everyone and reduce stress across the center. Agents know they are being treated fairly and goals can be achieved without extraordinary efforts. When the agent is late arriving for work, stays out longer than scheduled for breaks, or simply does not log in as scheduled, no exception entry should be made as these are the out-of-adherence issues the goal is meant to track.

Some centers establish procedures to minimize adherence issues around these long calls. They may suggest to agents that they log off early for their breaks if they are within a set number of minutes of the time. Therefore, logging off 2 minutes early is encouraged to avoid being caught on a call that might take 10 minutes. This early log-off time is accounted for in the percentage goal, so no extra administrative effort is required.

It is not uncommon for centers to offer rewards for higher than minimum adherence results. This encourages agent to ask for added forgiveness for every minute possible. While exceptional achievement in many areas should be encouraged and rewarded, the impact in this case can be counterproductive.

Data entry for schedule exceptions for other types of issues are typically still required. If a team meeting runs long, a training class is canceled due to trainer illness, overtime, and early release times are still going to need to be entered into the schedules. These are reasonable and should not be counted against agent adherence goals. However, there should be a definition of how long an exception must be before it is worth the time for everyone to do the accounting, again assuming the forgiveness in the percentage will take care of small variations.

Conformance Goals

Conformance tracks the total time that the schedule expects the agent to log in and out of the workgroup. Even though the agent may have worked the scheduled 7 hours, if she was late 10 minutes and worked past the scheduled shift end 10 minutes, she is in conformance for the total work day, but she is out of adherence with the details of the schedule.

It may also track the specific skillset that the agent is supposed to be serving or the type of contact to be handled (such as calls, email, chats, etc.). Tracking whether each agent is working in the specific skill or work type takes the process to a higher level of detail. For example, if the agent is logged into the incoming call queue but is supposed to be in the chat queue at that time, the agent is out of conformance. If he is logged into the sales call queue and should be in the service call queue, he is working but not doing what is needed. Just like adherence, the result is understaffing in one case and overstaffing the another. Setting goals and tracking at only the conformance level may ensure an accurate payroll, but it may not produce the staffing when it is needed and the resulting consistency of customer wait times.

Summary

Adherence is an important measure of productivity in a contact center. The primary focus is to ensure that the center is receiving “a day’s work for a day’s pay” from each agent and that customers are served in a consistent and timely manner. Setting the goal requires analysis of the unique needs of each center (and perhaps for each unit within the same center). As with nearly every measure in contact centers, there is no “industry standard” that applies. Once the goals are set appropriately and effectively communicated, the management of the process can be minimized, and agents are more likely to feel fairly treated. The goal is to provide the best balance of service to customers while managing cost and maximizing agent satisfaction.

Maggie Klenke is one of the founders of The Call Center School (now retired) and an active industry consultant, assisting contact center clients in development of strategic and tactical plans, technology applications, forecasting and scheduling optimization, service level analysis, and overall management issues. Maggie teaches seminars on a wide variety of call center topics and is a popular speaker at industry conferences in the US and abroad.
She may be contacted at maggie.klenke@mindspring.com or 615-651-3324.