Principles for Contact Center Productivity Measures

Apr 15, 2009 7:45 PM  By

The first thing your contact center management team needs to do when sitting down to create an agent productivity metric is agree in principle about what this is all about. Following are four principles we believe will make your journey on the road to productivity a safer one.

Principle 1: A productivity metric needs to reflect a leader’s understanding of what impacts productivity.
We are all responsible for how we use our time. Performance measures should be designed to encourage people to take that responsibility. The measure—along with the rest of the performance management infrastructure—should be constructed to indicate that management understands that there are events in every job that people cannot control.

The metrics should indicate to the agents that you know they can control how they use their time. But you also want them to know that you understand there are certain things that happen throughout the day that they cannot control.

For example, agents can control, to a certain extent, how much time they spend in after-call work. The agents’ skill level contributes to the length of after-call work— those more skilled can complete the tasks faster.

But the complexity of the call can also determine the amount of time spent in after call work time. The more complex calls typically take longer.

And, let’s suppose an agent needs to regain his composure after a particularly trying call. That agent can choose to take a few extra seconds in after call work time to regroup before taking the next call.

Agents are not in control of the time they spend waiting for a next call. This is determined by call volume, staffing levels, and the workforce management team.

The productivity metric needs to reflect a leader’s understanding of all these subtle time utilization possibilities.

Principle 2: The same measurement for everyone doing the same job.
For ease of measurement and setting consistent expectations, your productivity measurement should be the same for all people doing the same type of work. The metric should allow you to compare any agent to any other agent no matter what time of day he or she is working.

This makes it easier to roll the metric into a center-wide performance metric. The goal of all good measurement systems is to have the metrics in one part of the organization relate to the metrics throughout the organization.

Let’s say you are using “calls per hour” as your productivity metric. If you calculate productivity by simply taking the number of calls offered and divide it by the number of hours worked, then your night shift can never reach the same standard as your day shift. The night shift agent usually has fewer calls offered per person than the day shift (thus leading to lower calls per hour).

Your job is to determine how to accommodate any such inequity and still utilize the same measurement and standard for all.

Principle 3: One metric—different standards.
There are times when you’ll need to have different performance standards. Let’s pretend you have two agent groups, one of which handles a call type with a much longer average handled time. You can still use the same measure for productivity.

But you can adapt the metric to this longer talk time by setting up a different performance standard. So, if your productivity measurement were “calls per hour,” the group with the longer handled time would have a lower performance standard. By definition, this group cannot complete as many calls in the same number of hours since the handled time is longer.

Principle 4: Fairness
The most important principle is fairness. There are many things that can happen throughout the week that can affect an agent’s ability to produce.

What if you introduce a new process that is quite complex? It may take the agent a while to get up to speed.

Or, in an outbound telemarketing environment, the list quality and campaign requirements can cause a varying degree of productivity. The management team must always be on the lookout for factors of system variability and adjust the standard for fairness.