[April 20, 2005]

What Kind of Goal Is Service Level Anyway?

BY BILL DURR

Service level, the percent of calls answered within some number of seconds, is perhaps the best-known and most widely used metric in the contact center industry. While there may be some slight variations in how it is calculated – do we count quick abandoned calls or not – everyone understands what the metric means and how to interpret it … or do we?

Here’s something to consider: What kind of goal is service level? Is it like profit? Profit is the goal of every enterprise. On the other hand, perhaps service level is more like a tire’s air pressure. Tire manufacturers specify the air pressure required to inflate the tire properly. Having less air pressure creates safety issues, but like-wise, over-inflation also creates the same concerns. Unlike profit, a tire’s air pressure needs to be managed so as to avoid having too much or too little.

Most centers operate as if service level is a goal identical to corporate profit – more is better and less is not. If the goal is 80 percent in 20 seconds, the operations team is overjoyed to make 90 percent in 20 seconds. However, when the goal is not being met, it goes into crisis management mode. Schedules are re-run. Agents are shuffled around. Coaching and training is halted. Disruption ripples throughout the center as all energies focus to return the service level to the target.

We understand why operations teams get edgy when the center is running below the service level goal. A queue that is too long is one of the major sources of frustration that customers have with contact centers today. And increasing your customers’ frustration can’t be good for business. But we believe that service level is a goal, more like tire air pressure than profit. Operations management ought to be just as concerned when service level is above the goal as it is when it falls beneath the goal.

For any given amount of work, specifying the service level goal automatically specifies how much labor expense the center will require. Operations teams tend to pad the agent requirements, so as to avoid underachieving the service level goal. The pad is used to counter the negative consequences of minor variations to the plan – such as the center getting more calls than expected, calls taking longer, some agents not adhering to schedules and so forth. If the day unfolds without encountering any of these minor variations, the service level target is surpassed, unless the operations team decides to offer time-off without pay or pulls agents off the phones to perform project and back-office work. While some tightly run centers do get concerned when service level is above target, most don’t. But they should, and here’s why …

A service level goal represents a compromise between economic benefit and expense. The reason we answer calls is because the enterprise receives some economic benefit. Unless you got your service level goal in a box of Cracker Jacks, or actually thought there was such a thing as an industry standard, someone in the company decided that your service level goal represented the best solution to the benefit/expense equation. If the center is over-achieving the service level goal, it is likely they are spending more than they need to. This only makes sense if the benefit the enterprise receives also increases. That said, except in rare instances, we almost never see proof that a service level of 90 percent in 20 seconds delivers any incremental economic value whatsoever to the enterprise beyond what a service level of 80 percent in 20 seconds brings.

If there isn’t an economic benefit to the enterprise when the contact center exceeds the service level goal, why do center management teams push to accomplish this? Perhaps it’s a consequence of misunderstanding the nature of the goal coupled with a natural competitive spirit. For many goals in the corporate world and in life, exceeding the goal is desirable and positive.

The problem with service level goals is that the link between them and corporate well-being is tenuous at best. We have argued in the past that centers ought to experiment with service level in order to understand that linkage. We suspect that every organization will discover a different co-efficient between service level and corporate profit. Some will discover that the linkage is linear. Some will find that the linkage is geometric. And others will discover that the linkage is nonexistent. Knowing which pertains to your organization is key.

Strategic consulting services are often employed by centers to help them determine what their service level ought to be. Contact center consultants understand that service level goals should never be arrived at in an arbitrary fashion. It is intellectually lazy to suggest that your center’s service level should be aligned with an illusory industry standard. Didn’t mom warn us about blindly following the pack? Would you jump off the pier just because everyone else was? Yet, this is the state of our industry.

Don’t misunderstand the point. Benchmarks are extremely important. They inform us about how other organizations view the cost/benefit analysis associated with their particular market and competitive situation. But it could be a mistake to accept what other firms are doing as the standard or template that you should emulate. Rather, use benchmarking as a guide – another point of reference. If your center is ever going to assume strategic importance to the enterprise, you need to determine what the appropriate service level goal is for your company in your market given your unique competitive environment.

Bill Durr is Principal Solutions Consultant for Witness Systems, provider of workforce management software and services.