Be Careful What You Measure

By Peter DeHaan

January/February 2007

"My labor percentage is down to 28%," boasts the owner of a mid-sized call center.  "Have you ever heard of anyone lower?" he continues, as though seeking affirmation.

"Twenty-eight percent!" I exclaim. 

"So, it's good?" he probes, still seeking validation.

I carefully consider my response.  "No, it's terrible!" I pronounce. 

"Terrible?" he asks incredulously.

"Yes, terrible," I confirm.  "What are you going to do to fix it?"

"Fix it?" he responds, dumbfounded.  "I worked hard to get down to 28%.  I can't get it any lower."

"You need it to be higher," I state matter-of-factly.  After pausing for dramatic effect, I add, "By stating that your labor is at 28%, you are implying that your overhead is at 72%.  That is way too high and needs to be much lower.  I fear that your call center is suffering from an inflated overhead, and it needs to immediately be brought under control."

There is a slight glimmer of resigned comprehension in his eye, but wanting to avoid any additional challenges to his entrepreneurial equilibrium, he mumbles a reluctant thanks and makes a hasty retreat.

I didn't actually say those things - but I sure have wanted to!  The truth is that when call center managers and owners focus on labor percentages, they are often looking at the wrong thing for the wrong reason.

Yes, it is correct that unchecked labor costs can quickly escalate, threatening to run out of control.  As such, skyrocketing agent expense is the most likely cause of the fiscal demise of a call center.  On the other hand, the too aggressive reduction of labor costs is the most likely cause of the quality demise of a call center.  Therefore, a requisite balance exists between cost and quality that is all too often out of balance, sacrificing customer service on the altar of cost-containment.

The Situation: For the sake of illustration, let's manufacture a fictitious, yet nonetheless realistic, situation for a typical call center.  To keep the math easy, we will assume that the call center has annual expenses of $1 million and spends 50% of its budget on call center labor.  For simplicity's sake, we will lump everything else into the broad category of overhead.  This assumption is not unjustifiable, as it is a call center's labor that directly provides the service, and everything else, albeit important, is ancillary or indirect.  In summary, the call center's financial picture looks like this:

Expenses:     $1,000,000      100%

Overhead          $500,000        50%

Labor:               $500,000        50%

So, we have a $1 million call center, spending $500,000 (50%) on labor and $500,000 on overhead.  To ensure profitability, it has been determined that overall costs need to be reduced by 10%.  Now what?

Scenario 1: According to conventional wisdom, you address the biggest cost area, which is agent labor.  Additionally, labor is also a variable expense, which means that it can be cut relatively easily (fixed expenses are much harder to scale back).  Plus, the effects of any labor adjustment can be recognized quickly, generally within a month (reductions in nonlabor related areas take much longer to materialize). 

In any schedule, there are some debatable details.  These include the number of agents required at certain times and the length of specific shifts.  Even by eliminating all of these disputable items, there is still considerable - and painful - cutting to do.  Eventually, the sagacious scheduler will be able to make the required cuts, resulting in the target reduction of 10%.  The annualized results look like this:

Expenses:        $900,000      100%

Overhead          $500,000        56%

Labor:               $400,000        44%

The targeted 10% cost reduction has been accomplished, profitability has been restored, and things are good, right?  Not necessarily so.  The 10% reduction was completely realized by cutting agent labor.  With overhead remaining unchanged, agent labor was actually subjected to a 20% reduction.  This will unarguably result in a noticeable drop in customer service levels, both measurably by the call center and perceptibly by the callers.  This will produce an increase in complaints, adding work for the supervisory and management staff, while further taxing the agents, who are now working harder than before.  It is also likely that some client defections will result, causing income to decrease and the newfound profits to evaporate.  This scenario could very well exemplify the old adage of "winning the battle, but losing the war."

To extend this scenario to reach the preceding and overly ambitious goal of 28% labor cost, agent labor would need to be reduced an additional $205,556!  This would result in:

Expenses:        $694,444      100%

Overhead          $500,000        72%

Labor:              $194,444        28%

Scenario 2: The prudent businessperson, however, will realize that the call center's carefully crafted agent schedule is essentially correct.  The agents (who are provisioned via the labor expense) are the primary determining factor in the quality of service offered and the resulting client satisfaction.  Once it has been determined that the agent schedule is on track, the cost reduction efforts can focus on overhead - that is, those activities that do not directly affect the provision of quality service.

It is important to note that while agent labor is a highly monitored and scrutinized effort, overhead, or non-agent labor areas, receive much less attention and on a considerably less frequent basis.  Therefore, these areas are much more likely to be inflated and merit reduction.  That is not to suggest that cutting overhead will be easy.  It won't; it will be difficult, especially since these reductions reside much closer to management.  Perhaps unneeded perks have crept into the budget; these can be axed without a detrimental effect on service.  Likewise, there are probably costs that are no longer necessary, but have continued unabated.  Other costs, left unchecked, have escalated over time and need to be trimmed back to a more reasonable and appropriate level.

Lastly, there is a labor component in the overhead category as well.  This applies to management (at all levels) and support personnel.  It could be that a certain position is no longer needed, but retained because everyone likes the person filling that position.  Other jobs could have become bloated with unnecessary effort or busywork that produces no real benefit to the call center.  Bureaucracy and self-preservation activities are also prime targets for elimination.  Finally, there is the possibility that complete departments or management levels might no longer be necessary, or at least warrant major reductions.

These types of cost reductions are not easy to make, and they are often harder to spot.  However, they can be made with the least impact on the callers - the very reason that the center exists.  In reducing costs by focusing on areas other than agent labor, the provision of service is not directly affected.  The annualized numbers become:

Expenses:        $900,000      100%

Overhead          $400,000        44%

Labor:               $500,000        56%

Another Extreme:  In the first scenario, we looked at reducing the labor percentage.  In cutting labor by $100,000 and then by another $205,000, a resultant 28% labor figure was realized.  There is, however, another way to accomplish this same result.  Quite simply, by holding agent labor constant and increasing overhead by $780,000, a 28% labor figure can also be achieved!

Expenses:     $1,785,714      100%

Overhead       $1,285,714        72%

Labor:              $500,000        28%

In conclusion, we see that there are two ways to reach a 28% labor figure: detrimentally slashing labor or obscenely increasing overhead.  However, in both cases the wrong target is being measured for the wrong reason.  Rather, the intent should be to establish an agent schedule that will produce the proper service level to callers and then shrink overhead to the minimal level.  This will effectively increase the labor percentage, while decreasing the overhead percentage - a right and worthy goal for any call center to measure.

To read other articles written by Peter DeHaan, go to From The Publisher or check out his blog, Musings of Peter DeHaan.  In addition to publishing Connections Magazine and AnswerStat magazine (for healthcare call centers), Peter also publishes several websites, including ArticleWeekly.com. He may be reached at 616-284-1305, dehaan@connectionsmagazine.com or the Peter DeHaan Publishing website.