Keeping Score - The Right Way

By Bryan DiGiorgio

March 2010

"How's business?"  That's the question forever on the minds of executives and business owners.  What is the true state of my business?  For those of us working in the call center environment, the answer might seem readily available - simply measure the performance of each center.  But how can we know that we are measuring the right things?  Do the metrics reported by individual centers offer an accurate reflection of their operations?  Or are the performance reports weighed down by scores of miscellaneous data points that diffuse the focus rather than pinpoint the areas of concern?

It's true that measurements serve a vital role in assessing effectiveness and highlighting potential points of failure.  However, if you aren't tracking the right metrics, prioritized in the right order of importance, you could be failing to spot performance issues, defocusing what matters most, or even missing out on additional revenues or cost savings.  Therefore, instead of relying on a massive laundry list of metrics and reports, I recommend approaching measurement, site efficiency, and overall success with this three-step strategy:

Ensure Clarity around Your Goals and Prioritize Your Metrics: Regardless the type of call center or queue (i.e., inbound sales, customer support, or tech support), there are some standard measurements that apply across all lines of business: average handle time (AHT), average speed of answer, cost per call, call quality, and customer satisfaction, to name a few.  However, there are also measures that are equally important or even more so for discrete queues.  For example, what if your call center is primarily a technical support queue?  In that instance, while the standard metrics are important, I would propose that the most important metric is issue resolution - this single metric indicates whether your call center is solving a caller's questions. 

Issue resolution is a single, clear metric that describes the business - and customer - purpose of the queue.  Think about it: if you don't solve the caller's issue, what else matters?  All other metrics become meaningless.  Customers will either call back - doubling your costs - or simply stop being customers.  Moreover, if you typically up-sell in this queue, it makes sense to up-sell only after the issue is resolved.  Therefore, when you increase issue resolution, you increase up-sell opportunities.  This single metric can also be a leading indicator of your ability to maintain customer loyalty and reduce customer turnover or "churn."

This does not mean that other metrics are not important - they are.  Nor does this mean that you should not address other outliers.  However, no action should negatively affect the primary metric.  Consider placing risk and reward on this single metric; then performance-manage the rest.  The level of clarity this approach provides will have a dramatic impact on your business results.  This leads us to the next part of the approach.

Examine Metrics from a Holistic Viewpoint: The customer management function is an interdependent one.  Every transaction either supports or furthers another transaction; every stage of the continuum is equally important.  If you focus on an individual transaction without considering the downstream repercussions, you will find yourself trying to improve one aspect of your operation only to see another problem spring up down the line.

How do you avoid the exhaustion - and exasperation - of this approach?  I believe that you can only do so by embracing a centralized point of accountability, using the "hub-and-spoke" model.  With this approach, the hub acts as the centralized support structure of the people, processes, and technologies that drive customer management effort; the spokes are the contact and fulfillment centers.

The best model for the hub is one that includes centralized management functions such as intraday performance management, root cause analysis, integrated reporting, statistical modeling, process mapping, knowledge management, curriculum development, and project management. 

When operating at full potential, a well-run hub will result in fewer variations in your metrics.  Additionally, it will improve other support metrics such as quality, staff utilization, the ability to quickly adjust staffing to react to call volumes, and expenses. 

While many companies generally perceive themselves as operating a "hub-and-spoke" (even in a single center environment), few organizations operate an effective one.  Sometimes "guesswork" and assumptions may substitute for root cause analysis.  For example, you may know you receive x amount of technical support calls, and you may know the top five products causing your issues.  However, if you cannot tell someone specifically what is broken, the likelihood of fixing the root problem is generally low.  In most cases, process mapping is not well documented and routine audits/assessments are not completed.  If you do not have processes documented, how do you know how long the AHT should be?  How do you determine what processes (steps) are used to evaluate and reduce AHT? 

Integrated, centralized reporting is typically cobbled together from individual reporting at an aggregate level - not from true centralized management reporting.  This nuance will prevent you from correlating a delay in fulfillment to the effect on call volume or accurately assessing the negative impact on customer retention.  Management oversight and the commitment to holistic measurement is an instrumental part of achieving your goals.  The lack of oversight will result in poor performance by the overall customer management system.  Like an all-star team where players are selected for individual performance, the function will never work to its full potential unless it operates as a cohesive, interdependent unit.

Establish Role Clarity and Capability Assessment Procedures: One of the greatest obstacles to success in any business is a lack of role clarity, particularly when portions of the customer management continuum are outsourced to a number of different transactional suppliers.  In these cases, there may be an unreal expectation that the call center vendor can provide in-depth analysis to improve performance issues that have been identified. 

Let's be clear: outsource call centers excel at managing agents and following a prescribed process, but they are not typically experts in root cause analysis and process improvement.  That's why the hub-and-spoke model is so appropriate - it places the responsibility of applying Six Sigma quality assurance processes squarely within the boundaries of the client-side "hub" team and allows the call centers to focus on the agents and data collection.

I counsel many client-side call center executives to devote a significant portion of their organizational design process to defining roles, assigning responsibilities, and associating hand-offs for data collection and analysis.  Otherwise, like a major-league slugger in the midst of a batting slump, you'll end up with low scores without knowing which changes will result in the greatest improvements.

In an environment of frequent reports and abundant data, call center performance problems should be easy to spot.  However, there are many hidden pitfalls in the form of nonessential data or metrics that, rather than clarifying issues, obscure the true business drivers.  As a call center executive, you want to maintain a laser focus on the true indicators of success, tuning out the white noise.  You have to get everyone to agree to the rules of the game so you all end up with the same score.

Bryan DiGiorgio is president and CEO of CXO Global Solutions, a Kansas City-based company specializing in customer experience management.  A twenty-year customer management veteran, Bryan has extensive experience within the automotive, consumer services, financial services, and retail and telecommunications sectors.