The great American philosopher Yogi Berra once said, “If you don’t know where you’re going, chances are you will end up somewhere else.” Yet many utilities possess only a limited understanding of their call center operations, which can prevent them from reaching the ultimate goal: improving performance and customer satisfaction, and reducing costs.

Utilities face three key barriers in seeking to improve their call center operations:

  • Call centers routinely collect data on “average” performance, such as average handle time, average speed of answer and average hold time, without delving into the details behind the averages. The risk is that instances of poor and exemplary performance alike are not revealed by such averages.
  • Call centers typically perform quality reviews on less than one-half percent of calls received. Poor performance by individual employees – and perhaps the overall call center – can thus be masked by insufficient data.
  • Calls centers often fail to periodically review their processes. When they do, they frequently lack statistically valid data to perform the reviews. Without detailed knowledge of call center processes, utilities are unlikely to recognize and correct problems.

There are, however, proven methods for overcoming these problems. We advocate a three-step process designed to achieve more effective and efficient call center operations: collect sufficient data; analyze the data; and review and monitor progress on an ongoing basis.

STEP 1: COLLECT SUFFICIENT DATA

The ideal sampling size is 1,000 randomly selected calls. This size call sample typically provides results that are accurate +/- 3 percent, with a more than 90 percent degree of confidence. These are typical levels of accuracy and confidence that businesses require before they are likely to undertake action.

The types of data that should be collected from each call include:

  • Call type, such as new service, emergency, bill payment or high bill, and subcall type.
  • Number of systems and/or screens used – for example, how many screens did it take to complete a new service request?
  • Actions taken during the call, such as greeting the customer, gathering customer- identity data, understanding the problem or delivering the solution.
  • Actions taken after the call – for example, entering data into computer systems, or sending notes or Emails to the customer or contact center colleagues.

Having the right tool can greatly facilitate data collection. For example, the call center data collection tool pictured in Figure 1 captures this information quickly and easily, using three push-button timers that enable accurate data collection.

When a call is being reviewed, the analyst pushes the green buttons to indicate which of 12 different steps within a call sequence is occurring. The steps include greeting, hold and transfer, among others. Similarly, the yellow buttons enable the analyst to collect the time elapsed for each of 15 different screens that may be used and up to 15 actions taken after the call is finished.

This analysis resembles a traditional “time and motion” study, because in many ways it is just that. But the difference here is that we can use new automated tools, such as the voice and screen capture tools and data collector shown, as well as new approaches, to gain new insights.

The data capture tool also enables the analyst to collect up to 100 additional pieces of data, including the “secondary and tertiary call type.” (As an example, a credit call may be the primary call type, a budget billing the secondary call type and a customer in arrears the tertiary call type.) The tool also lets the analyst use drop-down boxes to quickly collect data on transfers, hold time, mistakes made and opportunities noted.

Moreover, this process can be executed quickly. In our experience, it takes four trained employees five days to gather data on 1,000 calls.

STEP 2: ANALYZE THE DATA

Having collected this large amount of data, how do you use the information to reduce costs and improve customer and employee satisfaction? Again, having the right tool enables analysts to easily generate statistics and graphs from the collected data. Figure 2 shows the type of report that can be generated based on the recommended data collection.

The analytic value of Figure 2 is that it addresses the fact that most call center reports focus on “averages” and thus fail to reveal other important details. Figure 2 shows the 1,000 calls by call-handle time. Note that the “average” call took 4.65 minutes; however, many calls took a minute or less, and a disturbingly large number of calls took well over 11 minutes.

Using the captured data, utilities can then analyze what causes problem calls. In this example, we analyzed 5 percent of the calls (49 in total) and identified several problems:

  • Customer service representatives (CSRs) were taking calls for which they were inadequately trained, causing high hold times and inordinately large screen usage numbers.
  • IT systems were slow on one particular call type.
  • There were no procedures in place to intercede when an employee took more than a specified number of minutes to complete a call.
  • Procedures were laborious, due to Public Utilities Commission (PUC) regulations or – more likely – internally mandated rules.

This kind of analysis, which we describe as a “longest call” review, typically helps identify problems that can be resolved at minimal cost. In fact, our experience in utility and other call centers confirms that this kind of analysis often allows companies to cut call-handle time by 10 to 15 seconds.

It’s important to understand what 10 to 15 fewer seconds of call-handle time means to the call center – and, most importantly, to customers. For a typical utility call center with 200 or more CSRs, the shorter handle time can result in a 5 percent cost reduction, or roughly $1 million annually. Companies that can comprehend the economic value and customer satisfaction associated with reducing average handle time, even by one second, are likely to be better focused on solving problems and prioritizing solutions.

Surprisingly, the longest 5 percent of calls typically represent nearly 15 percent of the total call center handle time, representing a mother lode of opportunity for improvement.

Another important benefit that can result from this detailed examination of call center sampling data involves looking at hold time. A sample hold time analysis graph is pictured in Figure 3.

Excessive hold times tend to be caused by bad call routing, lengthy notes on file, unclear processes and customer issues. Each of these problems has a solution, usually low-cost and easily implemented. Most importantly, the value of each action is quantified and understood, based on the data collected.

Other useful questions to ask include:

  • What are the details behind the high average after-call work (ACW) time? How does this affect your call center costs?
  • How would it help budget discussions with IT if you knew the impact of such things as inefficient call routing, poor integrated voice response (IVR) scripts or low screen pop percentages?
  • What analyses can you perform to understand how you should improve training courses and focus your quality review efforts?

The output of these analyses can prove invaluable in budget discussions and in prioritizing improvement efforts, and is also useful in communicating proposals to senior management, CSRs, quality review staff, customers and external organizations. The data can also be the starting point for a Six Sigma review.

Utilities can frequently achieve a 20 percent cost reduction by collecting the right data and analyzing it at a sufficiently granular level. Following is a breakdown of the potential savings:

  • Three percent savings can be achieved by reducing longest calls by 10 seconds.
  • Five percent savings can be gained by reducing ACW by 15 seconds.
  • Five percent savings can be realized by improving call routing – usually by aligning CSR skills required with CSR skills available – by 15 seconds.
  • Three percent savings can be achieved by improving process for two frequent processes by 10 seconds each.
  • Three percent savings can be realized by improving IVR and screen pop frequency and quality of information by 10 seconds.
  • One percent savings can be gained by improving IT response time on selected screens by three seconds.

STEP 3: REVIEW AND MONITOR PROGRESS ON AN ONGOING BASIS

Although this white paper focuses on the data collection and analyses procedures used, the key difference in this approach is the optimization strategy behind it.

The two-step approach outlined above starts with utilities recognizing that improvement opportunities exist, understanding the value of detailed data in identifying these opportunities and enabling the data collected to be easily presented and reviewed. Taken as a whole, this process can produce prioritized, high-ROI recommendations.

To gain the full value of this approach, utilities should do the following:

  • Engage the quality review team, trainers, supervisors and CSRs in the review process;
  • Expand the focus of the quality review team from looking only at individual CSRs’ performance to looking at organizational processes as well;
  • Have trainers embed the new lessons learned in training classes;
  • Encourage supervisors to reinforce lessons learned in team meetings and one-on-one coaching; and
  • Require CSRs to identify issues that can be studied in future reviews and follow the lessons learned.

Leading organizations perform these reviews periodically, building on their understanding of their call centers’ current status and using that understanding to formulate actions for future improvement.

Once the first study is complete, utilities also have a benchmark to which results from future studies can be compared. The value of having these prior analyses should be obvious in each succeeding review, as hold times decline, average handle times decrease, calls are routed more frequently to the properly skilled person and IT investments made based on ROI analyses begin to yield benefits.

Beyond these savings, customer and employee satisfaction should increase. When a call is routed to the CSR with the requisite skills needed to handle it, both the customer and the CSR are happier. Customer and CSR frustration will also be reduced when there are clear procedures to escalate calls, and IT systems fail less frequently.

IMPLEMENTING A CALL CENTER REVIEW

Although there are some commonalities in improving utilities’ call center performance, there are always unique findings specific to a given call center that help define the nature and volume of opportunities, as well as help chart the path to improvement.

By realizing that benefit opportunities exist and applying the process steps described above, and by using appropriate tools to reduce costs and improve customer and CSR satisfaction, utilities have the opportunity to transform the effectiveness of their call centers.

Perhaps we should end with another quote from Yogi: “The future ain’t what it used to be.” In fact, for utilities that implement these steps, the future will likely be much better.