Call Center Software Featured Article

Analytics in the Call Center

In the last five years, the role of analytics in call centers has advanced to the point where companies can better and more quickly identify patterns of call center agent and customer behavior.  In part, this is because CRM and call center data systems are becoming better integrated with sales and service performance management software, offering complete dashboard views of customers down to the last interaction, with the most recent call center agent notes attached.

The business value of recognizing these patterns accurately, consistently and early in most cases is significant: increased sales, reduced operational costs and employee churn and more satisfied customers. 

At stake are millions of dollars to the bottom line.  The call center for a major telecom handles 400-500 million calls annually, and a single call can cost upwards of $10. Reducing call volume, shortening call duration and removing the reason for customers to make repeat calls would cut these costs dramatically.

What are some of the behaviors that are being more closely watched in sales and service centers that offer the best options for improving service and reducing costs?

The most important behavior is customer call reasoning. Why do customers make the call in the first place? The implications upstream involve a host of potential changes to services.   A company might be able to direct customers away from the call center and to the web site for a number of common and frequent issues. New features on a product might be generating a large number of questions that can be addressed in a product fix. Agents can be retrained on how to quickly resolve queries and improve their own performance gaps.

One U.S. telecom recently used analytics on customer call reasoning to change their service offerings. Because a high number of calls that the company received were about billing issues, the company calculated that it would cost less to offer unlimited calling plans and make that entire category of call essentially disappear. This adjustment was made based on recognition of a pattern of past customer behavior, and it paid off for the telecom in a solution where everyone benefited.

Customer segmentation is another evolving trend impacting call center resource allocation, with companies adopting different approaches. Overall customers can be segmented according to their level of technical sophistication. Research has determined, for example, that a teenager posting to a company site on Facebook (News - Alert) is more likely to want tips from an IVR (voice recording) on how to self-serve, than on a live conversation with an agent.  A separate queue might be created for iPhone (News - Alert) or Blackberry callers, who have a discrete level of sophistication. However, a less digitally astute retiree with a basic phone would want more hand-holding by an agent. 

At a leading online bank well-known for its enviable customer loyalty, agents are trained to recognize four distinct personality types. Within a few seconds, call center agents can identify the type of caller and adopt the appropriate response strategy, leading to a positive experience for the customer. Is the caller in a rush? Do they want basic-level support, or do they have a grasp on the technology? The agents are measured on customer satisfaction, not the more traditional metric of length of time on the call.

About five years ago, the standard operating procedure in call center management meant a supervisor would collect and aggregate data on call volume, types of calls, and agent responses. The data would be presented to agents in weekly management meetings. By then it was a week old. Today, agents can use personal analytics dashboards to self-monitor against clear goals on call volume and resolution, and customer satisfaction. They can see the frequency and reason for repeat calls, search for team performance notes on types of calls and adapt better behaviors on the very same day.  The transparency provided by the dashboards are such a powerful force on agent performance that one large telecom was able to decrease the number of supervisors while allowing agents to see for themselves what customers say about them, and give them access to information on “the next best action” to adopt without intervention by a middle layer of management.

By applying analytics to the operations of call centers, companies can understand customer behaviors and respond with an empathetic approach that is almost always felt by the customer as a positive experience. The call center employee, in turn, enjoys the transparency analytics provide on how they performed and how to make it better.

The next step? Predictive analytics that can anticipate customer behaviors, call volumes per geography, and more precise customer segmentation to determine agent resource allocation. Coming soon at a call center you know.