Convergys Corp., a Cincinnati-based customer service outsourcer with more than 60 call centers around the world, recently started using a quality management system, from Nice Systems Ltd. of Israel, that can detect the emotional states of both customer and representative. “It’s not just what you’re telling the customer,” says Bob Lyons, senior vice president of global services at Convergys. “It’s how you’re telling them. That ‘softer side’ contributes hugely to overall customer satisfaction.”
By analyzing both the customer’s and agent’s voices—all within the first 10 to 15 seconds of a call—the software can flag the call for review based on changes in pitch, tone, cadence and speed. The system also records how long the customer spent on hold, and the information that was available to the agent on his or her computer screen during the call. Each call is digitized and stored, making it keyword- and emotion-searchable.
Convergys uses the emotion data to assist with accent neutralization in Indian call centers, as well as to help HR departments identify the kinds of personalities best suited to customer service. “Emotion detection has been especially useful for employee coaching and retention,” says Lyons. “We create scorecards for employees to show them, based on their call histories, where they’re succeeding and where they could use some work in their approach to customers.” Convergys is planning to make the software standard at all its call-center locations.
For now, the system’s purpose is limited to improving the quality of call-center service. The sort of real-time analytics that could allow irate phone callers to be transferred directly to supervisors, for instance, is a few years away. Still, Lyons says the system has helped Convergys deal effectively with thousands of the calls they receive each day. “The customer-service agents like to see how they’re doing,” he says, “and based on the data we’re collecting, we can recruit the types of people who will do the best in our centers.”