Last year, Sprint discovered that the handle time at one of its more than 70 call centers had suddenly shot up by 25 seconds. Twenty-five seconds may not seem like a lot, but any call center manager can tell you to the penny just how much that one sixtieth of a minute costs the company.
What they may not be able to tell you is what exactly is behind an unexpected surge in handle time.
Sprint, however, had a system to stitch together data from multiple sources throughout the organization create a picture of the problem and figure out how to solve it. Using NICE Systems Contact Analytics, they were able to pinpoint a small group of new call center agents responsible. They were spending ten to twelve minutes walking customers through complete factory resets of their Blackberry devices—often more than once per customer.
Ultimately, Sprint worked with RIM, its smartphone vendor, as well as more seasoned call center agents to figure out a more effective approach and the customer hand-holding time dropped down to normal immediately. (Sprint declined to define the normal time for handling a call.)
Contact centers are one of the most data rich departments in the corporate environment, says Tony Filippone, executive vice president of research for sourcing analyst firm HfS Research. “Every contact is counted, routed, measured, and scored,” Filippone says. But call centers are only just beginning to incorporate big data—large volumes of structured and unstructured data from across the company—into their management and strategy.
“There is a lot of data—and a lot of messy data,” says Jason Pointelin, manager of Sprint’s analytics and performance team. “Everyone thinks you can put the puzzle together quickly. But it takes a lot of legwork to integrate data from different systems.”
Data About One Call Culled from Multiple Systems
Sprint currently sends transaction data from billing systems, voice data from their calling systems, agent information from their performance management system, quality data from their customer surveys, and process data from their knowledge management systems to NICE Systems every night where the information is crunched in the cloud. “They tie it all together so for every call we have a whole lot of information and we know everything about the customer and everything about the agent, so we understand what happened and [we can] make better decisions going forward,” says Pointelin.
That helps call center leaders not only better measure and improve individual agent performance, but also corporate systems and processes. “Certain processes take too long. Certain call types cause more pain,” Pointelin says. “This gives us a cold, hard fact-based view.”
For example, a group that specifically handles calls with new customers or telephone service requests from customers with new handsets had a repeat call rate that was unusually high. By analyzing the data on those calls, they discovered that most of the callbacks were related to old device activation. “I got a brand new, shiny iPhone 5 and I want to give my dad my old phone,” says Pointelin. It was an easy fix—add a question at the end of the call asking the customer if they would like to activate their old phone for someone else or recycle it—that reduced repeat calls and overall handle time and improved customers satisfaction.
That’s a key success factor for big data analytics in the call center. “The analytics have to drive action, not just insight,” says Filippone. “It isn’t enough to count issues and score sentiment. New solutions have driven agent behavior change or transformed CRM strategies.”
Since 2008, Sprint has risen from last among carriers in the American Customer Satisfaction Index to first, while cutting the budget for its call centers in half. Customer satisfaction and first-call resolution metric have both improved by more than 30 percent. The company has also reduced the amount of credits or adjustments it offers customers for mistakes or problems. “The other side benefit is that when the customer is happier, the agents are happier,” says Pointelin. “They’re not on those seven-minute phone calls where the customer is reliving in great detail what happened in all the previous phone calls.”
Sprint is now working with NICE to do it all faster, better—and potentially—cheaper. NICE is using Sprint call centers as a test environment for its new big data offering which incorporates IBM’s big data analytics software, including InfoSphere BigInsights, to collect, integrate and analyze larger, more complex datasets more quickly, potentially impacting customer interaction in real-time. “We’ve been talking to NICE for a while about it,” says Pointelin. “We’re going to use our data to see what the benefits of the new system might be.”
Stephanie Overby is a Boston-based freelance writer. Follow her on Twitter: @stephanieoverby.