Analytics Converting Call Centers into Insight Centers

by   |   October 16, 2012 6:44 pm   |   1 Comments

For call centers, “handle time” – how long it takes to deal with a customer’s issue – is money. So it was a significant problem when Sprint found out that that one of its 45 call centers had a handle time that was four times longer than average.

Related Stories

Inside Wharton’s research center for customer analytics.

       Read more»

The potential and challenges of workforce analytics.

       Read more»

Normally, in an operation with tens of thousands of agents spread across the country, such a costly problem could lurk unaddressed for weeks. Sprint was able to pinpoint and correct the issue – bringing the handle time back to normal – within hours. The reason was traced to a group of new agents who were struggling with calls about a new device.

The fast action is a direct result of the company’s sophisticated use of call-center analytics, which ties together information from phone calls, billing systems, customer surveys, and other systems. “And by integrating analytics with performance management and coaching, you have actionable information,” says Jason Pointelin, manager of analytics and performance management systems at Sprint.

Call Center is Data “Nerve Center”
As people grapple with the often-confusing and frightening buzzword that is “big data,” many are looking to call centers as ground zero. Call centers are the nerve center of companies’ intelligence gathering efforts, where data can instantly translate into profit-enhancing customer service. “Call centers are used to working with large amounts of data and managing people with data,” Pointelin says.

Tips for Turning Call Centers into Insight Centers

  • Think proactive rather than reactive: Move beyond “workforce optimization” to use analytics to identify customer insights.
  • Different channels—email, phone calls, customer surveys—provide different types of insight that, wedded together, give a more complete picture of the customer interactions.
  • Integrate data from multiple systems, and then tie that data back to coaching systems to make the data actionable.
  • Consider new call center analytics tools to guide agents while they are interacting with customers—though realize such tools are embryonic and imperfect.

While social media and online communities have changed the sobriquet of the “call center” to the “contact center,” voice still reigns supreme. The Aberdeen Group’s September 2012 “Speech Analytics: Listen to your Customers” benchmark report found that nearly two-thirds of interactions in contact centers involve voice — phone and/or Interactive Voice Response (IVR) conversations. Currently, approximately half of all contact centers are using speech analytics tools to analyze voice conversations. However, this number is expected to rise rapidly over the next few months as companies are grasping the importance of voice within their broader multi-channel customer conversations.

Ninety-five percent of companies that use speech analytics study the calls after the fact, Aberdeen found. There has been a growing realization that this treasure trove of data can be put to new uses.

“In the past, the speech analytics was used for workforce optimization, such as to reduce repeat calls and customer frustration,” says Daniel Ziv, vice president, voice of the customer analytics at Verint, an analytics software vendor. “However, companies realize that the call center can be a place where customers express their frustration before they speak out on social media. So now, it’s evolved to use analytics to look for broader, emerging trends. This is rich data to correlate what our detractors or promoters are telling us.”

He points to one of his customers, Elavon, a merchant payment processing firm, which used speech analytics to identify the reasons for customer churn. The 300 agents in the Elavon’s Knoxville, Tennessee, contact center take up to 15,000 calls a day, all of which are recorded and transcribed. The company analyzed the chain of calls to determine words and phrases that hinted a customer could be on the path to switching to a competitor.

As a result, Elavon identified 2,000 calls a day that indicated a potential issue. By contacting those at-risk customers immediately, the company figures over a three-month period it saved nearly 600 accounts worth $1.7 million.

As this suggests, analytics can allow a contact center to become proactive rather than reactive in dealing with problems and finding opportunities. For example, a cable TV provider that realizes customer calls about receiving service on mobile devices have doubled in the past year has important insight on the direction of its business. Such information can suggest new products and services.

Aberdeen’s research separated out speech analytics “leaders,” which was marked by their first-call resolution and other important metrics of call center success. Aberdeen found that 68 percent of the leading businesses use speech analytics tools to identify emerging business trends, compared to 38 percent of companies that are trailing top performing organizations.

The Moment of Truth
Omer Minkara, a research analyst studying contact centers with Aberdeen Group, says the true value of analytics will be maximized when the integration of multi-channel information is coupled with the agents’ ability to use these insight in “the moment of the truth” – when they need to serve the customers. “You need to empower the agent with all the information on their desktop to see what the customer said about the brand pre- and post-interaction,” he says.

A smattering of companies is using speech analytics in real-time now. For example, if a customer uses the words “cancel” and “service” in the same sentence during the call, an analytics system would prompt the agent to provide a special incentive to reduce the likelihood of cancellation.

“This is the early stage for using analytics in real time,” Ziv says. “There are a lot of issues. For example, if you are calling from a noisy environment, the keyword can be miscategorized.”

How Call Center Standouts Use Analytics

Leaders in the field of call center analytics demonstrate better performance in the following areas, according to researchers at Aberdeen Group:

  • Process: Monitor speech content to determine keywords associated with customer satisfaction and dissatisfaction.
  • Organization: Control agents’ use of greetings and closings and ensure they comply with company policies.
  • Knowledge: Identify business trends through the use of speech analytics tools.

Even post-call analytics are not perfect.  While 60 percent of contact centers with speech analytics deployments are satisfied with their ability to capture the voice of the customer, Minkara’s research found that about one third of companies that use speech analytics are challenged to effectively use this technology to listen to and engage their customers.

Pointelin says the trick – and “and the next big thing” for call center analytics – is integration. “For a few years, there have been integration solutions that tie together your speech analytics, text analytics, customer survey, and performance management in one system,” he says.

He notes, however, in the past these systems were only available from niche vendors – “a couple of guys in a garage.” Now larger players have begun to gobble up the niche vendors, which will make the technology more widely accessible.

In roughly the last 12 months, for example, NICE Systems has purchased Merced Systems (performance management), IEX (workflow management), and Fizzback (customer surveys). Verint has also been an active buyer, grabbing up Vovici   to strengthen its position in customer feedback management and GMT Corp. to boost its enterprise-wide workforce management. As technology heavyweights ranging from Oracle to Avaya to Siemens announced deals last year, Ventana Research analyst Richard Snow noted “the contact center is undergoing more change now than I have seen in 15 years.”

The increasing functionality gives companies better ways to plan customer-facing strategies. “By tying together our data to understand every interaction, we’ve unlocked stuff we never had access to before,” Pointelin says.

Pointelin stresses this only happens if the technology is comfortable for agents and managers. “I get excited about data and analytics, but my customers are people in the field who want simplicity,” he says. “We’ve found if the system is too difficult, people won’t use it.”

Joe Mullich, a freelance writer based in Los Angeles, can be reached at

Tags: ,

One Comment

  1. Mukesh Singh
    Posted November 3, 2014 at 5:58 am | Permalink

    Nice description of contact center analytics.

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>