Analyzing recordings reveals critical BI about what customers are really saying.
Posted Aug 1, 2005
"This call may be monitored for quality assurance." We've all heard that familiar refrain. Every day businesses record tens of millions of hours of customer calls, yet the majority of this valuable store of information is completely ignored. The customer's voice remains a cry in the dark. Dedicating resources to live sampling is prohibitively expensive and the human capital required to listen to the calls makes manual listening impractical. This economic reality forces businesses to sample only a small portion of recordings. But, with the advent of call mining technologies, call centers need no longer ignore a single customer voice. Call mining technologies convert recorded calls into minable databases of text and data. When these are combined with industry-standard data mining and unstructured data analysis technologies, call centers now have a powerful tool to mine 100 percent of call content for business intelligence.
Uncover customer trends--what are they saying?
I recently purchased a PDA that turned out to have a defective memory card slot. At first it seemed that the card itself was defective, but after replacing the original memory card and finding out it didn't work either, it seemed that the device itself was the culprit. My next move was to pick up the phone and call the manufacturer. The customer service representative not only completely and accurately understood the problem; she apologized and sent a courier to pick up the PDA that day. The manufacturer overnighted me a new PDA.
That agent's actions not only lifted the cloud of frustration caused by my wasting hours tinkering with the device; they also engendered that priceless commodity--brand loyalty. Now, the electronics company that I spoke with records some customer interactions, but like most recorded calls, mine likely found its way into an archive that was ultimately deleted and forever lost. Had the electronics company employed call mining tools it would have learned valuable information about its relationship with me--that I owned several other devices from it, how I used those devices, and why they were important to me. Perhaps most critical, it would have learned that, if this interaction didn't go well, I was willing to move on to a competitor for a better solution. Only call mining can reach across all recordings to provide trends for call topics over time and yield effective, actionable information for the enterprise.
Call mining can be used to identify key topics that point to customer dissatisfaction, like naming a competitor or asking to speak to a supervisor. Moreover, since topic identification technologies can be applied to recordings in near real time, companies can identify customer satisfaction trends before they become issues. They can quickly create and implement programs that will satisfy and retain customers before they decide to take their business elsewhere.
Uncover critical corporate information
Analyzing recordings has typically been the domain of the call center. Yet there's a tremendous value in call recordings for managers in legal, marketing, and other departments. Since call mining creates a rough transcript of calls and stores the information in a standard data warehouse, the data is available for other departments to share.
A legal department can mine calls to mitigate business risk associated with liability claims or fraud. Companies can use speech analytics to classify or score calls to identify key topics and phrases that may cause potential risk to the enterprise. Agents can be instructed to modify their language for future customer contacts before the behavior results in a lawsuit. Businesses can use call mining technologies to profile calls over time and search for specific topics in customer/agent calls that could result in fraud. A simple example of this might be that agents are not properly identifying key personal and account information to confirm that the caller is the owner of an account.
Marketing departments can employ speech analytics to mine what customers like or dislike about particular products. Customer interactions can be profiled for specific topics to determine how customers react to existing or new products. Because call-mined data occupies a very small storage footprint, data can be kept for several years, giving marketers a valuable historical record of customer responses.
Call mining provides enterprises with a rich tool set to enable the analysis of all recorded calls. The technologies are now being employed to understand what customers are saying and how agents are performing. They make use of what may be an enterprise's most valuable asset: recorded calls. The refrain we may hear in the future will be "this call is being minedfor quality assurance." Then every customer's voice may be heard.
About the Author
Cliff LaCoursiere is senior vice president of sales and marketing at CallMiner. He leads the company's strategic marketing and sales initiatives that have cemented CallMiner's leadership position with best-of-breed quality management and workforce optimization vendors. CallMiner provides software applications and tools that uniquely enable companies to mine hours of recorded call center agent and customer interactions. He can be reached at www.callminer.com
Sponsored By: Informatica