Speech Analytics: The Time for ‘Listening’ to the Voice of the Customer Is Upon Us
May 15, 2012 Leave a comment
- Hundreds of millions of minutes of customer care conversations are recorded annually in contact centers, but less than 0.1% of recorded conversations are ever replayed and analyzed.
- Automated speech analytics, which can mine most of these conversations for useful information, is finally getting the attention of customer care executives as the technology improves and the ROI is validated.
I have been following the speech analytics market as it pertains to customer care and contact centers for the better part of a decade. The application has only come into its own as it gained credibility through successes in the past two or three years. Recently, there has been an uptick in sales at companies that provide the technology to monitor the ‘voice of the customer,’ such as CallMiner, Nexidia, NICE, Utopy and Verint. There has also been a flurry of merger and acquisition activity among companies that provide contact center solutions and those that deliver customer feedback applications. Verint acquired Vovici, a provider of feedback management solutions; Avaya purchased Aurix, a speech analytics company based in the UK; and Hewlett-Packard bought analytics platform provider Vertica. I believe the industry is now leaving the embryonic stage and moving into a more mature phase of growth that will continue for the next decade.
Simply stated, the basic speech analytics engine can scan previously recorded customer service voice calls looking for specific words, phrases or more subtle customer issues and trends, analyzing the information to deliver insights. Early systems were slow to catch on, because they were expensive and the sheer amount of data they generated and analyzed from customer service conversations made them processor hogs and very impractical in many environments. In addition, few enterprises had the resources to dedicate to analyzing the collected data and putting together action plans to capitalize on the wealth of information mined. With today’s modern systems, companies can select from a range of offerings: from those that simply spot individual words, to those that recognize developing trends and strategic information, such as persistent mentions of a competitor’s name, or the words “cancel my service” that might indicate growing customer service problems or opportunities. There are also more complex approaches which involve the use of phonetic (speech to phoneme) analysis and large-vocabulary continuous speech recognition (LVCSR) capabilities. There seems to be a solution for companies of every size and budget.
Using automated analysis of speech to extract useful information about the content of conversations allows mining of 100% of calls – to assess agent performance and customer attitudes and emotions, as well as business trends. Companies are using speech analytics to: pinpoint cost drivers in their business, business trends and opportunities; identify strengths and weaknesses of organizational processes and offerings; and help understand how the marketplace perceives their products and services. The wise use of this wealth of new and valuable customer information is leading to great enhancements in the levels of customer satisfaction and longevity as well as revenues per sale. It can also result in a reduction of the number of minor issues turning into major customer care disasters through early detection and correction.
The investment in ‘voice of the customer’ programs can quickly be very worthwhile, as recent reports suggest ROI is reached in as little as six to twelve months. So, I advise companies looking to implement a program using speech analytics not to scrimp, but to invest the money, time and effort to understand customer profiles, plan and select solutions, and ensure the appropriate levels of staff and other resources to turn the intelligence that is revealed into workable action plans. Otherwise, the new information will go to waste, and the raised customer and management expectations will lead to bigger disappointments.