|
Despite all the buzz that surrounds it, speech analytics are seldom fully understood and often underestimated.
By applying advanced speech recognition technology to spoken conversations, speech analytics solutions reveal critical insights that would not otherwise be apparent. And by measuring not just contact center performance, but all of the back-office processes and other customer touch points that shape customer perceptions, speech analytics extend the benefits of contact center recording from workforce quality to enterprise-wide customer process optimization.
Redefining Quality from an Enterprise-Oriented Perspective
Today, sophisticated speech analytics solutions, such as Verint’s ULTRA IntelliFind, are addressing quality within the context of each company’s business objectives, rather than as an independent contact center entity with no relationship to overall business goals. These speech analytics solutions are redefining quality in a way that promotes greater customer satisfaction, loyalty, and value.
An effective, goal-oriented speech analytics solution is built on three distinct and robust technologies — word spotting, call categorization, and root cause analysis — each providing an increasingly important layer of value to your enterprise quality initiatives.
Word Spotting to Quickly Locate Calls of Interest
Word spotting identifies words or phrases (such as Dissatisfied, Angry, Unhappy, or Cancel) spoken during customer interactions. With word spotting, you can quickly locate interactions in which competitors’ names are mentioned, new products are discussed, late shipments are cited, and so on. Although this information has value, word spotting alone does not provide an objective means for detecting trends and finding ways to improve performance outcomes. That is because word spotting requires you to define in advance the words that the technology should “spot” in conversations. Simply stated, word spotting requires you to know what to look for before you find it!
Plus, word spotting can yield misleading or invalid results. For example, use of the word “dissatisfied” in a customer interaction does not necessarily indicate a dissatisfied customer or that dissatisfaction is the subject of the call: I don’t want you to think I’m dissatisfied with the price. It’s probably right. I just found this invoice difficult to understand.
Call Categorization: Calibrating Quality to Business Goals
Call categorization is the definition of business-related categories (such as Billing Issues, Product Feedback, Repeat Calls, etc.), the automatic placement of calls into those categories based on call content, and the ability to drill down into each category to listen to the most relevant calls. Call categorization positions you to monitor your company’s performance with respect to real-world business goals, so that the quality that you measure is the quality that matters most.
For example, you undoubtedly are interested in all calls in which customers say they are happy or unhappy with your company. But, you will probably find calls in which customers cancel their accounts because of frequent BILLING ERRORS highly relevant to your customer retention targets and immediately actionable. (Based on this information, your can institute changes to billing procedures to improve customer satisfaction and stem the tide of defections.) Your Sales and Order Departments are likely to be very interested in the BACKORDER category of calls when they hear many customers say they are buying competitive products because they can no longer wait for your back orders to be filled. And your Shipping Department managers may find the MISSING PARTS category of calls particularly informative when they hear the number of customers calling about the same parts missing from specific product shipments.
Call categorization helps you identify and track trends, pinpoint ways to reduce call volumes, improve customer-related processes, and maximize revenue generating opportunities. But, while an essential element of any speech analytics solution, call categorization alone will not provide you with objective information about why specific business outcomes occur.
Root Cause Analysis and the Power of Why
Root cause analysis automatically reveals the underlying reasons for specific behaviors and business outcomes. Unlike other speech analytic technologies, root cause analysis reveals not just what is happening, but why it is happening, to provide greater understanding of the factors that shape business results.
For example, your call center might be experiencing a significant spike in call volume. Root cause analysis may reveal that a new promotion has been the cause of a large percentage of calls, as customers try to clarify the precise details of the new offer. This is actionable information that you can immediately use to take effective action — such as clarifying promotional details in new advertisements and direct mail — thus reducing call volumes, freeing agents to handle more profitable calls, and increasing customer satisfaction.
Root cause analysis complements call categorization, surfacing real-time cause/effect relationships. For example, a REPEAT CALLS category can enable you to listen to customer calls related to unresolved issues discussed in prior phone calls. But, the reasons why customers call back can change over time: one month, many customers may be confused about a new service price plan; the next month, a significant percentage of repeat calls may focus on a new sales promotion; and the month after, delayed product shipments might be the cause of repeat customer inquiries. Root cause analysis doesn’t just gather similar calls together; it can reveal and quantify the reasons for repeat calls as new circumstances emerge, without requiring you to listen to every call.
Conclusion
A comprehensive speech analytics solution, such as ULTRA IntelliFind, uses word spotting, call categorization, and root cause analysis to reveal strategic insights about the factors that shape customer satisfaction, drive operational costs, and generate business revenue. This actionable intelligence transforms quality into an enterprise initiative and increases the success of your quality programs by addressing customer touch points in your contact center and company wide.
----------------------------------------------
Daniel Ziv is Director of Business Analytics at Verint Systems.
|
Internet Telephony Magazine
Click here to read latest issue
CUSTOMER 
Cloud Computing Magazine
Click here to read latest issue
IoT EVOLUTION MAGAZINE
