Making the Business Case for Speech Analytics
Selling, Upselling, and Compliance
Marketing and sales are another area where strides are being made. Here, the business case is often easy to identify: greater sales conversion rates. For instance, a firm can examine its current processes and refine them. "Agents can use different wording and increase the likelihood that a sale is made," says John Busby, vice president of advertising platforms at Marchex. In addition, agents might struggle selling to customers who called for product information. In this case, coaching might help improve their performance.
Furthermore, new upselling opportunities can present themselves. An insurance company receives a call from a customer who wants to change her son's address on her car insurance policy because he is leaving for college. Analytics could create an alert for a possible sales opportunity. The address change indicates that the son will have a separate residence, which represents an opportunity to add a renter’s policy to the family’s coverage.
Compliance departments could benefit from speech recognition as well. "Recently, we have been seeing compliance drive interest in the products," Genesys's Kolman says. A few factors are driving that increase. The number of regulations that enterprises need to adhere to has been growing. Businesses need to put mechanisms in place so they conform. Speech analytics can help, for example, pinpoint how well employees adhere to scripts for greetings and closings in agent-customer conversations. The initial results have been promising. One client increased its compliance rates from 35 percent to 87 percent in one region and 59 percent to 99 percent in another region, according to Genesys's Kolman.
Viewing the Entire Interaction
A complete snapshot of the client's experience, otherwise known as the customer journey, is the customer service holy grail. Consumers now often interact with a series of autonomous company representatives and applications. The customer starts with a salesperson at a retail store, who works with an order entry application; calls a person from the billing department, who works with another system; and ends up in the contact center, where a third person uses a different solution. The consumer sees one company and expects an integrated experience, but the employees see only their slice of the interaction and have limited (and oftentimes no) visibility into the other systems.
To operate in an integrated fashion requires a number of technical and organizational changes. On the technology side, data has to be consolidated. Currently, information is stored in departmental systems. In a call center, information is often lost during a call transfer process. To properly serve the customer, the information coming from all communication channels—calls, chat, email, and social media—should follow the caller on her customer journey.
While these systems have been getting better, they are not yet perfect. Data cleansing has been an ongoing challenge. Organizations work with information from a variety of different database management systems, which categorize data in different ways. The contact center might have a nine-field customer record and the accounting department may have a 15-field record. As information is consolidated, developers need to make sure the data looks the same, a process dubbed data cleansing.
Organizational change is also required. "No department should own the speech analytics system," notes DMG Consulting's Fluss. "For the systems to be most effective, they need to be purchased and managed at the enterprise level." She recommends that businesses form an analytics team that has support from the highest levels of management. Only with that support is the analytics group able to translate analytical findings into actionable items. Sometimes, the system points to radically revamping current business processes. For a variety of reasons, including a natural resistance to change and self-preservation, departments often balk at radical changes.
Getting Information in Real Time
The ability to perform analysis in real time is the next step for these tools. "The manager wants to coach the agent while the event is taking place, not after it has occurred," notes CallMiner's Gallino. For instance, individuals who have placed multiple calls to a contact center have a high likelihood of churning. In real time, hostile tones and words such as confused and cancel can be flagged and sent to a call center supervisor for resolution. The supervisor might offer a refund or a special gift to a customer whose experience has been less than stellar. Ideally, such moves keep the caller a customer rather than have him turn to a competitor.
An emerging feature in this area is emotion detection, and here the analytics can help not only customer service but sales as well. Calls analyzed using this type of tool provide data on how customers feel about the company—and armed with this kind of information, an agent may have a better chance of upselling them a product or service.
Speech analytics have tremendous potential for helping organizations streamline business processes and interact with customers more effectively. But these solutions may also require corporations to dramatically overhaul their operations. "With speech analytics, corporations can reduce expenses, increase income, and streamline operations, if they are used properly," concludes DMG Consulting's Fluss. "The key is figuring out how to use the solutions properly."
Paul Korzeniowski is a freelance writer who specializes in technology issues. He has been covering speech recognition issues for more than a decade and is based in Sudbury, MA. You can send him a note at paulkorzen@aol.com or follow him on Twitter at @PaulKorzeniowsk.
[Editor's Note: In the original story, Teri Navin at Nexidia was incorrectly quoted as saying, "One of our customers listens to 15,000 hours of conversation each day." The original quote has been paraphrased and now reads this way: "One of Nexidia's customers analyzes up to 150,000 hours of audio each day, according to Teri Navin, product marketing manager at Nexidia." We regret the error.]