Avoid the Phone Slam with a Finely Tuned IVR
Once targeted calls are identified through the speech analytics process, a supervisor still needs to listen to them to ensure that the system is working properly and has detected the calls that require intervention in the form of correction, coaching, or changes in business processes. Speech analytics streamline the process. Managers no longer need to listen to all calls. The difference from the past is that supervisors typically have to listen to only the 10 or 20 percent of calls to find those that require their direct and often immediate action, instead of wasting a lot of hours reviewing as many calls as possible just to find the few that would benefit from intervention.
Listening to Customers
The customer side of conversations provides tremendous value to organizations by pinpointing the reasons behind customer complaints and identifying quality gaps in interactions, products, services, and internal processes.
Many times, miscommunication between the company and customer occurs. A firm’s marketing department checks to see whether callers are mentioning their recently launched loyalty program. Few callers refer to the program by its given names—say, Silver, Gold, or Platinum membership—so the promotion seems like a failure. However, after conducting a content audit, the business understands that the consumers were referencing the program but ignoring the official package names and instead asking about special rewards. Also, a business may learn, for example, that customers are more likely to say “terminate my service,” “stop my subscription,” or “get rid of my subscription” in place of expected phrases like “cancel my service” or “cancel my subscription.” Gone undetected, these minor differences can distort results significantly.
Breaking Out of a Vacuum
Also, IVR applications do not sit in a vacuum. They are part of the firm’s operations and as noted make up a significant portion of the customer experience. As a result, businesses need to work proactively and think in advance of the actions they will take whenever certain phrases and trends are detected. They need to build efficient business processes into their speech-enabled IVR systems—for instance, ways to upsell customers who reach out to the contact center.
They need to view the software as a filter, one that can help them get to what is most important to the customer more quickly. Automation and integration among the recorder, analytics system, and other applications makes business processes more efficient. This integration enables companies to identify problems and fix them. For instance, a few customers call the contact center because information was missing from the website. Armed with this information, the organization then updates the page, reducing the number of calls and resolving its customers’ frustration, a classic win/win.
As vendors deliver speech functionality that is embedded in, integrated with, or tied to other applications, such potential benefits grow. These solutions perform tasks such as powering best-action recommendations, offering real-time guidance, carrying out prescriptive analytics, and supporting predictive analytics.
Finding the Right Person
Another challenge: Speech tuning requires special expertise, so you need to find someone with the necessary skill sets. “As for experience, you definitely want a qualified speech scientist involved for tuning,” Genesys’ Underdahl notes. “Usability should be done by someone who has specific experience with that style of testing, as there is an art to putting the test together and interviewing the caller to gain unbiased answers.”
There are not a lot of individuals with such expertise. With enterprises spending more money on their IVR solutions, demand for these professionals has been growing. Whenever demand is high and supply is low, salaries and corporate expenses rise. Currently, that scenario is true: The job site Indeed.com listed 1,360 open positions for IVR system analysts, and close to three out of five (57 percent) of the positions offered salaries of $85,000 or higher.
System testing and tuning has been an ongoing issue for speech recognition IVR systems. These solutions have been getting better lately, so fewer consumers now slam down their phones in frustration. But more work remains before businesses (and customers) are completely satisfied with how these systems function.
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, Mass. You can send him a note at paulkorzen@aol.com or follow him on Twitter at #PaulKorzeniowski.