Speech Technology Magazine

 

Are We Underestimating the Role of Emotion?

Consider it along with data when evaluating caller experience.
By Kevin Brown - Posted Jul 28, 2014
Page1 of 1
Bookmark and Share

Over the past decade, the trend of outsourcing speech-enabled contact center solutions has accelerated. Changing along with this trend are application design and development practices. Many organizations rely on their hosted platform providers to decide who will design and develop their applications.

Whether you retain your own user experience experts or rely on your outsourcing provider, you should be aware of the various approaches used and the consequences of relying too heavily on a single approach.

A topic frequently discussed in customer experience (CX) circles is whether emotion is given enough significance in assessing current state and designing future state experiences.

Often, reliance on either emotion or data is given a preponderance of importance in current analysis and future design. Drivers for out-of-balance approaches include CX and UX professionals' personalities, clients' personalities, clients' industries, and channels addressed, among other factors.

A balanced approach will produce the best outcome. A balanced approach will address emotional aspects that may not be discernible from data, coupled with analysis of large amounts of data to highlight measurable opportunities for improvement that can't be identified by a limited amount of human-generated calls.

To address the emotional facet of the voice channel, voice user interface experts must monitor customer experiences. This includes understanding who is calling and why. Of course, having good data about call types and caller information helps speed this portion of the assessment. A sampling of calls recorded or monitored live in side-by-side monitoring sessions with customer service reps can usually provide a reasonably accurate perspective into who called and why.

Understanding callers' mental states places us in a better position to understand why certain activities could anger some customers while not having any impact on others. A frequent comment we hear from callers who feel that they have been ignored by a company is that not knowing how long to expect that they will be on hold especially angers them. For those who do not feel as though they were being ignored, long call queues without an estimated wait time are reported as a negative experience, but nowhere near the degree of those who feel ignored prior to their call.

Conversely, caller experiences can also affect callers' states of mind. Poorly tuned and infrequently monitored IVR routing applications can offset the excellent customer experience provided by a friendly, helpful agent to the point of leaving the customer with a poor opinion of the company.

Attempting to address caller experience on emotional state alone would be short sighted. It is impossible to accurately judge the emotional state of every caller. The best that can be done is to group callers into a few potential emotional states and eliminate the negatives and enhance the positives wherever possible. Using data to analyze callers' behavior provides another large piece of the puzzle.

Until very recently, capturing data on end-to-end calls was nearly impossible. If available, it was very expensive, and the solutions were problematic because they used IVR logs and frequently broke. So we resorted to telecom, ACD, and IVR reports to stitch together an accurate picture.

With the advent of advanced speech analytics, gathering valuable data for end-to-end caller experience is becoming easier, faster, and more cost effective. Data, coupled with audio recordings, allows voice channel experts to accurately track caller activities, while also providing a glimpse into callers' emotional states or, at minimum, their verbal reactions to prompting and menus.

Using a speech analytics solution allows voice experts to choose a prompt and then analyze caller activity across hundreds of utterances in just a few hours. Furthermore, the utterances can be easily listened to, thereby weaving the emotional connection into the analytical approach. It is at this point where discoveries are sometimes made that would be highly unlikely to be found through random sampling of hundreds of calls across many days. Combining data to drive the focus to transactions where callers do not interact with the IVR as expected, coupled with the ability to quickly and easily listen to utterances, provides emotional and data-driven insights that improve the caller experience while driving high ROIs.


Kevin Brown is managing director at VoxPeritus, where he specializes in speech solutions and caller experience consulting. He has more than 20 years of experience designing and delivering speech-enabled solutions for on-premises and hosted environments. He can be reached at kevin.brown@voxperitus.com.


Page1 of 1