One of the known challenges encountered daily by those in the call center industry is that call center agents do not know the voice at the other end of the line. The minute they are on the phone, they do not have an ounce of an idea about the person’s character, mood, and disposition. Each and every time a customer calls, and much more during an outbound sales call, everything about the person at the other end of the line is a mystery yet to unfold for the call center agent.
Emotion & speech analytics in action
That is probably one of the reasons why Beyond Verbal, an emotion analytics company, came up with (and is further developing) a software that can algorithmically detect emotions through the person’s voice intonations, as New York Times reports.
“It is not what you say. It’s how you say it.” They use this technology to detect 400 variations of different moods and emotions beyond the meanings of words. It is real-time – emotions are detected as the calls happen, and an agent can act or decide based on the analyses.
Two more companies mentioned in the article likewise intend to qualify and measure a person’s mood, but instead of focusing on voice intonation, they employ speech analytics, which classifies the words and phrases that customers use during the calls. They collect these data to parse and try to measure the level of satisfaction or dissatisfaction of a customer.
Can mood detection technology be a valid basis of action?
Are speech analytics and voice intonation enough to detect a person’s mood at a certain time? How do these technologies address the other factors you can get from semantics and semiotics? What about culture? What about the individual’s personality? What about setting?
Filipinos, for example, are generally patient people. It doesn’t automatically mean that a person is calm when they speak softly on the phone. There are people who take issues and matters very patiently while deep inside, they are extremely angry at a particular product or service. There are people who suffer from inferiority complex, that they cannot push authority even when they are on the right. There are people who cannot communicate very articulately and comprehensibly that it can be impossible to decipher what they mean to say over the phone.
Filling-in the gap
These emotion detection technologies are coming out in the market by the end of this year, but are these types of software worth the bucks? How valid and accurate can the data be?
Phone support is just one of the channels to receive feedback from customers. It is also one of the hundreds of marketing channels a salesman can use to sell a product or service. This compels customer service companies to not focus on one channel alone in order to get the overall feedback data from consumers, but from all other bourgeoning technologies that are widely used and accepted by the general public.
Reaching out to customers is more than just about detecting their emotions over the phone, but also about identifying a collective behavior of a select group of market segment over a period of time. We can integrate all channels (social media, surveys, phone calls, etc.), regularly ask for feedback, involve front-liners of salespersons in understanding the clientele, and many more strategies.
But one thing is for sure: emotion is just one of the many things we need to know about our customers. Customer service companies have to think of more ways to know and understand this behavior before making actions and decisions.
If you are in the call center industry, do you think that these types of emotion-detecting software will be able to improve your initial interaction with a customer? Do you think that mood detection will be an accurate reading of what a caller’s personality would be? Do you think that the Philippine call center industry should get a hold of such technology?
Perhaps these are some things these mood analytics companies need to resolve before getting their products out in the open. Or maybe a feedback loop.