Natural Language Understanding Forces Rethink of Contact Center Analytics


According to a new whitepaper “natural language understanding” which is a new approach use by the contact centers are forcing a rethink of traditional center analytics. Generally, contact centers measure the number and duration of calls as its key metrics to get more customer calls completed quickly as soon as possible. But it is not necessarily the best approach they say.

The researchers are exploring how natural language understanding is an emerging technology that helps to convert customer conversation using voice bots to the text which gains new insights about customer sentiment. Sharon Melamed founder of the matchboard says that some voice bots conversation is providing new data sources. voice bots are helping people by providing a new way to engage with brands where they can reveal their issues, frustration, likes, and dislikes. The insight obtained from this can help companies identify problems, fix broken processes, and can also understand what is making customers angrier since voice bots can detect raised voice and emotions.

Especially for business purposes, this information can be combined with analytics to gain a better understanding of cost or revenue impact by addressing certain issues that they have never gained without the help of a voice bot. Dave Flanagan’s director of digital and conversational AI says that customer experience and operational efficiencies are the one thing organization must have in their mind when looking at the contact center and voice bot analytics. Also, he said that the bot should have a solid foundation, to begin with, to keep it successful when handling the customer request.

According to the whitepaper, the way to do this is to focus on critical use cases that add value to the customers. By providing the bot rich data around its core capability by ensuring that it does a few things well. Once there is a surety that the bot can handle most of the user expressions then the organization can look to develop new capabilities. Also, another key metric identified in the research is that the confusion triggers which can be used to check how often the bot don’t understand user request and how it responds in this evitable situation.

Flanagan also says that it is very important to check the user engagement and identify the bottlenecks that are occurring in a conversation that results in errors, abandonment, or escalation to a human agent.


Please enter your comment!
Please enter your name here