Smart purchasing and project managers in the B2B enterprise have used the internet to research product solutions and then contact their vendor or supplier of choice to complete the buying process since the mid to late 1990s. Over time, companies evolved and started to employ a “live-chat” feature to help prospective buyers find what they were searching for or to make suggestions. That normally involved having live agents on hand to chat or text with prospects and customers about their needs.
In the harried world of today’s business life, getting quick online answers to questions can be a true competitive advantage over those companies with endless phone-based menu options when prospects or customers call them – press 554 for sales, press 369 to hear a duck quack, and so on.
Keeping with the theme of this issue of SD&I, Artificial Intelligence is powering the development of software and hardware that could replace or at least greatly augment the human interaction of live chat, call centers and CSR departments for all inbound inquiries.
Todays’ breakneck speed of technical advances in machine learning and chatbots will revolutionize the way we interact with companies for information, services and products.
Machine Learning
According to Wired.com, “machine learning is a branch of Artificial Intelligence, which involves the design of computer applications that are able to learn based on their data input and/or output.”
Machines are being developed that can learn based on experiences and continued interactions. Think of Big Data and using machine learning to extract just the information you specifically are looking for – a machine that can remember what you searched for in the past, what preferences you have in receiving information and is constantly learning how to refine that vast data.
This is quite possibly one of the biggest leaps forward when it comes to customer service. Machine learning can help those customers who want to maintain their own level of independence in the buying process or help customer service agents go above and beyond in their traditional level of service.
Chatbots
The Oxford dictionary defines a chatbot as “a computer program designed to simulate conversation with human users, especially over the internet.” This takes us back to the live chat feature – except chatbots can work autonomously 24/7. While this would not entirely replace human interaction, you could now have human employees resolving complex customer issues while leaving the chatbots to answer inquiries that are asked on a regular basis.
Dramatic changes have occurred in our personal lives as well as in business. Just look at voice interface devices such as Amazon Echo and Google Home. While these are consumer applications, there are literally thousands of skills these devices are learning and employing. Consumer acceptance and consistent usage of these products will eventually help bridge the gap of business acceptance and usage.
And this brings me to the “grumpy old man” portion of our story – but instead of yelling at today’s kids to get off my lawn, I will just point out a cautionary tale of not getting ahead of ourselves in replacing human interaction with chatbots.
According to CBS News, researchers at Facebook Artificial Intelligence Labs built a chatbot earlier this year that was meant to learn how to negotiate by mimicking human trading and bartering. When the social network paired two of the bots, nicknamed Alice and Bob, to trade against each other, they started to learn their own bizarre form of communication. In human terms, they developed their own language and started to barter with each other in “machine-speak” since no one ever told them they had to communicate in English. Facebook was forced to shut down the project.
As we look at technology to improve our personal and business lives, it is good to keep it all in perspective and make sure that these new technologies are effectively elevating the customer service experience at every turn.
Ric McCullough is vice president of sales and customer service for PSA Security Network. To request more information about PSA, please visit www.securityinfowatch.com/10214742.
