two women talking
PHOTO: Trung Thanh

At its core, conversational experience is an easy-back-and forth with a customer on a specific channel such as chat or voice. The customer is able to speak or type naturally and ask spontaneous questions without having to navigate cumbersome IVRs, thank you very much. In short, said Jim Freeze, CMO of virtual assistant provider Interactions, customers are able to talk or type to a virtual assistant exactly how they would if they were speaking with a human. Or put another way: “The idea is to move away from forms and buttons to a process that enables an exchange that resembles what we have in a 1:1 conversation with a friend,” said Ajeet Kushwaha, director of bots and automation at customer engagement software provider Freshworks.

Understanding Customer's Intent With AI 

This process is typically supported with the latest artificial intelligence (AI), including automated speech recognition, natural language processing and machine learning that continually improves the effectiveness of the virtual assistant, Freeze added. “The most effective AI has been extensively trained to understand a wide variety of words and phrases, and can understand a customer’s intent even in a complex sentence,” he said.

Related Article: Natural Language Processing Is Hitting Its Stride

Choose the Right Channels

Ideally, a company is able to implement a conversational experience for every channel through which a customer contacts it. But that isn’t always practical, said Lesley Yarbrough, director of customer support at helpdesk software provider Groove HQ. “Yes, it needs to be easy for your customer to contact you and start a conversation, but this doesn’t mean that the customer gets to choose where it happens. Just because a variety of communication channels exist, you don’t need to try and support them all,” she said.

Instead, she continued, based on the company’s resources and the type of business, it should be easy for customers to contact it on the channels where it can provide a great experience. “Choosing the channels you support and don’t support will likely be a difficult decision,” Yarbrough said. “Certain channels require more staff, others require more tech, some require both. Every business has its own unique requirements and limitations.”

Think about it: A software-as-a-service (SaaS) company probably doesn’t want to try and communicate via SMS, but an online retailer might have success there, Yarbrough said.

Related Article: Google Duplex's Conversational AI Shows a Path to Better Customer Service

Not a Free-For-All

It is also important not to be misled by the term conversational experience this is not a venue where customers can chat about anything. Conversational experiences need to be designed to respond to very specific aspects of the customer journey, said Dave Bruno, managing director at unified commerce platform provider Aptos. "We must resist the urge to deploy experiences simply because we can." Why? "Well, for starters, despite the massive adoption of smart speakers and digital assistants, the technology is still very immature, making complex interactions frustrating and ineffectual," he said.

One aspect companies must deliver on, however, is timeliness. “In today’s digital age, customers expect brands to respond to their message in a timely manner,” said Katie Wilson, founder and CEO of direct advertising platform TapOnIt. “There is little excuse for brands to keep a conversation with a customer hanging [given that there are] so many channels to contact them, from social to email to phone to text.”