LogMeIn’s Sr. Manager for Bold360 Products Discusses AI for CX: Lessons Learned and Best Practices for Success - Argyle Executive Forum Events
Akhil Talwar
ABOUT AKHIL TALWAR Akhil Talwar is the Senior Product Manager for Bold360 at LogMeIn, Inc. Working at the intersection of technology and business best describes the past 13 years he’s spent in various industries including financial services, energy, legal, and high-tech. His career track includes managing mission-critical projects at General Electric and NSDL, serving some of the largest global utilities and stock exchanges. He also spent several years at smaller startups, including Recommind, bringing to market machine learning-based products for legal/enterprise search and document management. His recent work at LogMeIn includes helping several global customers successfully manage ecommerce and online support within their contact centers. In his present role, he’s focused on building the next generation of customer-engagement platforms. He’s passionate about the application of latest advances in artificial intelligence and user-experience design to drive better customer-service outcomes.

Akhil Talwar, Senior Manager for Bold360 Products at LogMeIn, talked about principles for creating a positive chatbot experience for your customers.

“There’s a lot of apprehension in the industry overall about AI, and rightfully so,” stated Talwar at the outset of his thought leadership presentation at the 2019 Customer Experience Leadership Forum: Beyond Satisfaction, Building Loyalty, held on April 9 in Atlanta. “Why should we even be having this discussion about AI and chatbots? I’m going to look at some of the bad experiences we’ve all had and analyze where we go from here by looking at the better things we should strive for by viewing AI from a self-service aspect for customer experience and engagement,” said Talwar.

Consumers are going more and more to self-service before they reach out to contact centers or anyone in customer service within our organizations. People don’t want to talk to customer-service reps, especially the younger generation. Another trend is acquiring products or services without actually talking to anyone inside the organization. By 2020, at least 80% of businesses will have deployed some kind of a chatbot. Conversational commerce is a broader way to acquire customers rather than just being useful from a post-purchase standpoint. A study found that agents inside contact centers are only spending 41% of their time solving problems for customers. What they’re spending the other 60% of their time doing is potentially ripe for automation,” he observed.

“With all of this, you’d think the industry is primed for more pervasive chatbots, but the reality is CFOs are pushing hard to get more out of their customer-service and support organizations, so there’s the cost component. Who here has had a bad chatbot experience?” asked Talwar. “Looks like about 70% of you in the room. Part of that is due to bad chatbots and part of it is customer expectation. Customers are expecting something more intelligent, and their expectations aren’t being met. Chatbots are creating a bad, self-serve customer experience,” he noted.

“To create a good chatbot experience, I’ve come up with a set of seven principles. The first is a seamless coexistence between humans in the contact center and AI. One of the ingredients to creating that harmony is recognizing that consumers expect these experiences to be conversational and the results to be relevant. Chatbots are scripted, so as soon as a customer deviates from the script, there can be problems, so we need to make this a more conversational, freeform experience. Also, chatbots don’t need to be a typed experience. You can have engaging experiences using visual cues,” said Talwar.

“It’s necessary to have an intelligent foundation, and that core foundation is based on natural-language processing. Your AI has to understand nuances of language as well as context and be able to carry that forward from one conversation to another. Also, chatbots aren’t just about providing an FAQ or a knowledge base. Bots need to have a transactional component so customers can actually do things without talking to an agent. Also, it’s necessary for the customer to be able to seamlessly transfer from a chatbot experience to a live-agent experience if the bot isn’t able to solve the problem—and it’s crucial for the agent to be empowered with the full context of everything that transpired with the bot,” he said.

“AI can empower your agents, as well, to provide consistency and improve productivity by helping them answer mundane, repetitive questions quickly. The other big thing is that this technology has to have learning built into it, so it can adapt over time. It should be able to capture not just what customers are asking but what the front line of your workforce is experiencing that needs to be encapsulated in your self-service journey. Last, but not least, you need to adapt to constant change by visualizing, analyzing, and taking action on insights you have.”