AI is the game-changer needed in customer service
The holiday season is a busy time for call center agents. However, problems in the CX industry are leading organizations to turn to machine learning for optimization.
The holiday season means an influx of orders for businesses. It can also mean long wait times for those contacting call centers and customer service agents.
However, this year with the Great Resignation and many leaving call center positions, contact center and customer service agents are scarce.
It's getting harder for organizations to find employees to fill service departments, according to Shawna Wolverton, executive vice president of product at customer experience vendor Zendesk. The vendor provides SaaS products to customer support and sales departments. Wolverton said even when organizations hire new employees, the hires require training, which can take time. During the holidays, organizations need to respond to their customers quickly.
In this Q&A, Wolverton discusses how AI, automation and machine learning (ML) can play a role in helping customer service agents as they deal with the influx of customers, not only during the holidays but also in the future.
What are some of the problems agents are facing in the CX world?
Shawna Wolverton: Long gone are the are the days when customers feel like they had to get on the phone with someone to get an answer. Right now, they want to get their answer fast. They're sort of used to Googling for answers. So, we're really optimizing for that and we're seeing our [users] really want to optimize for that.
It comes down to a bunch of things. One is around automation and this idea of bringing together powerful, conversational experiences that allow customers to get those answers fast. Not necessarily without having to talk to an agent, but … freeing up agents who maybe were sort of bogged down in: 'Where is my order, and when will it arrive? Can I reset my password? Can you help me change my reservation?' Being able to self-serve some of those things. Then freeing up those agents for those higher-value, more intensive conversations that you do sometimes need to have that one-on-one time to dig in, person to person.
Shawna WolvertonExecutive vice president of product, Zendesk
What are some specific ways can AI and ML help free up customer service agents?
Wolverton: One of the most valuable information sets a business has is all those tickets that they've solved before. We're finding great benefit from helping that new agent get the context, not just from the customer, but from all the questions and answers that have come before. So, understanding intent and surfacing that for the agent and then providing suggested responses.
We have the ability for automated responses called macros that we can suggest based on doing some machine learning and detection on the issues that come in and then surfacing the answers, either a help desk article or a previously closed ticket that was solved well, and providing those to customers.
On the other end is really like that sort of full automation and building out a chatbot that allows you to recognize those intents and then give answers automatically to customers sometimes without even having to go to an agent. Then the ability sort of through, you know, conversational APIs that exist to build out the kinds of systems that recognize intent and then actually offer an interactive solution so you can maybe do the password reset in the messaging conversation or do the reservation change without having to talk to an agent.
When you know there's a ton of volume, you can really use some of the machine learning and intent detection to understand how angry a customer is or how to route that issue. If someone's really upset about shipping, you can get them directly to someone who can help with that problem, rather than having to escalate that through multiple lines of agents and getting transferred. It's a much better experience for the end user and then much better experiences for the agents as well.
Is this reliance on automation, machine learning or AI something that will continue to grow in the customer service business?
Wolverton: What's great is that this technology is moving so fast. Even in our own portfolio, our bots used to be able to suggest an article and that was sort of the end of the game.
The more this technology continues to evolve with stronger intention and action with the ability for these bots to have more natural conversations with customers and to learn more and more from tickets that have been solved already, then I think it is going to be a critical part of the growing customer service and customer experience teams.
It's going to be a way that they can differentiate in the marketplace of being able to get great answers to customers even more quickly.
Where do you see this technology going as we go into 2022 and beyond?
Wolverton: I think we're at the beginning of the of the curve here. As this technology advances and develops and becomes democratized … more people are going to get the power of this and they're going to see the benefit for their customers and the benefit for their agents. I think we'll continue to see more evolution here and more and more customers will be adopting this kind of AI and ML technology, especially across channels, like messaging, where you have these long-running, ongoing conversations. With this idea of messaging … you can easily switch between automated conversations with bots for your quick answers, and hand off easily to agents and those conversations can really live both in an automated and a human-to-human world.
Editor's note: This interview has been edited for clarity and conciseness.