New roles, new rules: How AI is reshaping jobs in customer service

Wes Dudley’s first move at Broad River Retail was to roll out a bolt of paper. 

As a new VP of customer experience in 2022, he set out to map the complete customer journey for a business covering 33 Ashley furniture stores across the southern U.S. As Dudley and other CX leaders plotted it out on paper, a not-so-pretty picture emerged: Customers frequently ran into roadblocks after making a purchase—such as clunky delivery scheduling, or waiting up to three days to resolve their queries.

“The process just didn’t work,” says Dudley.

Dudley realized he didn’t have the right combination of technology and people working together to smoothen the experience. So he tried a new approach: First, he deployed a team of AI-powered chatbots to automate delivery schedules. Next, he opened up a new role on the CX team: a dedicated “bot manager” to manage quality control of the chatbots and ensure they interacted with customers in the right way using the right content and workflows. It was a natural step to hand this position to a manager who had been responsible for populating the knowledge base. 

“The bot manager oversees the content and the tools to get what we need to improve the experience,” says Dudley. With the new system in place, Broad River has already slashed customer resolution times from 36 hours to six hours. And the AI chatbots are helping improve customer satisfaction by giving shoppers an easy, self-serve way to schedule deliveries—previously an inefficient process reliant on phone calls. 

Dudley’s creation of a “bot manager” role might have seemed out of left field just a couple years ago, but today, more companies are seeking ways to creatively deploy AI in customer support—not to automate their teams out of work, but to pair them with AI in strategic ways that drive results.

“AI-powered tools are helping service leaders scale their operations in new ways,” says Malcolm Koh, senior manager of customer experience consulting at Freshworks. “Customers see more personalized engagement, leaders see greater team productivity, and workers are getting back time and meaning in their days.”

Time flies with customer copilots

Even as AI takes on new tasks, highly skilled agents will always be a critical part of a company’s success—there will always be customers with more complex issues who require more personalized attention. But even here, AI can play a role in helping maintain high levels of job satisfaction, productivity, and performance by pairing agents with AI co-pilots.

Three months into his new job as a customer support agent at iPostal1—a leading global provider of digital mailboxes—Conner Fitzsimmons worked in the direct line of fire: talking nonstop to irritated customers with a range of issues, from late package deliveries to missing or incomplete postal forms that cause service interruptions.

Read also: 3 ways AI is redefining jobs in IT

Even with software tools to help manage phone and chat support, the range of questions, requests, and gripes Fitzsimmons faced was tough to manage, and often risked leaving customer issues unresolved.

“A lot of the time I'm in a rush,” says Fitzsimmons. “I’ll have four or five calls in the queue at all times, and I'm constantly rushing, trying to get through everything as fast as possible.”

Help finally arrived in the form of a new AI-powered copilot tool that helps Fitzsimmons quickly draft customer responses and answers, helps adjust and rephrase language to deliver the desired tone, and fact-checks and corrects everything using the company’s knowledge base. Called Freddy AI, developed by Freshworks (publisher of The Works), it’s an AI agent—a tool that autonomously performs complex tasks and routines in collaboration with other software and systems.

Fitzsimmons is now enjoying significant relief from constant stress and time pressure, while customers are receiving faster, more effective answers to their queries. In business terms, it’s a big win-win: using AI to simultaneously improve both the employee and customer experience.

Making CX measurement more intelligent

As agents tackle more complex problems and customer engagements grow longer, support organizations will no longer be able to use metrics like average handle times to measure the efficiency of their operations. Managers must rethink workflows and reassess how they measure performance, notes Peter Murphy Lewis, a growth advisor and principal of the Strategic Pete consultancy.

Also, as AI changes the relationship between support agents and customers, how managers measure team effectiveness will shift as well, says Shep Hyken, a best-selling author and speaker on customer experience. 

“As AI handles the easiest issues, the harder ones are escalated to front-line customer support," says Hyken. “That means service reps are spending more time on each issue, which is a great opportunity for companies to create a better experience for their customers.” 

These higher-touch engagements can, in turn, drive higher customer satisfaction, says Paul Daugherty, chief technology and innovation officer for Accenture, and co-author of Human + Machine: Reimagining Work in the Age of AI. After Accenture helped a major telecom company deploy generative AI to support its clients, the productivity of the telco’s customer support teams increased by 30%, while customer satisfaction jumped 60%.

AI brought speed and comprehension: It was able to understand the context of each interaction, including the customer’s entire history with the company and the types of equipment they were using. That enabled the telco’s support personnel to find the right solutions faster.

“Customers got the information they needed, they got it quicker, and were happier,” says Daugherty.

Success starts at the top

When it comes to how his CX team will evolve, Broad River’s Dudley is taking the long view: “Many of the jobs of the next five years aren’t even created yet,” he says. Defining a new role as a bot manager is only the first step. As the AI grows more sophisticated and works more autonomously, Dudley expects the bot manager’s role will evolve to “bot supervisor”—less hands-on and more managerial and strategic, like managing a team of workers, only the workers will be AIs.

If AI is going to drive that kind of organizational change, the process needs to start at the top, adds Daugherty. Executives need to be learners as well as leaders, to understand the potential and limitations of the technology, then develop a strategy that uses AI to drive innovation, not simply to cut costs. 

“The most significant shift has to come at the C-suite level," says Murphy. “They need to create a culture where AI is viewed as an augmentation to workers, and not a replacement.”