LAS VEGAS — United Airlines is obsessed with trying to improve its customer service experience, especially for “the hard stuff,” according to Bryan Stoller, VP and global head of customer care and contact centers at United Airlines.
The easy part is the standard inquiries that can be quickly and efficiently handled by AI, Stoller said during a keynote presentation at Customer Contact Week in Las Vegas. The real challenge, and where United wants to truly stand out, is resolving the complicated problems that are left behind for human experts.
“This work is harder,” Stoller said. “This work is ambiguous. This work is emotional. This work is cross-functional. This is what I like to call the hard stuff — situations that don't fit neatly and cleanly in a policy workflow in a chatbot.”
United has a team of “hard stuff hunters” who spend their time building knowledge that will improve the company as a whole, Stoller said. These employees are on the lookout for high-effort, high-stress experiences that don’t lend themselves to easy answers.
The goal is to transform customer service into a support operation that can find reasonable answers to difficult questions and equip customer service teams with the tools and autonomy to solve problems that are outside standard operating procedures.
Stoller noted that United is only partway through its journey. The “hard stuff hunters” have already found success in some areas, but there is still plenty of experimentation and failure ahead.
Finding reasonable solutions to hard stuff
The real challenge in customer service is defining the standard operating procedure for things that don’t have standard operating procedures, according to Stoller.
One example is the last flight of a night being grounded due to inclement weather, according to Stoller. That’s a straightforward problem with a straightforward solution: Put the customers on a flight the following morning.
The challenge comes in the details, according to Stoller. What if someone has important medicine in checked luggage? What if an affected family has a child who is melting down? What if there are no open rooms in the local hotels?
These situations are impossible for an AI and can be very hard for even an experienced customer support agent, Stoller said. It’s no longer enough for customer service teams to try and offer rote solutions for complicated problems.
“Most customers, they're not trying to get around the rules,” Stoller said. “They just want someone to look at the situation and make a reasonable call. The future of service is less about us just being the answer department and more about us being the reasonable department.”
Autonomy leads to better outcomes
United updating its operations across four pillars, Stoller said: Detect, route, resolve and learn.
“These pillars probably sound familiar,” Stoller said. “They can be applied to today's business, but the difference is what's in them and how we achieve them.”
Detect requires teams to identify when an interaction is beyond the routine and will require a higher level of support, according to Stoller. This is especially important in automated channels to ensure smooth escalation.
From there, teams need to figure out how to route the call to whoever has the best toolset for solving the problem, according to Stoller. The resolve pillar is easy enough — provided human agents have what they need to make decisions that may step outside the usual solutions.
“You have to give these new teams context, tools, authority and judgment frameworks to resolve this, and you cannot constrain them with black and white policy,” Stoller said.
The final pillar, learn, creates a feedback loop that enhances the business, according to Stoller. Every customer call is a signal, and it can potentially identify where something isn’t working somewhere along the line.
The “reasonable department” operating model doesn’t seek to create another escalation queue, according to Stoller. It is aimed at becoming a new way to serve customers while elevating the experience, supporting people on both sides of the business in an increasingly AI world.
“I see amazingly great success for AI, and we will enjoy it, but let's not forget about the human experts who will be left with the hardest, most complex and most important customer moments,” Stoller said.
