Qantas Loyalty soars past automation goals with Zendesk AI Agents

Qantas Loyalty runs one of the largest membership programmes in the Asia-Pacific, with around 17 million Frequent Flyers earning and redeeming points across a growing ecosystem of platforms including Qantas Marketplace and Qantas Wine.

“We have about 17 million Qantas Frequent Flyers, and quite a lot of them shop on Qantas Marketplace and Qantas Wine,” says Luca Giaccardi, Senior Operations Excellence, E-Commerce at Qantas Loyalty. “They use our platforms to redeem and to earn points.”

That scale comes with volume. Each month, Qantas Loyalty’s team fields around 3,000 inquiries through Zendesk. A mix of bot and form-based contacts on top of roughly 7,000 calls a month through its phone lines.

“We get quite a number of interactions through different channels,” Giaccardi says. “And all the AI Agents that we use are integrated with Zendesk. It has made quite a lot of difference to be able to triage all of the inquiries coming through.”

Two clear goals: Reduce noise, speed up resolution

When Qantas Loyalty began exploring AI Agents, the team set out with two specific objectives.

“The first two objectives we had at the beginning were to triage the most common knowledge inquiries that would come through—where we wouldn’t need to essentially be part of the inquiry at all,” Giaccardi says. “We wanted to be invisible behind it. A lot of general questions would be answered through the AI Agents, and that was one of the main key drivers that would release a lot of noise.”

The second objective was about smarter handoffs.

“The other objective was to grab all the information from our members and pass it on to our agents in a quicker, smarter way to be able to resolve the inquiry with one touch point,” he says. “That was one of the things we struggled with: going back and forth with the agents.”

Ticketing, AI Agents and knowledge working as one system

Qantas Loyalty leans on Zendesk in three main ways.

“The ticketing system—it’s not very sexy, but it’s very, very efficient for us,” Giaccardi says.

The second is the AI Agent, deployed as a chatbot across all three platforms the team operates. “That has been the one that has changed things for us in terms of contact reduction,” he says.

The third is the knowledge base—a library of FAQs and articles that serves both members on the public website and internal and third-party partners through restricted visibility, so they too can find answers and be helped by the bot automatically.

Doubling the automation target—twice

Qantas Loyalty’s automation results outpaced expectations almost immediately.

“We had an initial goal of about 30 to 35 percent automated resolutions when we first launched,” Giaccardi says. “We went up to 60 percent within the first two weeks—and then we had peaks of 85 percent. Which we did not expect.”

For general knowledge inquiries specifically, resolution rates climbed even higher. “With regards to general knowledge inquiries, it goes up into the high 90s on some months,” he says. “It just works, so long as we put all of the information there and the intent is clear—and the AI Agent can actually resolve, not only close the tickets, but resolve the inquiries itself.”

The team has also found that performance compounds over time.

“One thing that we’ve noticed is that the more we refine our agents, our AI Agents, the more we tweak, the more we review our cases, our inquiries, the better the AI Agent itself becomes,” Giaccardi says. “It’s a sort of a self-fulfilling prophecy. So the more we do work at it, the better it becomes—and the higher the resolution that we get, which is great.”

Freeing up the team for the inquiries that need a human touch

For Qantas Loyalty, the value of AI Agents isn’t just about ticket counts—it’s about routing the right work to the right place.

“The AI Agent has helped triage all of the inquiries that come through—and basically trickle back to the humans the cases where a little bit of TLC was needed, and a little bit more action with integrations that are more complex,” Giaccardi says. “All of the tools we used through Zendesk just empowered us to be able to do everything at once—quicker for the member, quicker for us as well.”

With automation now handling the bulk of routine, high-volume questions across Qantas Marketplace and Qantas Wine, Qantas Loyalty’s team is positioned to keep refining its AI Agents, sharpen its knowledge base and continue building a digital experience that matches the scale of its 17-million-member programme.