AI And Decades Of Data: How Expedia Group Helps Business Partners Navigate Travel Industry Turbulence

Economic pressures and AI are rapidly reshaping the travel industry. But Expedia Group is taking advantage of its decades of data to help partners navigate the turbulence.

Many consumers know Expedia Group because of consumer brands like Expedia, Hotels.com, and Vrbo. Or from eye-catching ad campaigns, including a 2022 Super Bowl commercial in which actor Ewan McGregor urged everyone to prioritize experiences, not things.

But Expedia Group also operates a robust B2B network that totals over 70,000 businesses and 200,000 travel advisors, spanning corporate travel departments, airlines, hotels, and banks. Increasingly, this ecosystem is a key growth driver. In the last quarter, the unit's revenue grew 38% to $3.55 billion, well ahead of Wall Street expectations.

“B2B has been the fastest growing,” CEO Ariane Gorin told the Wall Street Journal. “So it will be a bigger part of the company.”

These partners are under increasing pressure to deliver outstanding experiences for their own customers, according to Karen Bolda, Chief Product & Technology Officer, B2B. And it falls to Karen and her team to make sure they have the technology capabilities that they need to stay ahead.

“We’re trying to take on some of the hardest technical challenges they have … so our partners can focus on innovating in their business and growing stronger,” Karen told CIO News in a recent interview.

AI is rapidly accelerating the ability of Karen’s team to prototype to deliver on this mission. But there’s a wide gap between experimentation and production-grade quality: “Demos are so easy and look magical, but you're not going to put enterprise-wide scale traffic on that software,” she said.

To bridge that divide, Expedia Group’s B2B team focuses its AI strategy around three core pillars: Make current products better, test and experiment with new technology, and run more efficiently. Internally, teams utilize all the leading AI models, and there’s a focus on quickly turning cutting-edge technology into real-world results. As a result, Expedia Group is continually shipping new AI-powered features and capabilities — and getting value from them. Using AI, for example, Expedia Group has also helped to detect over $4 billion in fraud across its network.

A software engineer by training, Karen spent the last 20 years in travel technology. We talked to her about the value of proprietary data in the age of AI, the importance of quality integrations, and how her team brings innovative AI technology to customers.

Data: The cause of, and solution to, bad AI

Expedia Group’s secret weapon is its own data. With over 70 petabytes of intent data under management, Expedia Group maintains “one of the world's largest travel data sets,” said Karen.

That access to rich first-party context helps Expedia Group B2B give partners the intelligence they need to be successful. For example, an AI-powered engine can automatically surface recommendations for travelers unique to their itineraries. This might be suggesting a car rental at airports where arriving passengers typically require them, for example.

“AI is as strong as all of the data it’s trained on. It’s not the AI models, it’s the underlying foundation that makes AI reliable at an enterprise scale,” she said. “Data is the foundation, AI is the accelerator."

A clean, trusted, and unified architecture is important. Otherwise, AI will create a very confident wrong answer, Karen said, or people will make bad decisions regardless of what AI is telling them. Expedia Group worked over the last few years to consolidate its various data platforms. The aim was to simplify the landscape and gain more control over the estate: “Having data is one thing. But having data that you can use to get intelligence, that's really the critical ingredient,” Karen said.

AI also requires companies like Expedia Group to rethink how they manage data in motion. The corporate world is awash with failed AI projects. A key reason is the lack of quality connections between disparate data sources: “Real-world integrations have got to be secure, they've got to be scalable, they have to be production-grade, they have to be guardrails,” Karen said.

Problem first, technology second

Part of the challenge facing teams like Karen’s is the need to keep up with the pace of change. For example, the company has experimented internally with Model Context Protocol, or MCP. Expedia Group also makes all the leading frontier LLMs available to AI builders through a single development platform.

“Whether it's an API, whether it's MCP, whether it's a UI widget; it's really about what the person is trying to do and what the best, lowest-error method to make sure they are able to do what they need to do,” said Karen.

Internally, an innovation center helps Expedia Group B2B quickly transition emerging technologies into user-ready tools. For example, Expedia Group’s AI-powered trip planner for partners, dubbed Smart Trip AI™, was born out of the hub.

“We have strategy, roadmaps, schedules we’re delivering, but we also have a space that lets us continuously test and experiment with new technologies,” said Karen. “It’s really the engine that accelerates how we are prototyping, testing, scaling products.”

It’s that constant cycle of iterating and doubling-down on the successful projects that keeps Karen’s team one-step ahead of the industry — and partners one-step ahead of their own customers’ needs.