One company pioneering technological innovations in travel is KAYAK, and the company is taking a self-declared bullish approach to using artificial intelligence to forward the next phase of travel planning. For example, KAYAK recently was the first in the travel industry to launch of a hands-free booking capabilities on Alexa for hotels.
KAYAK assistants currently are also found on platforms like Facebook Messenger, Microsoft Teams, Apple TV, Google Home, and Slack. To understand how travel bookings and planning are changing, Digital Journal caught up with Matthias Keller, the Chief Scientist at KAYAK.
Digital Journal: What are the main challenges facing the travel sector in terms of AI?
Matthias Keller: One challenge we’re facing is that voice transactions are brand new and people are not yet used to spending money on these platforms. Because of this, our first iteration currently offers hotel content that can be refunded–no strings attached.
Another challenge is educating the user and properly presenting a hotel to them. Looking ahead to the future, we hope to add flight bookings to our skill but flights pose a major challenge due to the regulations involved. And, we want to make sure the user is getting all the information they need in order to make an informed decision, so they feel confident in their choice post-booking.
On platforms like Alexa, listening time is premium. The goal here is to get in early on AI and learn as much as we can, so that when these devices and our comfort with voice becomes mainstream, we’ll be ready.
DJ: How can AI help consumers with travel planning?
Keller: AI will help make travel planning more personalized (real-time recommendations), context aware (such as canceled flights? or a family or business trip?) and geo-relevant (like when you’re stranded at airport or already at hotel), as well as make customer support more efficient post-booking.
Imagine, you are standing in your kitchen making dinner and you realize you forgot to book a hotel for an upcoming business trip. Without ever touching your phone, you can ask Alexa not only how much is a hotel room at your usual hotel, but you can also book that room in a matter of seconds without ever lifting a finger. This is the future of travel planning.
DJ: And your service can help with this?
Keller: Our goal with AI and machine learning is to create user experiences that fit seamlessly into a travelers’ daily lives, on the platforms they’re interacting with most. Take our Messenger bot, for example- with the KAYAK Travel Planner for Facebook Messenger, you can plan your group trip from start to finish with just a few taps and without leaving Messenger. We are making it easier to get your group trip off the ground on the platforms you know and love most.
Airport wait times are another example of an intricate problem that might only be solved by machine learning coming to the phone.
Current KAYAK in-app security wait-time estimates are based on data from official sources like the TSA and feedback from helpful travelers. With AI, KAYAK may someday allow you to stay home an extra hour because you know the exact wait times at the airport security line.
DJ: What advantages does AI bring to the travel industry?
Keller: AI gives us the ability to meet our users on the platforms and devices where we know they are spending the most time in order to ultimately make the travel planning process more seamless. While AI is still in its early days, it is opening new doors to help take the friction out of travel planning from start to finish. When it comes down to it, we want to be useful to consumers wherever they are.
DJ: Please can you explain more about the KAYAK service and its artificial intelligence
Keller: Facets of AI have been in existence in some travel brands for years. At KAYAK, machine learning has always played a key role in our product – it’s what powers predictive rate caching for hotels and flights, fare accuracy models and flight price predictions. But with the recent rise of chatbots and voice platforms, it’s being recognized by everyone as a major driver of innovation in the space.
Since 2016, travelers have been able to use KAYAK’s Alexa skill to search for flights, hotels and cars, to explore destinations they can go to within a certain budget, and to track flights. Recently added personalization features give access to KAYAK Trips (i.e., “Is my flight on time?”), allowing users to create price alerts and save search results for later use. Most recently, KAYAK became the first travel company to make booking a hotel room – using just your voice – possible.
We’ve been experimenting and investing in natural language processing and machine learning for several years, taking a leadership role in AI innovation within the travel industry.
DJ: How did you go about developing the technology?
Keller: We decided to invest in in-house technology not only because there were less offerings from when we started compared to the offerings available to us now, but also to fully integrate with other parts of KAYAK (that is our world-class travel search technology and our expertise in building international products).
DJ: Does the technology help businesses to plan their workforce?
Keller: It can. In May 2016, KAYAK launched collaborative travel planning on Slack and in March 2017, we were one of the first bots to launch within Microsoft Teams. As more and more teams globalize, being able to collaborate no matter where you are in the world is increasingly important. With the launch of Microsoft Teams, we wanted to make sure you and your co-workers could making planning your business travel as easy as possible.
DJ: What has the interest been like?
Keller: Out of the platforms we are currently on, Facebook Messenger is used the most often right now, and we’re the second most popular app in the travel category on Slack. It is early days but we are excited about the progress.
DJ: What’s next for AI and travel?
Keller: We’re just getting started. The future of travel is hands-free, simple and seamless and making Alexa a travel expert continues to be our mission, so you can expect more updates to our skill in the coming months.
