In tech circles and beyond, AI is the mot du jour lately, often thrown around in speculative conversations as the magical key that will unlock previously unthinkable technological progress. While often such conversations involve long-horizon views, AI has already made a tremendous impact on the consumer experience. This impact is continuously growing and becoming increasingly prevalent in the travel industry.
Consider, for instance, chatbots and voice-activated assistants. Chatbots are already helping airlines and online travel agencies handle some of the simpler customer transactions (booking, boarding passes), so their human agents can focus on more complex interactions. KLM Airlines, for example, was one of the first to allow travelers to request their boarding pass within Facebook’s (News - Alert) Messenger app. Taking it a step further, other airlines have recently begun letting passengers book flights using the Facebook Messenger interface.
On a similar note, travel companies such as Kayak, have begun using voice-activated assistants (Amazon’s Alexa in Kayak’s case) to offer hotel search and book functions. Between Amazon’s Alexa, Microsoft’s Cortana, Google’s Voice Assistant, Samsung’s (News - Alert) Bixby Voice, the market of voice-activated assistants is definitely heating up. Even more so in travel, where large brands, such as Marriott, are not only testing out the general-use assistants, but also developing their own signature ones. Cosmopolitan of Las Vegas recently unveiled its new AI concierge, “Rose.” This is both a brand differentiation effort and a way to improve guest services; after all, Rose never sleeps and is marketed with a catchy personality. And they’re not alone, Edwardian Hotels have launched their chatbot “Edward” and Hilton Worldwide teamed up with IBM (News - Alert) to develop their AI-based hotel concierge “Connie.”
As AI-based services mature, they have the opportunity to simplify and personalize the user experience throughout the travel lifecycle, from facilitating searching and booking, to enhancing the user experience during travel and, finally, collecting feedback post travel. For example, besides helping a traveler search and book her travel, the same voice-activated assistant could detect a guest’s profile upon arrival to her destination hotel and tailor her experience to her preferences. One can definitely imagine Alexa, or another voice assistant, detecting a user, alerting her to any important notifications, and suggesting a restaurant or a local event that matches her preferences.
Overall, whether branded or not, done right, AI could be a great way to deliver exceptional experiences for guests who would welcome digital brand interactions. But full, successful implementation of AI into the travel search and booking process will require infrastructure work to accelerate the machine learning process. After all, AI is only as “intelligent” as the data used in its creation, so automated data collection, organization, and usage are crucial to its continuous improvement and ultimate success.
One of the challenges of machine learning is getting feedback, while not being overly intrusive. Integrating voice assistants and apps could achieve just that. Imagine the voice assistant getting feedback on an earlier suggested restaurant, after verifying, through cross-referencing with the guest's mobile device, that the guest indeed dined there. Such feedback could be used to enhance a guest's profile and accelerate the improvement of the assistant’s future interactions. As this process of learning becomes ubiquitous, the capabilities of voice assistants will grow. The more we can thread the needle between data collection, learning, and privacy, the quicker AI will be adopted and the larger its impact will be.
About the Author
An academic turned entrepreneur, Carl is an expert in using data, algorithms, and machine learning to deliver compelling business value and drive company growth. He leads Sojern’s data science team and focuses on leveraging the company’s rich data assets to develop new products and maximize online ad efficiency. Prior to Sojern, Carl led RTB optimization at Nanigans and managed the build-out of their RTB machine learning pipeline and bid evaluation tier. Before Nanigans, he held senior leadership positions managing engineering teams at inPowered and Kayak and worked on cybersecurity research at BBN Technologies and Intel (News - Alert) Research. Carl earned a PhD in Computer Science from MIT.