Cloud Computing

Making Travel Personal Again With Watson

Share this post:

Did you know that the average traveler visits 38 sites before making a booking?* Or that 44% of corporate travelers spend up to 60 minutes booking travel during working hours?** This is considerable lost productivity.

Online travel agencies have focused their resources on optimizing the booking experience on their sites for years, and have become very good at it. Today, true increases in conversion rates for these companies are only incremental, a product of cosmetic upgrades such as a change in button color, button size or an improvement in the call to action and copy. is famous for running 1,000 conversion experiments a day.**

The challenge of making the simple online booking has been solved. But how are customers to know what hotel is in fact best suited to their needs? A quick input of destination, check-in/check-out dates, number of travelers, and number of rooms will yield an overwhelming list of results. A basic search on a major travel site for Rome will return 3,000 hotels!

Consumers want choice, whether it’s what they eat, how they spend their free time, or where they travel. Travel sites must now have large inventories to satisfy the wide-ranging needs and tastes of their customers. A site that can immediately present highly relevant and personalized results will have a significant advantage over sites that offer generic search results. Finding the perfect hotel should be convenient.

As a travel technology company, ZUMATA is deeply engaged in the entire travel process—from the sourcing of hotel inventory, shepherding the customer’s online booking journey, to after-sale customer support. Our technology is used by some of the leading travel companies in the world, including DHISCO, Expedia and Amadeus. We continue to seek ways in which we can complement the existing and future needs of our clients, while trying to solve the largest challenges the industry faces through innovation.

We have a wealth of content and user data, and strive to use the right tool for the job. With that mission in mind, our team has started to explore how to solve the travel customer’s problem of personalization, relevance, and convenience. After much research, we concluded that IBM Watson was the clear choice to help us address the problem of time and customization as it relates to finding the perfect hotel for the customer.

The products we have built with Watson encompass all areas of our expertise. We use Watson Alchemy API, Retrieve & Rank API, the Image Recognition API to help us to identify the best available content (images, descriptions, surrounding activities, sentiment, and review, etc), the most relevant hotel results, delivery of an enhanced user experience through various channels and devices (mobile, messaging, desktop), and even assistance to customers after booking. For its part, Watson continues to learn from previous interactions, (self) improving the technology, and building an insurmountable competitive advantage for ZUMATA over time.

THINK travel photo

On June 9th, IBM launched its Watson Centre in Singapore where we showcased how our existing Hotel API—which includes over 500,000 hotel properties and an enormous amount of images, descriptions, and user data—could be combined with Watson. Our demonstrations featured Facebook messenger bots that used Natural Language to search for a hotel booking as well as communicate customer service queries.

The Facebook Messenger bot, integrated by mobile specialist DMI, allowed customers to ask questions such as, “I need a luxury hotel in Sydney with a view of the opera house, an indoor pool and free wifi.” The hotel results returned will display images with the view from the hotel, its amenities, and even sentiments about the hotel from other travelers. This new level of individualization cuts the research time for the customer and improves conversions for travel sellers.

We also worked with Qualitance, to integrate our cognitive customer service agent bot onto both Telegram and Facebook messenger in order to demonstrate how portable our technology is between platforms. The cognitive agent allows customers to ask questions related to their booking, as if they were talking with a live human agent and enables instantaneous access for customers.

No more waiting on hold or waiting for the chat agent to reply to ten other customers. By integrating a combination of Watson Bluemix services like Dialogue service, Natural Language Classifier, customers can communicate with our partner companies and make any customer service related enquiry they like, even after placing a booking, and have action taken on their behalf. As the agent is connected to our Hotel API, requests such as cancellations, amendments and special requests can be addressed immediately. This not only improves the customer experience, but also dramatically reduces operating costs for our partners.

It truly is an exciting time for travel technology and hotel distribution. We believe that with the power of Watson we will be able to make travel personal again.



Director Business Development and Strategic Partnerships, Zumata Technologies

More stories

Toronto Raptors Use Data-Driven Command Center on Path to NBA Finals

For the first time in their 24 years as an NBA franchise, the Toronto Raptors are in the NBA Finals, squaring off against the Golden State Warriors in the best of seven games series. The Raptors are going up against one of the most talented and accomplished NBA teams. The Warriors are the defending NBA […]

Continue reading

Medicine and the Message: Consumer Engagement and the Power of Nontraditional Data

The journal Health Affairs estimates that the U.S. spends nearly $500 billion annually on pharmaceutical treatments. Focusing on clinical value alone isn’t enough to meet business growth targets — customer experience plays a critical role in long-term brand success. But experience doesn’t occur in isolation: To win trust and boost spending, pharma brands must use […]

Continue reading

How IBM is Using AI to Speed Partner Lead Sharing

Passing quality sales leads from our salespeople to partners is crucial to bolstering a thriving partner ecosystem, and data shows that the faster we can pass these leads on, the greater the chance of closing the deal. According to a 2017 study by IBM’s Chief Analytics Office (CAO), sales leads that are passed within a […]

Continue reading