Your EPS account

Tap into all of our EPS partner resources: update your account details, log support requests, access API documentation and much more

Your email or password is incorrect. Please try again or reset your password.
If the issue persists please contact your EPS representative.

Forgot password?

Reset your password

Please enter your email address below. We will send password reset instructions to the email associated with your EPS account.

Thank you

An email has been sent to your account email address with further instructions on how to reset your password.

Back to sign in


'​The Future of travel'

Aug 8, 2017

TTG MENA interview

Ariane Gorin, Senior Vice President and General Manager, Expedia Affiliate Network Brand, grants Tatiana Tsierkezou (TTG MENA) some insight on EAN’s very latest technologies and future forecasts.


1. How do you predict online travel booking technology will change in the coming five years? Why will these changes happen?

Machine learning, a form of Artificial Intelligence (AI) whereby computers learn without being explicitly programmed with new information, is advancing fast and is a hot topic at the moment. AI relies on the ability to interact successfully – or intelligently – with a consumer in some way.

‘Chatbots’ are particularly visible examples of machine learning at work and are potentially the next evolution in discovery. As the name suggests, chatbots are essentially machines - messenger apps - with which customers are able to have conversations.  Armed with the knowledge of the customer’s past books, the chatbot can offer targeted recommendations highly likely to be converted into sales. The chatbot keeps learning from each booking the customer makes, meaning recommendations become more relevant with every new ‘chat’ and customer interaction.

A report EAN recently launched with Phocuswright called ‘2017 Phocus Forward: The Year Ahead in Digital Travel’, found that chat and voice are new mediums slowly being adopted by travelers. More travelers are comfortable using chat when they travel, but book by chat is still in its infancy. Chat-powered search and customer service, however, will become integral to travel planning. In fact, it has been predicted that voice is going to constitute 50% of all searches in the next five years. Already, Google voice search queries in 2016 were up 35 times higher than in 2008, according to Google trends via Search Engine Watch.


2. What noteworthy trends has EAN picked up on?

We work with a range of travel partners from airlines and travel agencies to loyalty and corporate travel companies all around the world, so it is important to stay ahead of market trends in different parts of the globe.

According to the report we commissioned with Phocuswright, ‘2017 Phocus Forward: The Year Ahead in Digital Travel’, in the most mature online markets, including Europe and the US, travelers are searching and shopping more on mobile, but holding onto entrenched desktop habits. Interestingly however, the Asia-Pacific region, led by China, is really breaking away from the pack in terms of mobile bookings. Also, in less mature online markets such as India, Indonesia and Brazil, where there is no ingrained habit of booking on desktop, it is predicted that mobile will overtake desktop more quickly than in the US and Europe. Another significant change to note is the advent of the millennial traveler, who is 23% more likely to travel abroad than non-millennials. In terms of preferences, millennials are interested in more authentic and local experiences in the travel destinations they seek. 

This trend is especially relevant to MENA, where eMarketer predicts that digital travel sales will top $40 billion by 2020. The population in MENA is young with almost 65% under the age of 30 and 50% under the age of 24.   An extensive Visa study in Saudi Arabia and UAE found that millennials are actively online for between 4.5 and 6.5 hours a day, and according to Google, 82% of under-35s in Saudi and 59% of under-35s in UAE researched their last purchase on a smartphone. It is therefore especially important for travel companies and businesses entering the region to understand that a growing proportion of their target audience are using smartphone for search and purchase.


3. Please tell me a bit about Expedia’s collaboration with Amazon’s Alexa. 

The Expedia skill for Alexa offers a voice-activated method to hear details on upcoming trips purchased on Consumers with Alexa-enabled devices, such as Amazon Echo and Amazon Dot, can use the Expedia skill to instantly hear details about their hotel and flight reservations, easily book a rental car or check Expedia+ rewards loyalty point’s balance with just a few simple commands.
We are delighted with how well our new Expedia skill has been received by Alexa device owners, and this this is just the beginning for us. We’re in conversations with other companies and product offerings about similar ways to use some of our long-term work around Natural Language Processing, or the task of turning travelers’ common questions into searchable content, to enhance voice and chat-based search options. 


4. What’s next for EAN? 

EAN is working on using a type of machine learning called ‘deep learning’ to rank and sort hotel images. We’ve learnt that the very first thing people glance at within a hotel listing, before considering the hotel name or price, is the image. In fact, it takes us around one twentieth of a second to process an image, so the quality and relevance of the images and the order in which they are displayed to travelers is crucial. 

Currently we rely on a manual process to select the featured image for a listing, while the other images are randomly ordered or grouped. EAN has over 300,000 properties, so ranking and sorting the images for these manually is difficult. Enter AI to do this automatically.

The aim is to automatically order and sort the images according not only to image quality, but also to traveler types, customer preferences and seasonality, so that the images most likely to encourage a booking are displayed to each individual customer. Geographic and cultural preferences are also an important part of our dataset.  We work regionally in the Middle East and elsewhere in the world to make sure that all images are handled sensitively.