As it is becoming more and more important to glean insights from big data, companies are increasingly seeking to use Hadoop. Mariam Sharp reports

So what is Hadoop? Apache Hadoop is one of the most used technologies for big data. Created in 2005, it is an open-source software framework for storage and large-scale processing of datasets. It helps companies understand the complete value of all their data while extracting insights from previously out-of-reach data to drive fresh revenues. The fact that the data is stored across large clusters of computers means that it can work with much larger datasets and at much faster speeds.

But how are travel brands using Hadoop? Let’s take a look.

1) To manage unstructured data

Travel websites are extremely popular online, however these websites generate extensive quantities of unstructured data, and are chockablock with photos, reviews, comments, blogs and other sources of valuable consumer information. Hadoop was designed to process both structured data - such as pricing, customer addresses and web analytics, and unstructured data such as videos or photographs. Indeed, so much available information is contained within data sources such as videos, audio files and email.

2) To boost marketing and sales

Big data analysis is also important in marketing. In an interview with EyeforTravel Jonathan Isernhagen, director of marketing analysis, Travelocity was asked: “How can big data’s power improve marketing initiatives?” His response: “We are closer than ever to realising John Wanamaker’s dream of discovering which half of his advertising budget was wasted. ”That’s pretty powerful stuff! He continues saying: “Big data in the Hadoop/Revolution Analytics sense of the term enables channel attribution and predictive modeling—analysing every interaction that produces a transaction and every interaction which doesn’t, sampling from among billions of rows, without necessarily shelling out for a huge structured data warehouse and a SAS licence.”

3) To understand the importance of quality over price

Orbitz Worldwide, found through their use of the Hadoop system that Mac users tended to pay more for their hotel rooms, and did not just differentiate on price. This led to a Wall Street Journal article in June 2012 that got wide press coverage sensationalising the fact that Mac users paid more for hotels when booking on Orbitz. The insight was gained after collecting 750 terabytes of unstructured data on consumer trip-planning behaviour, using a cluster of 100 servers housing the Hadoop database.

4) To gain fresh insights

Another example of Hadoop usage comes from Disney. This giant of the entertainment and travel industry collects data to gain insights via audience analysis & segmentation, a recommendation engine and analysis of in-park traffic flow. They started small, building their big data platform like a startup builds a sustainable company; with a small and flexible team, failing fast and early, but improving along the way. Disney created a ‘big data’ platform based on Hadoop that was complemented by a suite of other tools. They moved to their first Hadoop cluster in 2009 to a complete Data Management Platform in 2011. Disney take this so seriously now, that they were in the news recently for investing a £1bn into a customised big data system.

5) To inform strategy

On the destination front, Japan Tourism Agency has implemented a massive project that will include the collection of data from mobile GPS and social media streams from about 700,000 tourists across the nation. This will be used to build their strategy with the ultimate goal of boosting the number of overseas visitors by an additional 30 million by 2030. They are forming a panel of experts in using big data analytics to establish a framework based on Hadoop.

Are you using Hadoop? If so how? Or perhaps you’ve discovered another useful technology. If so share your thoughts in the comments box below.