HPCC systems and opera solutions deliver a comprehensive suite of tools to support Travel Agents identify and capture new-sell/cross-sell/up-sell opportunities A case study

Summary form only given. New-sell/cross-sell/up-sell opportunity identification and capture for Travel Agents. Volume and velocity are two of the three V's that define Big Data [1]. This is especially true in the global travel industry, where billions of messages are being passed between travel agents and travel providers on a continual basis, arriving at rates of hundreds/thousands per second (this includes for example availability requests, rate information, bookings, changes to booking and cancellations.) This presents a challenge for traditional Business Intelligence reporting based on relational databases since a `state' needs to be maintained continually for multiple millions of bookings, and even more so for advanced predictive analytics where the task is not only to retrieve data, but to compute complex derived variables and statistical models based on continuously evolving patterns in the underlying data. The only solution is to use a distributed platform that can scale to the volume and velocity required [2]. In this talk we will discuss case studies in the travel industry. The goal is to provide a comprehensive suite of tools to support travel agents identify and capture new-sell/crosssell/up-sell opportunities from this continuous flow of data. This includes extraction of `true performance' for each agent/agency necessitating clustering to construct like-for-like peer-groups based on the travel fingerprints for bookings being made, through to the real-time recommendations of specific named hotels to go with selected air segments that is based on sophisticated collaborative filtering as well as a K-Nearest Neighbors approach on a feature space calculated over a rolling period of historical bookings. Extracting these features, what we call signals, brought in the third V - variety: We use information from va