Page 148 - Building Big Data Applications

146 Building Big Data Applications

Demographic Geographic
Customer Transac ons
Data Data
Channels Agents Vendors Preferences

Revenue Profit Calendar

FIGURE 8.1 Sample data block layout.

In a travel industry perspective, the bottom of this data layer is a swamp as it contains
raw data and dirty data which are not aligned to the user yet. This is the layer for data
discovery and in travel industry the data layer contains searches, links, log times,
clickstream data, reprocess, purchases, repeated searches based on price, calendar,
destination, links to hotels, and reservations aligned with travel, searches for tourism
packages, deals, and incentives. The data can be in multiple languages, have images,
videos, user group comments, independent reviews, and sentiment outbursts. This is the
world of data which we need to build an application for performing data discovery, data
analysis, data segmentation, data classification, and data categorization; data discon-
nects and provide outcomes on how to connect all this data to deliver value. This is the
first application to build.
Once we have streaming data identified as a source, we also see that there are several
opportunities missed where the customer or prospect could have received personalized
services based on their searches and how that tipping point could have shifted the
behavior. Especially in the travel industry, such a personal approach is of vital impor-
tance and the opportunities for delivering analytical services at that layer of personali-
zation to each prospect is very important in the travel industry. If we look at the
conversion rate on travel websites, we see that approximately 92% of shoppers will not
convert and 60% of visitors never return after a first visit. This behavior has been
transformed into a positive outcome with engaged stay and in many cases a purchase by
the shopper, all of this happening with the utilization of big data and analytics across
online travel companies to deliver the right message at the right time to the right person,
and provide services which ultimately deliver revenue.
The application storyboard for the operational and raw data analytics including
streaming and online data, is that knowing what your customers like, or do not like, can
have a severe impact on your brand, its services, and revenue. Everyone shares
information on social networks, especially personal travel stories. Collecting and
studying this data with a neural network algorithm for text analytics will deliver vital
information that can be used to provide a user a more tailor-made message. The goal of
the storyboard is to increase stickiness and realize revenue. The best example of this is
the “tripadvisor” website.