Big Data Analytics

Agency behavior on GDS

The corporate travel market exceeded $1 Trillion in 2013, and it’s an endlessly expanding market. As the demand for travel grows, so does the need for travel agencies. Travel agencies, in turn, use a global distribution system to find the appropriate service for their clients. On the other end of the spectrum lie the suppliers of these services who have to find a way to optimally use these global distribution systems and incentivize agencies to purchase a service from them.

In this paper, we look at one such travel service market segment: hotel shopping. In such a massive market, how does an agency know what to search for, and how much search effort should they apply before finally settling on a price? We identify search behaviors among agencies while they look for the right hotel room for their client. We also explore whether the ultimate utility to the client as a function of search behavior is diminished so that they pay less for a service as the search behavior changes.

Finally, we identify the trade-off for the global distribution system itself. Unlike search engines such as Google, which benefit from increased searches, these travel information aggregators invest millions on setting up the search infrastructure and maintaining it. Their goal, therefore, is to increase the number of bookings with the minimum number of searches. However, if more searches lead to a better deal for the customer, then agencies are driven to search more, given that they have no other incentives. We are able to demonstrate from a planner’s perspective how this trade off can continue to be economically beneficial.


Techniques and Skills used: SAS, MYSQL, Big Data optimization, 2SLS estimation, Fixed Effects

Working Paper.