Case Study: Travelport achieves 100x faster, real-time, multi-field traveler search with Elastic

Travelport Powering the search for a 360° view of travelers

Travelport is a global travel-commerce platform that processes more than 200 million bookings a year and manages massive Passenger Name Records (PNRs) on legacy mainframes. While mainframes handled transactions reliably, they couldn't perform complex, real-time searches across free-text metadata or multiple traveler fields, limiting agencies to simple queries and preventing personalized offers despite ingesting up to 850 PNRs per second.

Travelport implemented Elasticsearch (with X‑Pack) to index PNRs, enabling nested and percolator queries that power real-time mobile updates, rich demographic insights, and targeted alerts. The new search layer cut query time from about 500 ms to 5 ms (100x faster), expanded searchable data from a couple of fields to hundreds, and added role-based security—unlocking faster development of customer-facing apps and more personalized travel services.

Travelport

Abdoul Sylla

Sr. Director of Data Products