Cendant / Orbitz Affiliate Growth, ML Itinerary Generation & GDS Cleanup One-Pager

Cendant / Orbitz

Technical Marketing / Travel Data / Machine Learning Contributor

Context

Supported affiliate-program growth and technical marketing by developing ML-style methods for generating high-value niche multileg flight itineraries, and contributed to geographic waypoint, destination inventory, and GDS cleanup work to improve the structure and accuracy of location-linked travel inventory used across search, routing, affiliate, and booking workflows.

Problem

Travel affiliate growth depends on capturing high-intent search demand across routes, destinations, trip patterns, seasonal interests, and niche itinerary combinations. Manual content and link-building do not scale well across multileg routes, long-tail destinations, and specialized travel intent. The opportunity was to use data and machine-learning-style generation to identify and create valuable itinerary combinations that could power affiliate traffic and technical marketing.

What I did

• - Supported affiliate-program growth through technical marketing and data-driven travel-content generation.
• - Helped identify high-value niche travel demand around destinations, routes, multileg trips, and long-tail search patterns.
• - Used machine-learning-style or algorithmic methods to generate candidate multileg flight itineraries from travel inventory and geographic data.
• - Connected itinerary generation to affiliate link-building and programmatic landing-page opportunities.
• - Used travel data-quality work, waypoint cleanup, and GDS-linked inventory context to improve route plausibility and destination relevance.
• - Applied filtering and ranking logic to avoid low-quality, implausible, or low-value itinerary combinations.
• - Focused on commercially meaningful travel concepts rather than generic route permutations.
• - Connected technical marketing, SEO, affiliate strategy, and inventory-backed booking experiences.
• - Helped create a scalable approach to long-tail travel-content expansion.
• - Built early experience that later translated into AI prompt generation, schema-driven enrichment, recommendation systems, and automated content workflows.
• - Reviewed and normalized geographic waypoint and destination inventory data — cities, airports, regions, hotels, attractions — to reduce duplicate, stale, or misclassified location records.
• - Connected data quality work to GDS-linked inventory so downstream search, routing, and booking experiences reflected accurate travel metadata.

Outcome

• - Supported affiliate-program growth by improving the ability to generate niche multileg itinerary concepts at scale.
• - Helped connect travel inventory, geographic data, route logic, and search intent to technical marketing execution.
• - Created a practical early example of applied machine-learning-style generation for commercial content and affiliate growth.
• - Improved the foundation for programmatic travel pages, route-aware content, and search-driven demand capture.
• - Built early-career experience in ML-assisted generation, ranking, filtering, technical marketing, and data-driven growth.
• - Established a through-line to later AI work: using structured data and models to generate useful, constrained, commercially meaningful outputs.

Evidence

• - Supported affiliate-program growth using technical marketing and algorithmic or ML-style generation of high-value niche multileg flight itineraries.
• - Connected route logic, geographic metadata, travel inventory, and long-tail search intent to scalable affiliate content opportunities.
• - Built early-career experience in applied ML-style generation, filtering, and technical marketing automation.
• - Public-safe metric language; exact dates, algorithms, itinerary counts, traffic impact, revenue impact, affiliate partners, GDS sources, and proprietary ranking logic require source confirmation before publication.