The product
trivago is a widely known global hotel metasearch.
Trivago’s hotel search allows users to compare hotel prices in just a few clicks from more than 300 booking sites for more than 5.0 million hotels and other types of accommodation in over 190 countries, helping millions of travelers each year compare deals for hotels and accommodations.
My Role
As the Product Owner of the Core Ranking team, I was responsible for the back-end application that exposes and ranks the hotel on every single search result.
The position was a highly technical role that involved: the understanding of machine learning algorithms, data analysis, high-traffic A/B tests, and coordination with other product teams.
Our everyday challenge was to come up with new ideas to improve our algorithm, and in consequence maximize the value of search results for trivago, advertisers, and travelers.
An example worth mentioning was the project to adapt our algorithms to alternative accommodations in search results. It involved testing our machine learning model to a sort of inventory that differs significantly from traditional hotels regarding room configurations, capacity amenities, and pricing.
Trivago has a long-lasting history with the use of data and I highly recommend you read their tech blog to get a deeper view of how trivago uses data to improve its product.
I also found this article which provides an example of how a ranking algorithm works.