trivago builds ML anomaly detection for real-time partner internet traffic monitoring
trivago's highly dynamic environment—with daily product changes, hundreds of partners each having multi-level parameters that shift dynamically, and billions of partner-hotel-locale combinations—makes it difficult to distinguish genuine traffic anomalies from expected parameter-driven changes, while data outages across a large partner base add further complexity.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Historical traffic as training input
Historical partner traffic data serves as the training input for the time series anomaly detection model.
Tools used
ProphetARIMA+Elastic
Outcome
trivago deployed the Prophet model in production for ongoing partner traffic anomaly detection, implemented 24/7 monitoring for large partners and customized automated reporting for smaller partners, resulting in improved data quality and a broader selection of rates.
What failed first
Advanced machine learning solutions proved unhelpful for detecting anomalies at the granular partner-hotel-locale level due to excessive false positives, and the Elastic solution was not adopted because it lacked easy customization.