Search papers, labs, and topics across Lattice.
This paper introduces Proximity Features, a novel privacy-compliant system for enhancing cold-start personalization in two-sided marketplaces like Airbnb, where many users lack sufficient historical data. By leveraging geo-IP data and an adaptive clustering algorithm, the system aggregates user-level signals from groups of approximately 1,000 nearby users without relying on persistent identifiers, thus adhering to privacy regulations. The deployment of this system has resulted in statistically significant increases in bookings, particularly benefiting users with limited or outdated interaction history.
Aggregating user signals based on geographic proximity can significantly boost booking rates for first-time users while ensuring privacy compliance.
Personalization in two-sided marketplaces relies heavily on user-level features, yet for platforms with infrequent, high-consideration purchases, a large fraction of users lack sufficient history for effective recommendation, spanning both paid and organic channels. At Airbnb, a substantial share of search requests comes from logged-out or first-time users, with this challenge especially pronounced on paid-channel landing pages, leaving traditional user-level features unavailable for a large fraction of traffic. Privacy regulations and increasing restrictions on third-party cookies further limit identifier-based tracking for non-essential use cases. This paper introduces Proximity Features, a privacy-compliant feature system that groups users by geographic proximity using geo-IP data and an adaptive clustering algorithm, producing aggregated user-level signals for groups of approximately 1,000 nearby users without requiring a persistent individual identifier at inference time. Privacy is preserved by design: the pipeline operates on consented, aggregated data only within consent-gated privacy controls. The system is deployed in production at Airbnb, serving multiple surfaces including marketing landing pages and destination recommendation, with engagement emails integration under way. Online A/B experiments demonstrate statistically significant lifts in bookings, with the largest gains observed among users with absent or stale history.
