Case Study: Travelmob achieves fake-listing prevention and slashes chargebacks with Sift

How Travelmob stopped fake listings from harming customer experiences

Travelmob, a marketplace for unique short-term rentals across Asia Pacific (later acquired by HomeAway), was facing a rising tide of fake listings and credit-card fraud that threatened guest trust and growth. With no dedicated fraud team, manual review processes couldn’t scale and fraud was slipping through the cracks, including bad actors posing as hosts and soliciting offsite payments.

Travelmob integrated Sift’s machine-learning fraud solution via a simple REST API in hours and was fully live within a week. Using Sift Scores to prioritize reviews and Network Visualizations to link fraudulent accounts, the team nearly eliminated fake listings and stopped chargebacks; overall fraud dropped dramatically and detection accuracy improved within days as more data was fed into the model.

Travelmob

Jan Hecking

Principal Software Engineer