Case Study: Moz achieves automated fraud detection and auto-banning with Sift

Saving time and resources while fighting fraud

Moz, a marketing analytics company with about 40,000 customers and a 600,000-member marketer community, faced growing payments and account fraud — from credit-card testing to spammy fake accounts and API-key abuse. Their existing rules-based solution required constant manual tuning and couldn’t keep up with increasingly agile fraudsters, polluting order systems and burdening the team.

Moz integrated Sift quickly via API and a JS snippet, sent event data and labeled users, and began using Sift Scores within two weeks. Automated actions tied to those scores let them auto-flag and auto-block bad users, use API logs and network visualizations to spot linked accounts, and cut time spent on investigations — reducing pre-auth activity and preserving the community experience.

Moz

Devin Ellis

Software Engineering Manager