Sift

Overview

Sift (formerly Sift Science) is an enterprise digital trust and safety platform that leverages real-time machine learning to detect and prevent online fraud. The system analyzes over one trillion annual network events across global digital properties to generate dynamic risk scores for user actions. By continuously processing behavioral patterns, device fingerprinting, and account interactions, Sift protects businesses against payment fraud, account takeovers, and fake user signups.

Founder story

Founded in 2011 by Jason Tan and Brandon Ballinger in San Francisco, California, Sift was created to democratize machine-learning-driven fraud prevention. The founders recognized that smaller and mid-sized e-commerce platforms lacked the extensive engineering resources required to build complex risk engines. Sift abstracted this infrastructure into an accessible cloud platform that safeguards digital transactions.

What it does

• Calculates real-time risk scores using global machine learning consortium models
• Detects account takeover attempts by monitoring user behavior and device signals
• Prevents payment fraud and chargebacks during online checkout transactions
• Automates custom workflows and review queues to lower manual fraud monitoring.

Who it's for

• E-Commerce & Digital Merchants
• Fintech & Crypto Platforms
• On-Demand Delivery & Travel Services
• Online Marketplaces & Communities

Why it works

• Global data network feeds over 1 trillion annual event signals into predictive models
• Machine learning automatically adapts to emerging fraud patterns without rigid manual rules
• Real-time decisioning APIs deliver dynamic scores in milliseconds during critical user events
• Comprehensive coverage across account creation, login, and checkout prevents diverse attack vectors.

Growth strategies

• Expanding international enterprise sales targeting high- volume fintech and marketplace verticals
• Network effect acquisition strategies leveraging shared intelligence across member properties
• Strategic platform integrations with payment gateways and risk management infrastructure
• Continuous product expansion into dispute management and account takeover protection

Alternatives

• Riskified
• Kount
• Signifyd
• Fraud0

Comparison overview

• Sift utilizes a shared global consortium network that evaluates machine-learning signals across over 34,000 sites.
• Unlike chargeback-guarantee solutions that take full liability, Sift focuses on customizable AI risk scoring and workflow automation.