Fraud detection for
eCommerce businesses
Protect your online business in real-time
Powering
today's plugins
today's plugins
Protecting billions of
transactions annually
transactions annually
Used by online
businesses worldwide
businesses worldwide
With minFraud®, you can: analyze risk, prevent fraud, reduce chargebacks, and cut time spent on manual review.
Ready to use
from day one
from day one
Data analyzed
with machine learning
with machine learning
Enhanced with expert
human heuristics
human heuristics
Detect high-risk IP addresses
- IP geolocation and context
- VPN and proxy detection
Analyze payment card details
- Prepaid, business, and virtual card detection
- Issuer, card brand and type
Flag suspicious email addresses
- Free and disposable email detection
- Reputation data for email addresses
Assess billing and shipping address risk
- Billing and shipping address distance
- IP geolocation and card issuer billing verification
Identify devices
- Device ID information
- Uncover proxies and high-risk IP addresses
80+ data attributes, and one of the largest fraud detection networks globally
IP address
- Geolocation
- Confidence
- Anonymizers
- Reputation
- Risk reasons
- User type/count
- Static/dynamic
Email address
- Reputation
- Email first seen
- Domain first seen
- High-risk email
- Disposable or free email
Device
- Device ID
- Device local time
- Device last seen
- Device confidence
Phone number
- Reputation
- Original operator name
- Fixed/mobile
- VoIP
Credit card
- Card type
- Business card
- Prepaid card
- Virtual card
- Issuer information
Shipping address
- High-risk shipping address flag
- Distance to billing and IP addresses
- IP geolocation matching
Make smarter and faster decisions with risk scoring
Identify high-risk IP addresses with IP risk scores
- Assess the reputation of IP addresses, emails, shipping addresses, phone numbers, and devices
- Perform velocity analysis specific to your business and across the minFraud network
- Detect proxies and verify geolocation
- Access transaction feedback reports
Build custom rules with custom inputs
Customize rules and review transactions unique to your business
- Automatically accept, reject, or send transactions to manual review
- Facilitate fraud analyst review alongside minFraud data
Boost your fraud analysis with device tracking
Identify a device and track its usage
- Identify and track devices to detect fraudsters who use anonymizing proxies to cycle through IP addresses
- Link suspicious activities and flag the device and related IP addresses as high-risk
Understand the reasons behind a risk score
Gain insights into a transaction's risk score
- Use risk score reasons for manual review and risk modeling
- Perform additional checks and determine the validity of competing signals
- Avoid double-counting specific risk factors from your other data sources
Your anti-fraud journey starts here
Plans and pricingResources
Guides and help articles
Visit our knowledge baseTechnical details
Visit our developer portalFrequently asked questions
Machine learning, a subset of artificial intelligence, refers to machines that can learn from patterns and data. They improve and adapt over time, based on the data they receive and the environment in which they operate. This enables them to excel at tasks like recognizing patterns and adapting to new factors as they arise. In simpler terms, machine learning involves algorithms that continually learn and evolve, becoming highly effective as they encounter new trends and patterns. The minFraud network is powered by the collective input of all minFraud users, along with AI-driven machine learning and data analytics. The minFraud risk scoring algorithms prioritize performance to ensure low inference times and methods that offer a high level of explainability.
API stands for Application Programming Interface. It allows different systems to communicate with each other. An API can integrate a dataset into an existing application, or enable two applications to share information seamlessly.
A transaction risk API assesses the risk associated with transaction data like IP, email, credit card, device, physical address, and more, to return a risk score and risk data that can be used to automate transaction approval/denial and provide resources for your fraud prevention team. The minFraud service is a transaction risk API for assessing the likelihood of fraud in transactions.
The minFraud service returns an overall risk score, which uses an ensemble machine learning approach enhanced by continually evolving heuristics developed by fraud experts . The model behind the risk score adapts to new fraud patterns over time based on the data and feedback you provide. You can use risk scores to determine how to respond to a transaction, whether to accept, reject, or review it more carefully, or to complement your existing fraud models.
Custom rules allow you to automatically set a label for your minFraud transactions based on knowledge and experience about fraud patterns that affect your business. When you identify a pattern in fraud affecting your business, you can use custom rules to capture all transactions that match this pattern and apply a label of “accept,” “reject,” or “manual review” to those transactions. These rules do not affect risk scoring. Find out more about custom rules.