Learn about SEON
Overview
Fraud is a complex problem to which SEON is a comprehensive solution. Whether you want to stop multi-accounting, block bonus abusers, protect your customers from scams, reduce your chargeback rates, or learn more about your customers, we've got you covered.
You must be new here
SEON Docs is all about helping you get the best results with SEON. You can find interface overviews, and in-depth manuals of SEON's key tools and functions alongside guides and tips to help you fight fraud successfully.
We're here to help you take SEON for a spin. Visit seon.io if you want to learn more about why you should try it. It is free, you know.
SEON at a glance
SEON is a supreme cocktail of intelligence with the perfect mix of data enrichment, device fingerprinting, advanced logic, and machine learning tools crafted to give you that perfect taste of being on the safe side of fraud.
Connect SEON to your systems and send us any data points you collect about your customers, say their email address, phone number, and IP address. We'll take that data and create hundreds of data points.
That data will then feed into what we call the Scoring Engine to help you make the right decision about every user action. You can customize the logic that analyzes each transaction by creating and editing rules.
Our machine learning tools will also help you decide, either by recommending new rules (whitebox for transparency) or giving you a fraud probability score (blackbox for speed and automation).
With SEON, you're always in control. You choose how heavily to automate the process. Want to review each transaction manually? Feel free. Want to create advanced rules to catch sophisticated fraudsters automatically? Go ahead. Want to leave everything to the AI? Well, they're taking over sometime anyway...
Why data enrichment?
The premise is simple: people use the same phone number and email address for years. Over time, they leave breadcrumbs online. These can help us prove they're real people. They create social media accounts, sign up for digital services, and get caught up in data breaches.
Fraudsters don't have that history. Their newly minted email addresses and burner phone numbers have one purpose: stealing from you. SEON helps to ensure that doesn't happen.
Rules and Logic to keep you safe
With rules, you tell SEON what to do when certain user actions (transactions) meet a preconfigured set of characteristics. For example, we know email addresses that haven't been seen elsewhere online are risky. In rule-speak this means: If email domain is a free provider, no online profiles were found, and the has not been involved in a data breach we'll add a fraud score of 10 to a transaction.
You can get started straight away with SEON's Default Rules. Once you're comfortable, you can dive into the deep end and start creating Custom Rules to counter fraud unique to your business.
You can use custom rules to match two data points in the same transaction or compare them to another transaction or set value. You'll find velocity rules super useful to analyze user behaviors and catch suspicious actions.
State-of-the-art device fingerprinting
SEON's device fingerprinting is designed to catch fraudsters trying to spoof your systems. Rely on four different hashes and hundreds of data points to unlock one-of-a-kind insights into your customers' gadgets.
We can help you see if a user is logging in through an uncommon browser or using plugins often favored by fraudsters.
We can also show you when fraudsters do everything they can to hide their true identities by recognizing Tor nodes, virtual machines, or uncommon hardware configurations.
Data crunching superheroes – Machine Learning tools
Machine learning algorithms identify the patterns and typical behaviors behind fraudulent transactions and help you catch them earlier with improved accuracy.
By processing hundreds (or thousands) of approved and declined transactions, our machine-learning solution can discover connections you didn't even know existed.
Our machine-learning models will create new human-readable rules (whitebox) or assign a fraud probability score to each transaction (blackbox). No matter which one you use, or if you choose to rely on both, we'll always help you understand why our AI tools reach the decisions they do.