HQ: London / United Kingdom
Ravelin prevents fraud and protects margins for online businesses. Companies all over the world are accepting more transactions with fewer chargebacks thanks to our unique machine learning-based approach to fraud prevention. By automating standard fraud tasks, fraud teams can spend time focusing on the root causes of fraud instead of day-to-day review of transactions.
- Risk Engine
- Data Provider
- Manual Review Cockpit
- Chargeback Handling
- Callcenter Fraud
- KYC/AML/ID Verification
- Machine Learning
- Device Fingerprinting
- Behavioral Analytics
- Chargeback Guarantee
Fraud Solution Profile
Ravelin extracts patterns of fraudulent behaviour in our clients’ data using a clever combination of machine learning, graph networks, and business rules. By integrating with a business’ website or mobile app to get a real-time feed of customer data, a machine learning model can provide the client with probabilistic scores of the likelihood of the customer being fraudulent. Machine learning allows us to do this in less than 500ms guaranteed, which means our clients can be assured they are being protected in real-time.
We also employ graph networks analysis so we can discover connections within a merchant’s network and across Ravelin’s network of merchants. This allows us to identify and block whole networks of fraudulent accounts, therefore quickly and efficiently stopping fraud from spreading.
Furthermore, it’s the unique combination of the following technologies working together which has lead to our significant success in fighting fraud:
Machine learning: Our machine learning models are the primary way we combat fraud and are effective as they learn what fraud looks like based on a merchant’s historical attacks. These models work in three ways:
- We run global models for details like email address where there is no variance across different industries or geographies.
- We also run industry-specific models based on fraud insights from similar companies we have worked with.
- We learn exactly what fraud looks like for individual merchants and run bespoke models based on historic chargebacks.
Graph networks: Fraudsters rarely attack once. Our networks find linkages between customers so that when you find one fraudster you can block the other accounts linked to that person. This works both within your own data and across our entire merchant data.
Behavioural Analytics: Since we track every interaction users have with our clients’ platforms, we score on hundreds of features extracted from the data they send us. This means we’re tracking not only transactional behaviour, but all customer behaviours prior to checkout.
Business Rules: We know you know your business best, which is why we have partnered a rules system alongside of our machine learning models.
The application of these artificial intelligence techniques and highly efficient graph network approaches is Ravelin’s true innovative USP. We take established academic best practice and apply it to solve real-world problems for our clients.