how to get the full potential out of your risk management solution

Key points:

Commercial risk management solutions come with built-in limitations due to their business model, their technological basis and how they are configured

To get the most out of your risk management solution, we created a Payment Analytics Framework, which recognizes fraudulent behavior and derives recommendations to configure your Risk Management System

This approach reduces the time to configure risk management systems to weeks instead of months and helps cutting fraud by up to 90% while increasing conversion

Today, merchants and banks are facing a different kind of fraud than some years ago. But the means for fighting it have stayed nearly the same.

In the past, addictive shoppers and small criminals were the main source of fraudulent activities in online shops. But today, merchants are dealing with organized groups of fraudsters, equipped with latest bot technologies and access to myriads of stolen payment data.

why your risk engine is not doing its job

Addressing these issues, a market for anti-fraud software flourished. All these vendors have in common, that they evaluate payment transactions within fractions of a second and calculate a risk probability. The result of this calculation leads to accepting or rejecting the transaction – or queuing it for further manual investigation via a risk agent.

With this approach, risk management software inherits major flaws, leading to many wrong rejections and high manual review efforts. But why?

• Business Model of commercial risk management solutions
  Anti-fraud / risk management solutions are configured too strictly due to wrong incentives. They are focused on “stopping the bad guys” instead of “letting genuine shoppers through”. This results in blocking the wrong customers
• Methodology of configuring and running a risk management solution
  Commercial solutions come only with a pre-defined set of rules, limiting pattern-recognition – and leaving it to the merchants to configure them on their own. This is also due to limited staff on vendor side for adapting individual merchant needs, resulting in weak configurations
• Technological Software Basis
  Depending on the software provider, there are only limited possibilities to extend the solution. And in many cases, necessary data like basket information, can’t be included at all.

analytics framework to improve risk management setups

At anlyx, we are choosing a different approach: Instead of focusing only on reducing fraud (and therefore declining too many good customers), we also aim to increase the authorization rate and cutting manual review efforts.

For this, we created a Payment Analytics Framework, which helps us to analyze fraud patterns within seconds and deliver recommendations to improve the merchant’s risk management setup immediately.

For this framework, we are extending the amount of data, used by the merchant’s risk management engine, e.g. through cart information, CRM data or 3^rd-party data like Open Streetmap. This allows us to run a variety of automated analyses to recognize various fraudulent patterns. On top, we are using custom-built algorithms to create recommendations to configure the merchant’s systems. No matter if the risk management engine is based upon a simple score card or is using a machine learning approach.

This helps our clients to:

• Setup individual settings with better performance, instead of relying on standard configurations from the solution vendor
• Illuminate fraud patterns, which otherwise would remain unidentified
• Generate additional insights like payment method optimization or identify issues with technical integration of your payment service provider, acquirers or distinct issuing banks

find out, how the payment analytics framework can improve your risk engine

At anlyx we protect leading brands and small merchants from fraud since 2013. We applied our Payment Analytics Framework world-wide, reducing fraudulent activities by up to 90% while increasing the check-out conversion by more than 25%. If you want to know more about how we apply our methods, read the use case for “[adidas]”. Or see the options for our Managed Risk service [here].