How do I adjust my fraud risk blocking threshold?

The latest version of Adyen’s risk engine uses machine learning models trained on data from billions of transactions processed globally through the Adyen platform.

There are two machine learning models responsible for identifying and blocking suspicious transactions:

• Bot attack risk model: Detects the likelihood that a payment is part of a card testing attack; available in both premium and basic risk.
• Fraud risk model: Detects the likelihood that a payment is fraudulent; available in the premium version of the risk engine.

The models screen transactions before authorization using the information provided in the payment request and historical payment data in the Adyen Ecosystem. Transactions are blocked based on the probability that they are fraudulent. 

Only available when using the premium risk engine, the fraud risk model classifies each payment into its relative risk level: Very low, Low, Medium, High, or Very High. Next, the transaction is compared to the blocking threshold configured in the Risk Profile. If the level of risk exceeds the profile block threshold, the risk engine blocks the transaction pre-authorisation. Note that the blocking threshold of the bot attack risk model can't be adjusted.

By default, the Machine Learning Block rule is set to use the High-risk threshold. This avoids blocking genuine shoppers while keeping fraud at acceptable levels. Should this setting not provide adequate results, we offer the option to use the different thresholds. The higher the risk threshold, the more risk you're willing to take on. Please remember that not all fraud cases will be within the scope of our machine learning support; in this case, you can create a custom Block rule using your domain expertise.

Adjust the risk threshold

To view or change the risk threshold for your risk profile:

1. Log in to your Customer Area.
2. Go to Revenue & risk > Risk profiles.
3. Select a risk profile.
4. Select Machine learning > Fraud risk > Adjust blocking threshold.
   
   

   
5. View the current block threshold, or adjust it using the slider and click on “Save”.