v2.1
Solutions
API DOCUMENTATION
Products
GUIDES
Fraud Prevention
Powerful tools to support fraud fighting professionals.
Overview
Protecting against fraud while at the same time ensuring customers have a positive shopping experience is a key objective in any successful payment strategy.
To help prevent fraud from happening, dLocal evaluates every single transaction that it processes using the data provided, and determines whether there is a high probability of it being fraudulent.
In such cases dLocal may decline these transactions, and in doing so ensures that:
- Chargebacks and disputes are reduced. In addition to the cost of the lost charge, chargebacks and disputes also imply additional costs to merchants for handling, fees and fines that may apply, and the reputational damage on the merchant's brand.
- Long-term conversion rates are maintained or improved. By keeping chargeback rates low, acquirers and issuers will be less restrictive with their own fraud controls. Since these industry participants have less visibility on merchants' business models and data, the controls usually configured at these levels act as broader sweeps which cause an adverse impact on conversion rates, and are often difficult to reverse.
Data is key
All fraud prevention techniques rely on the data that is included in the payment request. Depending on your industry and business model, some of the data that can be obtained and shared is very relevant for the applied fraud prevention controls. To check whether you are sharing the right data, please see the Requirements per Industry section.
​
Beyond the required and recommended data, as a rule of thumb the more data is shared, the better prepared our algorithms will be to detect new fraud patterns. We recommend that our merchants also check the full specification for fraud-related fields in our Risk Data documentation.
dLocal Defense
Our dLocal Defense Suite is an enhancement over the standard fraud controls applied. dLocal Defense relies more heavily on data enhancement techniques and ML models to identify market and industry-specific fraud patterns and prevent potentially fraudulent transactions from being processed.
Last modified 1mo ago
Copy link