Five reasons why adaptive fraud prevention should be at the heart of digital transformation

Legacy tooling can’t keep pace with sophisticated fraudsters

The underground digital fraud community has grown and matured significantly in recent years. They have numerous dark web forums on which to share knowledge and tools. A wide range of technologies, enabling everything from location spoofing to large-scale bot-driven attacks, has automated and simplified operations. Unfortunately, many fraud prevention systems have not kept pace. They are unable to spot sophisticated and continually changing patterns of malicious behavior, leading to high false positives, end user friction and operational inefficiency.

Machine learning offers a better way forward

Fortunately, machine learning can offer an effective way to tackle contemporary fraud. By training models with large data sets, machine learning-powered fraud detection solutions are able to accurately spot intricate patterns of fraud that humans may miss, and monitor transactions 24/7 to take the pressure off merchants’ in-house teams. Machine learning can be a fast, agile and highly effective way to help manage fraud risk; in fact, some 60% of organizations that use automation, machine learning, or behavioral analytics agree that AI technologies are essential to detecting online fraud incidents.³

An adaptive approach can future-proof risk management

One of the biggest benefits to a machine learning-powered fraud prevention solution is that it can provide value not only today, but long into the future. That’s because the technology can be leveraged to both spot fraud patterns in huge datasets, and also make recommendations to optimize the fraud rules it uses. In this way, the solution can be configured to become smarter with every transaction. As the bad guys adapt their techniques, the machine learning models evolve.

It can save costs and maximize revenue

When it comes to fraud management, the key is to reduce the number of declines without allowing fraudulent transactions to increase. With an effective adaptive risk management approach this is what you can expect: keeping out the fraudsters without incorrectly flagging legitimate customers. This will reduce fraud-related chargeback costs and drive profits by improving authorization rates.

Reduced friction and false declines mean a better customer experience

If your fraud prevention filters are working as they should, you will not only help reduce the number of false declines, but also reduce the volume of transactions flagged for manual review. The latter can add significant extra friction for customers which in itself may lead to cart drop-outs, lower revenues, and possible customer churn. Creating a better all-round payment experience will help build loyalty, engagement, and revenue.

Our 2-Sided Network of over 360 million active consumers and more than 28 million merchants worldwide provides a rich source of data. This data is fed into our machine learning models for more accurate, adaptive, and real-time fraud detection.

PayPal has four risk offerings for merchants:

Chargeback Protection: PayPal will refund the disputed amount and any associated fees (up to a predetermined limit) for any automatically approved eligible transaction results in a chargeback.

Fraud Protection: Fraud Protection is an out-of-the-box toolkit built into the PayPal Commerce Platform and Braintree, designed to help provide merchants with more visibility and control over the transaction decisioning process.

Fraud Protection Advanced: Built into Braintree, Fraud Protection Advanced helps enable a merchant’s fraud team(s) identify and investigate suspicious transactions, analyze patterns, and uncover key insights to mitigate fraud losses.

Risk APIs: Available to extra-large eCommerce merchants, merchants can easily integrate Risk APIs to help guard against several fraud use cases, such as signup, login, and payment fraud.