How the PayPal 2-sided network optimizes fraud prevention

Learn how machine learning and the PayPal 2-Sided Network work together for powerful fraud mitigation. Today’s digital businesses face both opportunity and threat. On the one hand, there is a commercial advantage for merchants who can offer a compelling eCommerce experience. Yet there are also risks: online fraud continues to grow and threaten not only the bottom line for internet retailers, but also the hard-won trust of customers.

The world online

The direction of travel for global retail, even before the pandemic, was online. Strict social distancing rules, long queues, store closures and government lockdowns made shopping in brick-and-mortar stores less and less appealing. As a result, global eCommerce is expected to increase to $8.14 trillion in 2026, growing by 56% since 2021.1

At PayPal, we’ve also noticed a change in buying behavior over recent years. Transactions made with PayPal Checkout increased by more than 35% in 2020,2 and we also saw an 82% increase in recurring payments.3

Yet where there are people and money, fraud will usually inevitably follow. A 2021 survey demonstrated the impact the pandemic had on eCommerce fraud. Three-quarters of online merchants around the world reported a net increase in fraud attempts since the beginning of the pandemic.4 It’s a trend that looks set to continue, with the eCommerce fraud detection and prevention market set to almost double in the years between 2021 and 2025.5

Most commonly, fraudsters will either use stolen card details to make purchases, or hijack existing user accounts to buy goods using stored details. However, so-called “friendly fraud” (aka first-party fraud) is also known to rise during recessions. Here, legitimate customers may claim that a product was never received, in order to keep the item and get a refund.

Such schemes can have a serious financial and reputational impact on merchants, especially SMBs. Legacy fraud prevention tools can even make things worse by forcing excessive customer checks and manual reviews, which can increase the rate of shopping cart dropouts and operational costs.

The PayPal 2-sided network uses machine learning to optimize fraud prevention without requiring any excessive checks or reviews at the PayPal Checkout. Machine learning fraud prevention uses a subsidiary of artificial intelligence (AI). It works by collecting and evaluating huge amounts of historical data to establish new risk rules. These can then be implemented to block certain users or payments, which include suspicious logins, possible identity theft, or fraudulent transactions.

The bottom line

The bottom line is that traditional fraud detection algorithms are only as good as the data you start with. So, begin with data on billions of transactions, tens of millions of merchants, and hundreds of millions of customers and you’re off to a good start. But then what happens? This is where the power of our advanced machine learning algorithms really comes into play.

Our capabilities work to stitch together a 360-degree view of each and every customer using multiple data feeds and intelligence. The application of machine learning models helps to pick out unusual activity that human eyes are likely to miss, enabling the automatic detection and blocking of payment fraud, account takeovers, and more. Machine learning is also used to continually optimize fraud filters, helping to ensure they are as efficient as possible in detecting scams – even as fraudsters change their tactics.

Machine learning fraud prevention can therefore be considered the next step in fighting eCommerce scams. Utilizing the power of AI, PayPal fraud prevention systems give businesses the most effective tools to combat this serious and growing risk.

Making fraud prevention simple

The machine learning process can be a powerful ally to help mitigate fraud. PayPal helps make it easy for merchants to leverage these tools. For example, PayPal Fraud Protection is integrated into the PayPal Commerce Platform. Through an intuitive dashboard UI, users are offered customized fraud filters based on their unique business metrics provided at onboarding. Merchants can test the impact of these filters on historical data prior to activating, and changes are continuously suggested to optimize their efficacy.

This all adds up to powerful fraud prevention, leveraging a vast global network of 2-sided intelligence. With these tools in hand your business can:

  • Minimize third- and first-party fraud losses
  • Reduce false positives and customer friction
  • Streamline the shopping experience for customers
  • Empower users with simple-to-use tools so they can spend more time growing the business
  • Reduce operational costs associated with fraud prevention

These aren’t the only benefits to merchants of joining the PayPal network. New marketing opportunities, higher conversion and authorization rates, plus the value of decades-long relationships with leading banks and card companies could all be major wins for your business. But the potential damage caused by persistent fraud makes our features in this area particularly compelling.

The benefits of machine learning for fraud protection are becoming increasingly apparent as today’s digital businesses face new levels of risk. However, by using the PayPal 2-Sided Network it is possible to mitigate these factors and enjoy the huge potential benefits of the surge in eCommerce.

Find out how your business can grow and mitigate fraud with PayPal.

Was this content helpful?

Sign up to stay informed

Share your email to receive the latest enterprise updates, top stories, and industry reports.


We’ll use cookies to improve and customize your experience if you continue to browse. Is it OK if we also use cookies to show you personalized ads? Learn more and manage your cookies