paysafecard.com is an online payment method that allows users to pay for goods and services securely and privately at a huge range of online merchants throughout the world in 43 markets. paysafecard vouchers are sold at more than 500,000 retail outlets, gas stations and grocery stores, providing a simple, prepaid alternative to online payments. This also opens online payments up to the unbanked and underbanked, negating the need for a bank account.

MyPaysafecard offers users the opportunity to extend functionality to include a personal payments account to manage balances and replenish accounts.

A Challenging Fraud Landscape

paysafecard.com has recognized the need to understand regular customer behavior. The world of fraud is a very fast-paced one, and fraudsters are constantly on the lookout to find new loopholes. The way that rules can be written nowadays by fraud detection teams has also evolved significantly. Fraud rules have, until recently, been simple “If-Then” solutions, and fraud detection also relied heavily on malicious device recognition.

“ThreatMetrix has delivered remarkable developments of its product which has coincided with paysafecard deciding to intensify the invested resources into the evolution of the rules engine. The result is we have achieved major targets by reducing fraud significantly.”

Hany Razi, paysafecard.com

Designing a Solution

paysafecard is now able to create far more sophisticated rules by setting dynamic thresholds (Smart Rules), creating digital personas (PersonaID) and therefore better differentiating between the good and the bad.

Reducing friction for customers was the first step in this process: the focus of the Fraud team was to apply positive scores for good, returning customers. By doing so, paysafecard was able to reduce the number of false positives by about 15 percent while growing the business at the same time. This lead to happier customers while simultaneously reducing operational costs.

“We relied heavily on the Persona ID functionality for convergent behavior of customers for the first stage,” says Hany Razi of paysafecard.com. “Analysis of customer behavior showed that returning customers showed highly predictable behavior, such as purchasing our product at the same POS, same merchant and so on. In the next step, we tried to identify high risk, divergent behavior utilizing the same technology. Previously unknown devices utilizing a wide range of POS and entering our systems with a high-risk transaction are raising our suspicion right from the start and can be identified via ThreatMetrix in a reliable way.”

“ThreatMetrix enables us to take a surgical approach, where we can cut out malicious transactions without affecting good, healthy customers. We are now seeing the first effects and the increased value of the Digital Identity Network created by ThreatMetrix.”

Hany Razi, paysafecard.com

Leveraging Knowledge With ThreatMetrix Trust Tags

Trust Tags are digital labels that can be applied to various combinations of entities within a user’s persona to indicate their trustworthiness, reducing friction for legitimate users and more accurately identifying high-risk behavior.

paysafecard uses this technology to mark good, returning customers, in order to reward them with an even higher positive score. Trust Tags have also enabled paysafecard to combat fraud far more effectively, as the complexity of rules has increased significantly.

“We look forward to each product release because it’s like Christmas for us. Every enhancement gives us an additional layer of complexity to make better fraud decisions.”

Hany Razi, paysafecard.com

“We are now using intelligence out of multiple touch points in order to promote better decision making,” says Razi. “Trust tags are used as a sort of “Dye Pack”, like banks might use. High-risk events are being marked and malicious transactions can be identified and verified later on during the payment event. Once the transaction is recognized, we place a very sticky fingerprint on all identifiers used by the fraudsters, helping to combat fraud.”

From Device-Based Decisioning to a Behavior-Based Approach, Enabled by ThreatMetrix Smart Rules

In order to decide where to place the Trust Tags in the system, paysafecard relied heavily on Smart Rules, which have enabled the company to assess risk by creating dynamic user behavior models. “Anomalous behavior is more complex than ever, and the capability to adjust to this fact by having a dynamic threshold significantly impacted the way we design rules today within paysafecard,” says Razi. “While everything was customer centered in the past, we can now also evaluate on a Merchant/Distribution level. A sharp increase of events is now very likely to be detected while slow, steady growth does not force us to constantly change the thresholds.”

Request a Demo or Meeting

Start Today
close btn