Millions of businesses rely on Stripe to accept payments safely. Our goal is to help those businesses maximize payments from legitimate customers, while minimizing their losses to fraud.
Online fraud causes unnecessary losses for most online businesses; for some, it can be so expensive that it threatens their viability. Accordingly, Stripe has been investing in industry-leading fraud prevention technology since 2015. Radar uses advanced machine learning (optionally augmented by human-supplied rules) to shield Stripe users from hundreds of millions of dollars of fraudulent transactions every month.
As the online economy grows, online fraud is becoming more sophisticated, and Stripe dedicates large teams to improving Radar’s accuracy and building defenses that better-protect Stripe users.
As part of Stripe’s fraud prevention, we use fraud signals detected by Stripe.js and our mobile SDKs. These signals are most commonly used to identify “bots”—scripted web browsers that make fraudulent transactions. Signals we pay attention to include things like screen resolution and browsing patterns. Stripe’s ability to use these real-time signals is part of why Radar is so effective: in a domain where even small improvements can be a big deal, they reduce the expected number of fraud-driven chargebacks a business will receive by 11%.
In response to some recent questions about how we use signals gathered via Stripe.js and our mobile SDKs, we have made some updates to provide additional clarity:
- We updated our privacy and cookie policies to make it clear that our advanced fraud detection signals are used only for security and anti-fraud purposes. Stripe doesn’t—and won’t—share or sell this data to advertisers.
- We put up a page with more details about the advanced fraud detection signals we collect.
- We’ve added the ability for site and app owners to disable advanced fraud detection signals.
We hope these updates make our commitment to our users even clearer. If you have any questions, please don’t hesitate to reach out.