Amazon Fraud Detector

Published in

AVM Consulting Blog

4 min readJun 11, 2023

Amazon FAmazon Fraud Detector is a fully managed service that makes it easy to identify potentially fraudulent online activities such as online payment fraud and the creation of fake accounts.

Globally each year, tens of billions of dollars are lost to online fraud. Companies conducting business online are especially prone to attacks from bad actors who often exploit different tactics such as creating fake accounts and making payments with stolen credit cards.

Companies typically use fraud detection applications to identify fraudsters and stop them before they cause costly business disruptions. However, these applications often rely on business rules that don’t keep up with the changing behaviors of fraudsters.

More recent fraud detection applications have tried to use machine learning. But, they frequently use a one-size-fits-all approach based on general data sets and fraud behaviors that aren’t specific to your business, which limits their accuracy.

Fraud Detector overcomes these challenges by using your data, machine learning (ML), and more than 20 years of fraud detection expertise from Amazon to automatically identify potentially fraudulent online activity so you can catch more fraud faster.

Detect common types of online fraud

New account

Accurately distinguish between legitimate and high-risk customer account registrations so that you can selectively introduce additional steps or checks based on risk.

For example, you can setup your customer account registration workflow to require additional email and phone verification steps only for account registrations that exhibit high risk characteristics.

Guest checkout

Spot potential fraudsters even among customers without a history of transactions. Customers who transact regularly typically use a registered account. As a result, you have a history of transactions which makes it easier to identify potential fraud.

Guest checkout, on the other hand, has no historical account usage or user behavior data which makes fraud detection much harder.

With Amazon Fraud Detector, you can send as little as an email and IP address from a guest checkout order to assess its potential fraud risk so you can decide whether to accept it, review it, or collect more customer details.

Try Before You Buy’ service abuse

Identify accounts that are more likely to abuse ‘Try Before You Buy’ programs such as fashion services that ship clothing and accessories for you to explore before sending payment.

With Amazon Fraud Detector, online businesses can assess the risk of customers violating service terms and set limits on the value of goods or services provided so goods aren’t stolen or returned in a condition that violates service terms.

Online payment

Reduce online payment fraud by flagging suspicious online payment transactions before processing payments and fulfilling orders.

With Amazon Fraud Detector, you can setup your checkout flow to evaluate new orders and flag suspicious ones for review prior to processing payments to reduce credit card chargebacks.

Benefits

Build high quality fraud detection ML models faster

Amazon Fraud Detector provides templates that you can use to easily create ML models to identify potential fraud without writing any code.

You simply upload your historical online event data such as transactions or account registrations and select the model template that matches your use case.

From there, Amazon Fraud Detector automatically trains, tests, and deploys a customized fraud detection model fit to your business.

Stop bad actors at the door

Amazon Fraud Detector helps you identify fraudsters right when they create an account by predicting risk in the information they provided so that you can flag suspicious activity before they can do real harm.

That’s because Amazon Fraud Detector uses advanced machine learning techniques that can be applied even to the limited data provided at account creation. Amazon Fraud Detector’s machine learning models can identify up to 80% more potential bad actors than traditional methods.

Built-in online fraud expertise

The pre-built machine learning model templates in Amazon Fraud Detector were developed from 20 years of experience stopping bad actors from attempting to defraud AWS and Amazon.com.

For example, Amazon Fraud Detector uses models similar to those used in the AWS account sign up flow to create different account verification steps for low risk and high risk sign ups.

Give fraud teams more control

By automatically handling the complex tasks required to build, train, tune, and deploy a fraud detection model, Amazon Fraud Detector enables fraud teams to move faster.

After the model is created, they can create, view, and update rules to enable actions based on model predictions without relying on others.