Stripe Radar for Fraud Teams: A First Look

11 Oct 2026

Radar is already watching your Stripe payments. Here is what it scores, what it blocks, and where your team goes next.

Your team opened the Stripe Dashboard and found Radar. Good news. It is built into Stripe and needs no additional integration work. Here is what it does and where everything lives.

What Stripe Radar is

Radar is a suite of fraud-fighting tools built into Stripe. You do not need to do any extra setup work. If you take payments through Stripe, Radar is already evaluating them.

It pairs a rules engine you can change with AI. Together they weigh the risk of each payment. Radar checks transactions, accounts, and customers as they happen. Its AI weighs the fraud risk of each one. If you want more depth on the machine learning side, read about stripe ai fraud detection.

How Radar scores risk

Radar uses its AI models to score each transaction and connected account for risk. Each payment gets a numerical risk score between 0 and 99, where 0 is the lowest risk and 99 is the highest.

Each risk score receives one of these risk levels: normal, elevated, highest, or not assessed. A score of 65 or above indicates elevated risk. A score of 75 or above indicates high risk.

What happens next depends on the level. Stripe reports payments as high risk when it believes they are likely to be fraudulent, and blocks them by default. Radar allows elevated risk payments by default. It can also put card payments with elevated risk into review on its own.

What Radar screens

Radar screens all payment attempt types. That includes successful, declined, blocked, and flagged for review. Your team sees the whole picture, not just the charges that went through.

One exception. Radar does not screen SetupIntents for non-card payment methods. If you save a non-card payment method for later use with a SetupIntent, Radar does not screen it.

The review queue

Radar can place unusual payments in a review queue. The review queue lets you look at odd payments. You do not have to check every payment one by one. That last part matters for a small team. You look at what Radar flagged, not every order.

For a payment in review, you can approve it, refund it, or refund it and mark it as fraudulent.

Where to configure Radar

Everything lives in the Dashboard. You can choose or modify your risk setting on the risk controls page. Risk settings use risk controls. They help you balance sales and fraud on your account. Loosen them and more good customers get through, along with more risk. Tighten them and you may lose sales to false alarms.

You can create custom Radar rules in the Dashboard if your Radar plan supports that feature. You can write custom rules for more control over which payments to review, allow, or block. You can also view and download a filtered list of every payment that a rule has been applied to. That shows you what a rule really did.

Extending Radar

The defaults are a starting point. Its default rules can block a substantial number of fraudulent payments, and custom rules give you more control.

Start with stripe radar rules. Rules let you hold or block based on what you see in your own orders. Then add a stripe radar block list. You can keep lists of trusted customers. Charges from them are allowed or blocked on their own.

For card payments, consider stripe radar 3ds. 3D Secure asks customers to take one more login step before they can finish the purchase. It adds friction, so use it where the risk justifies it.

What Radar does not do

Radar is not a promise of zero fraud. Payments with normal risk can still turn out to be fraudulent. And you are still responsible for the payments you accept. That includes payments later disputed or found to be fake. A fraudulent sale can later turn into a chargeback.

So watch your review queue. Write rules as patterns show up. Treat Radar as your first filter, not your whole fraud plan.