调至内容部分
创建账户
或
登录
Stripe 文档徽标
/
询问人工智能
创建账户
登录
开始
付款
销售收入
平台和交易市场
资金管理
开发人员资源
API 和 SDK帮助
概览
关于 Stripe 支付
升级您的集成
支付分析
线上付款
概览查找您的用例Use Managed Payments
使用 Payment Link
Use a prebuilt checkout page
Build a custom integration with Elements
构建应用内集成
支付方式
添加支付方式
管理支付方式
用 Link 更快结账
支付接口
Payment Links
结账
Web Elements
应用内支付
支付场景
处理多种货币
自定义支付流程
灵活收单
编排
线下支付
Terminal
超越支付功能
成立公司
加密货币
智能体商务 (Agentic Commerce)
Financial Connections
Climate
了解欺诈
Radar 欺诈保护
    概览
    Optimize fraud signals
    Radar 会话
    风险评估
    多处理器 Radar 评分
    风险设置
    审核
    列表
    规则
    本地支付方式
    Radar 分析
    Radar for Platforms
管理争议
验证身份
首页付款Radar fraud protection

注意

该页尚未提供此语言版本。我们正在加紧将我们的文档翻译成更多语言,很快将为您提供译文。

Optimize fraud signals

Adhere to our recommendations to maximize Stripe Radar's effectiveness.

Stripe Radar uses AI models that evaluate many signals to distinguish fraudulent and legitimate payments. It computes some signals automatically, but many depend on the data your integration sends. When you provide more relevant data, fraud prevention improves. Try to collect enough information to support accurate risk assessment without overloading your checkout experience.

Integrations and fraud signal completeness

Radar uses data from the Stripe network to detect and block fraudulent transactions across integrations. The integration you choose affects the completeness of the fraud signals you send to Stripe—the more payment data you capture, the better Radar can detect and prevent fraud.

If your integration doesn’t send enough payment data, you can add customer signals (user email, name, and billing address) to the Customer object. You can also add client signals (IP address, user agent, and checkout URL) to the PaymentIntent object.

Integration typeFraud signal completeness
Payment Links Recommended
Excellent
Checkout Recommended
Excellent
Elements with customer signals Recommended
Excellent
Direct API integration with Radar Sessions and customer signals.
Excellent
Direct API integration with client and customer signals.
Very good
Direct API integration with client signals.
Good
Direct API integration with customer signals.
Fair
Direct API integration with no additional signals.
Poor

Important signals to send to Stripe

Include the following information with your payments to improve fraud detection. Our recommended integrations automatically collect this data for you, while direct integrations typically require you to send it explicitly.

Data
Estimated fraud model improvement
Advanced fraud signals36%
IP address12%
Customer email11%
Customer name3%
Billing address1%

Best practices

To ensure that your conversion rate remains high while maximizing the performance of our AI models, adhere to the following best practices.

Collect advanced fraud signals

Stripe Payment Links, Checkout, Elements, and our mobile SDKs automatically collect important high-signal data, such as device information and IP addresses. If you’re not using one of our recommended payment integrations, consider using Radar Sessions to automatically collect advanced fraud signals.

Create payments using the Customer object

Use Customer objects when creating payments to let Stripe track the payment patterns for each one of your customers over time. This increases our ability to identify irregularities in purchasing behavior. To do this:

  • Set up payment methods for future use and add a billing address to Customer objects, using them to create subsequent payments.
  • Provide your customer’s email address when creating a Customer object.
  • Provide your customer’s name when you tokenize their card information.
  • Collect the customer’s shipping address, saving it to their associated Customer object if you ship physical goods.

Each Customer object can also store multiple payment methods, enhancing the checkout experience by letting your customers save multiple cards. Stripe continues to track payment patterns for each customer, regardless of which one they use.

If you’re manually creating a PaymentIntent, make sure to handle declines. If you reuse the PaymentIntent, track repeated attempts to help counter card testing.

Include Stripe.js

Include Stripe.js on every page of your site, not just the checkout page where your customer enters their payment information. This helps Stripe detect anomalous behavior as customers browse and provides additional signals that improve fraud detection.

<script async src="https://js.stripe.com/clover/stripe.js"></script>

Always load Stripe.js directly from https://js.stripe.com/clover/stripe.js. We don’t support using a local copy of Stripe.js, as it can result in user-visible errors, and reduces the effectiveness of our fraud detection.

Update your privacy policy

Radar collects information on anomalous device or user behavior that might be indicative of fraud. Make sure that your own privacy policy tells your customers about this type of collection. If your policy doesn’t include such a disclosure, consider adding the following paragraph:

We use Stripe for payment, analytics, and other business services. Stripe collects identifying information about the devices that connect to its services. Stripe uses this information to operate and improve the services it provides to us, including for fraud detection. You can learn more about Stripe and read its privacy policy at https://stripe.com/privacy.

Enable Radar for future use

Radar operates on a per-charge level, which means that during a PaymentIntent lifecycle, Radar might scan multiple charges if the payment has retries. By default, Radar doesn’t scan if you set up a Payment Method for future use without a charge. If you want to scan SetupIntents, go to your Radar settings and enable Use Radar on payment methods saved for future use.

此页面的内容有帮助吗?
是否
  • 需要帮助?联系支持。
  • 查看我们的更改日志。
  • 有问题?联系销售。
  • LLM? Read llms.txt.
  • Powered by Markdoc