Accertify

Accertify: Enterprise-Grade Fraud & Digital Risk Management

Accertify is a premier provider of fraud prevention, account protection, and chargeback management solutions. Originally built by in-house fraud analysts and backed by years of operation as a subsidiary of American Express (now operating as an independent entity backed by Accel-KKR), Accertify is trusted by some of the world's largest enterprise e-commerce, travel, and ticketing brands.

Accertify moves beyond simple rules-based engines, leveraging deep machine learning and artificial intelligence to assess risk across the entire customer journey.

When you connect Accertify to our orchestration platform, you gain access to several powerful capabilities:

• Global Community Data: Accertify pools anonymized transactional data from its vast network of global enterprise clients. This allows the system to recognize fraudulent patterns and bad actors across industries before they even hit your store.
• Advanced Machine Learning: It dynamically computes thousands of variables—including device intelligence, behavioral analytics, and connection data—to generate highly accurate, real-time risk scores.
• Comprehensive Account Protection: Beyond just payment fraud, Accertify helps prevent account takeovers (ATO) and fraudulent new account creations by analyzing login behaviors and device spoofing.
• Streamlined Chargeback Management: It offers end-to-end dispute management features, helping merchants automatically aggregate evidence and fight friendly fraud or first-party misuse.

How It Works With Our Orchestration Engine

When you integrate your Accertify credentials into our platform, we automatically ingest Accertify's detailed machine-learning risk scores and map them to our standardized risk tiers (Low, Medium, and High Risk). This allows you to effortlessly plug Accertify's enterprise-grade intelligence directly into your existing dynamic routing strategies—such as triggering 3DS for medium-risk transactions or routing low-risk users to a frictionless checkout.

Risk Tiers

The table below outline how the raw responses and risk scores from each supported provider are mapped to these standardized levels, along with the default routing actions triggered by each tier. Keep in mind that while these represent our standard configurations, you have full control to customize what triggers each risk level to perfectly align with your business's unique risk appetite.

Accertify returns a total score and a recommended action based on their machine learning models.

Requisites

The following information must be at hand before starting to configure the different environments

1. Admin Password

2. Admin Username

3. Environment:

   • test for sanbdox environment
   • live for production environment

Configure Accertify in sandbox

1. Set up testing environment

1. Access the DEUNA Admin.

2. ToggleTest Mode.

3. Configure Cybersource as an anti-fraud provider.

4. Use the testing credentials.

5. Setting test as Environment.

6. Configure a testing payment gateway and establish a Payment Strategy.

   1. Process Medium Risk if manual review will be used for the merchant.

📘

Currently, Accertify does not offer a dashboard to manage the merchant’s account, nor manual review support.

2. Configure device fingerprinting

Accertify has two levels of device fingerprinting:

• A basic script that Deuna has already implemented in Device fingerprint.
• An advanced script named User Behaviour Analytics (UBAs).

User Behavour Analytics is composed by scripts installed in your website pages.

You must collect the results of each script and send them in the purchase request.

📘

For more information, refer to Accertify Developer's Center.

UBAs are composed by the following fields:

1. ubaID. string. The ID used to reference behavior data already captured by Accertify. This value is retrieved from:
   • Browser from the JavaScript function getToken
   • iOS/Android from the SDK?function getUbaId()

• ubaEvents. string. A payload of the most recent fields. This is used to make sure we have up-to-the-second information.
• ubaSessionID. string. The user's Session ID defined by Accertify for the user's web browsing activity.
• pageID. string. The unique page ID defined from the Collector script.
• eventSource. enum. Where or how the event originated. It can be one of the following values:
  • agentPhone
  • agentStore
  • affiliate
  • kiosk
  • mobileApp
  • other
  • web

3. Configure your Accertify account

1. Follow the steps to configure the basic Device Fingerprinting for Accertify.
2. To configure the User Behaviour Analytics, get in touch with Accertify to receive guidance to install and get information from the UBA scritps.

4. Send the information to Deuna

For basic Device Fingerprint: A base64 value is sent in the

• device_id in v1 Create Purchase
• anti_fraud_info.device.id in v2 Create Purchase .

For User Behaviour Analytics:

1. Create a JSON with the UBA fields in camel case. For example:

{
  "ubaID": "accertify-12345-abcde-67890",
  "ubaEvents": "{\"device\":{\"browser\":\"Chrome\",\"os\":\"macOS\",\"screenResolution\":\"1920x1080\"},\"behavior\":{\"clickCount\":12,\"timeOnPage\":127,\"keystrokes\":45}}",
  "ubaSessionID": "uba-session-98765-fghij-43210",
  "pageID": "checkout-payment-page-123",
  "eventSource": "web"
}

2. Encode that JSON in base64

ewogICJ1YmFJRCI6ICJhY2NlcnRpZnktMTIzNDUtYWJjZGUtNjc4OTAiLAogICJ1YmFFdmVudHMiOiAie1wiZGV2aWNlXCI6e1wiYnJvd3NlclwiOlwiQ2hyb21lXCIsXCJvc1wiOlwibWFjT1NcIixcInNjcmVlblJlc29sdXRpb25cIjpcIjE5MjB4MTA4MFwifSxcImJlaGF2aW9yXCI6e1wiY2xpY2tDb3VudFwiOjEyLFwidGltZU9uUGFnZVwiOjEyNyxcImtleXN0cm9rZXNcIjo0NX19IiwKICAidWJhU2Vzc2lvbklEIjogInViYS1zZXNzaW9uLTk4NzY1LWZnaGlqLTQzMjEwIiwKICAicGFnZUlEIjogImNoZWNrb3V0LXBheW1lbnQtcGFnZS0xMjMiLAogICJldmVudFNvdXJjZSI6ICJ3ZWIiCn0=

3. Send the base64 value in:
   • uba_information in v1 Create Purchase
   • order.anti_fraud_info.uba_information in v2 Create Purchase.

Test and deploy to Production

1. Start testing

Make the following transactions:

📘

Save one payment strategy in the DEUA Admin.

Finally, review the transactions in the DEUNA Admin:

In the Fraud section, the following values shouls appear:

• Low Risk (Approved)
• High Risk (Denied)

2. Set up production environment

Follow the previous steps in your production environment.

Change the following values in Cybersource’s Dashboard

1. Untoggle Test Mode
2. Configure Accertify as anti-fraud provider using testing credentials
3. Setting live as Environment
4. Configure the payment gateways and wstablish a Payment Strategy.

📘

Process Medium Risk if you want to use manual review.