Identity fraud detection step

A step in the Journey Builder is a component used to configure and customize a login, registration, or self-service workflow.

The Identity fraud detection journey step evaluates customer-provided email, phone number, name, and physical address data for fraud risk during onboarding. It allows you to define rule-based risk checks and determine how risky accounts should be handled before they are allowed to proceed.

This step helps identify potentially fraudulent accounts early by analyzing risk signals and attribute associations sourced from third-party data providers.

Identity fraud detection is currently available only in the United States and Canada.

Capabilities

• Evaluate email, phone, name, and physical address data for fraud risk during onboarding.
• Apply rule-based checks to email and physical address signals (for example, email validity, address type, or vacancy status).
• Evaluate attribute associations between customer data points (for example, email-to-phone or address-to-name linkages).
• Assign a risk impact to each rule (pass, fail, or adjust the overall risk score).
• Control rule execution order and cumulative risk scoring behavior.
• Define a minimum risk score threshold that causes the step to fail.
• Require customer consent before performing fraud evaluation.
• Map customer data from local variables, context variables, or native claims.

📘

For conceptual background on fraud signals and attribute associations, see the Identity fraud detection section of the Identity verification policy.

Sample use cases

• Apply stricter scrutiny to high-risk onboarding attempts by routing customers to additional verification steps when fraud indicators are detected.
• Block account creation when strong fraud signals are present (for example, invalid email domains or suspicious physical addresses).
• Use cumulative risk scoring to balance fraud prevention with customer experience, allowing low-risk users to proceed while flagging borderline cases for further evaluation.
• Apply stricter fraud controls for specific journeys or customer segments.

Configuration

To add an Identity fraud detection step to your journey:

1. Open the Journey Builder in the left-hand menu.
2. Create a new journey or select an existing one to edit.
3. Select the + icon and choose Identity fraud detection.
4. Select the pencil icon to configure the step.

The initial configuration screen includes the following options:

1. Step name (optional): An internal label used only within the Journey Builder to help identify the step.
2. Consent requirement: Select the consent that must be accepted before the evaluation can proceed.
   If the customer has accepted the consent → The step continues.
   If no acceptance record is found → The step returns failure and includes metadata describing the reason.
3. Configure step: Select this button to open the detailed configuration dialog.

The configuration is divided into Risk rules and Mapping.

Risk rules

Risk rules determine how customer data is evaluated and how each result impacts the step outcome.

Email and physical address risk rules

Each rule consists of:

• Evaluation rule: The fraud signal being evaluated (for example, email validation or address type).
• Condition: The specific value returned by the evaluation (for example, “Invalid domain” or “Vacant address”).
• Risk impact: The effect on the evaluation outcome.

Available risk impacts:

• Fail evaluation
• Success evaluation
• Increase risk score (+1 to +5)
• Decrease risk score (-1 to -5)

Rules are evaluated in order and can be reordered. Multiple rules may be triggered during a single evaluation.

Attribute match rules

Attribute match rules evaluate associations between customer data points, such as:

• Email address to phone number
• Email address to physical address
• Phone number to full name
• Physical address to phone number

Each rule includes:

• Association type: Defines which attributes are compared.
• Match result: For example, no link, low accuracy linkage, high accuracy linkage.
• Risk impact: Using the same options as above.

Risk score threshold

The risk score represents the cumulative result of all evaluated rules.

• If a rule explicitly fails or passes the evaluation, the step ends immediately.
• If no pass or fail rule is triggered, the final outcome is determined by the configured risk score threshold.

Mapping

The Mapping tab defines where the step reads customer data from. Each field can be mapped using one of the following sources:

• Local variable: A value stored earlier in the journey.
• Context variable: A value available in the journey context.
• Native claim: A claim defined in the identity store and mapped to an account attribute.

Available mappings include:

• Contact-related: Email address, phone number.
• Name-related: Given name, family name.
• Physical address-related: Street address, city, state, postal code.

Outcomes

This step returns the following outcomes:

• success: The configured fraud evaluation rules do not trigger a failure condition, and the final risk score remains below the configured threshold.

• failure: One or more rules trigger a failure condition, or the final risk score meets or exceeds the configured failure threshold, or a required consent was not accepted.

Notes and limitations

• This step is only available in the US and Canada.
• Customer consent must already exist on the account; consent is not collected within this step.
• Rules are evaluated sequentially, and evaluation stops immediately if a rule triggers a pass or fail.
• If no explicit pass or fail rule is triggered, the final outcome is determined by the cumulative risk score.