Fraud performance overview

This dashboard shows visual reports related to fraud performance.

Fraud Performance Overview dashboard

Fraud Event Ratio widget

Fraud Event Ratio widget

The following table describes the elements of the Fraud Event Ratio widget:

Elements of the Fraud Event Ratio widget
Element Description
Visualization Fraud Event Ratio widget
Description Displays the number of engagements that have a fraud verdict applied by a fraud analyst as a proportion of all engagements. This metric is used to evaluate large changes in fraudster behavior over time. The data refers to a specific scope.
Type Metrics
Filter Engagement creation timestamp (date and time)
KQL language
Linkable Entity None

Fraud Missed widget

Fraud Missed widget

The following table describes the elements of the Fraud Missed widget:

Elements of the Fraud Missed widget
Element Description
Visualization Fraud Missed
Description The number of engagements that don’t generate a fraud alert but have a fraud verdict applied by a fraud analyst. This metric is used to evaluate the amount of confirmed fraud identified using external information, confirmed fraud engagement below the configured fraud engagement threshold, and new confirmed fraud engagements discovered using offline Gatekeeper fraud tools such as clustering. Display two set of metrics:
  • Fraud Missed - Verdict is Fraud but the engagement decision isn’t fraud.
  • Fraud Missed Percentage - The denominator of the fraction is the total number of engagements where decision is Fraud.
The data refers to a specific scope.
Type Two set of metrics
Filter Engagement creation timestamp (date and time)
KQL language
Linkable Entity None

Fraud Workload widget

Fraud Workload widget

The following table describes the elements of the Fraud Workload widget:

Elements of the Fraud Workload widget
Element Description
Visualization Fraud Workload
Description Displays the fraud workload set of metrics:
  • Fraud Alerts Generated - The number of engagements that surpass the fraud risk score threshold.
  • Fraud Alerts Worked - The number of engagements that surpass the fraud risk score threshold, have been reviewed by a fraud analyst, and have a verdict applied to the engagement.
  • Fraud Alerts Worked - The proportion of fraud alerts that have been reviewed by a fraud analyst.
The data refers to a specific scope.
Type Three set of metrics
Filter Engagement creation timestamp (date and time)
KQL language
Linkable Entity None

Real Time False Fraud Alerts widget

Real Time False Fraud Alerts widget

The following table describes the elements of the Real Time False Fraud Alerts widget:

Elements of the Real Time False Fraud Alerts widget
Element Description
Visualization Real Time False Fraud Alerts
Description Displays two set of metrics for the Contact Center:
  • Real Time False Fraud Alerts - The number of engagements that generate a real time fraud alert in the web agent console, have been reviewed by a fraud analyst, and a non-fraud verdict is applied.
  • Real Time False Fraud Alerts Percentage - The proportion of real time false fraud alerts as a function of all biometric validations including successful verifications, verification mismatches, fraudster matches, and fraudster mismatches.
The data refers to a specific scope.
Type Two set of metrics
Filter Engagement creation timestamp (date and time)
KQL language
Implicit Filter The widget selects SESSION_TYPE_CALL_CENTER.
Linkable Entity None

Verdict widget

Verdict widget

The following table describes the elements of the Verdict widget:

Elements of the Verdict widget
Element Description
Visualization Verdict
Description Displays the set of metrics:
  • Confirmed Fraud - The number of engagements that generate a fraud alert, and after review, a fraud analyst applies a Fraud verdict.
  • Confirmed Not Fraud - The number of engagements that generate a fraud alert, and after review, a fraud analyst applies an Authentic, Someone Else, or Uncertain verdict.
  • Fraud Accuracy Rate - The proportion of engagements that generate a fraud alert, and has a fraud verdict applied as a function of all fraud engagements that includes either Fraud, Authentic, Someone Else, or Uncertain.
The data refers to a specific scope.
Type Three set of metrics
Filter Engagement creation timestamp (date and time)
KQL language
Linkable Entity None