Confianza Consumer Fraud & Premium Leakage

The Confianza Fraud and Premium leakage model is a comprehensive prescriptive model that is designed to IDENTIFY & QUANTIFY ALL FRAUD across multiple product lines of insurance. The model will identify and quantify both “hard fraud” as well as “soft or opportunistic fraud”.

Definitions:

· The model identifies “hard fraud” which we define as when an entity (individual, agent, or company) purchases a policy for the purpose of securing a payment for a fictitious or staged incident.

· The model also identifies and quantifies “soft or opportunistic fraud”. This is defined as when an entity (an individual, agent, or company) misrepresents the characteristics of a policy during the application process to avoid premium or secure payment for an exaggerated event.

Eight Factor Summary:

1. Factor 1 - Identity

a. Can we positively identify the applicant?

b. How consistent has the applicant been in how they have represented themselves historically?

2. Factor 2 - Territory

a. Is the primary named insured associated with the address provided?

b. Potential to deceive the insurer regarding their territory / location?

c. Are there a multitude of unassociated identities associated with the address provided?

3. Factor 3 - Exposures

a. Can we identify all of the risks in the form of “exposure risk” associated with the application for insurance?

b. Construct household identifying all residents who have access to the property & vehicles on the policy?

4. Factor 4 - Asset Use

a. Are the assets being insured being used in a fashion that is commensurate with the type of policy?

b. Are the vehicles insured being used for commercial purposes?

c. Are there personal assets being insured by the company?

5. Factor 5 - Asset Ownership / Insurable Interest

a. Are the assets the property of the applicant or are they associated with a third party(leased)?

6. Factor 6 - Rating Variable Integrity

a. Are there discrepancies in the information provided that adversely affects the rating (pricing) of the policy?

b. Detailed address including unit designator and unit number for multiunit properties? Driver’s license numbers of the PNI and other identified exposures?

c. Driver & Vehicle exposure information associated with the primary named insured.

7. Factor 7 - Asset Condition and other Miscellaneous Factors

a. Has the operating condition of the asset been adequately disclosed? Existing damage? Branded title?

8. Factor 8 - Motive

a. Is the consumer or business under financial stress?

b. What is their current financial strength factor?

c. Is there financial condition improving or declining?

Instant access / Seamless IT Integration

· Data delivery & availability via single address or batch processing API’s

About Confianza:

· Driving profitable growth with instant 360 views of the Business, Consumers / Principals, Risk & Exposure.

Confianza’s AI/ML Predictive Loss Analytics and Data Overview:

Business Firmographics:

o NAICS Code & description, Sales, Employees, Year established, Bankruptcy indicators, etc.

Non-FBI Crime data:

o Includes all Incidents and at the near block level, outperforming FBI (based on arrest & at zip code) for Loss Signaling & Propensity.

Non-FRCA Business Owner or Consumer Risk & Behavior:

o Financial strength scores & momentum,

o Credit proxy, Assets, Stability: Own or Rent

o Credit Card Purchase History categories to Validate Operations or Risky Behavior

Property Risk & Exposure:

o Property Replacement Cost (RCE) to solve the undervaluation problem

§ Open, flexible model showing commodity & inflation factors & monitoring.

o Lessor’s Risk Occupancy: Identify all businesses, operations & tenants in building

o Proximital “Risky” businesses identified next to insured address

o Permits: Historical activity to identify building enhancements with related dates

o Flood Data such as: First Floor Height, Flood Risk Score and more

o Home Loss Factors & Indicators: Foundation type, Water heater & electrical panel location

Vehicle VIN, Registration details & Driver Violation History:

o Accurate, registry sourced