How IDFusion works

• Customer sourced application data, 
• Internal company data, and 
• External information sources, both public and private

As a prospective customer is completing their application, we take them through four analytical filters:

Entity Resolution – Are you who you say you are?
Linkage Analysis – What are you connected to?
Social Network Analysis – Who are you linked to and how strongly?
Predictive Analytics – What is your network doing?  What are you likely to do?

• Are there any missing family members (teenagers)? 
• Does the customer have multiple residences? 
• Are they claiming the insured property is garaged at the rural residence? 
• Do they have unreported accidents, citations or judgments relating to their driving? 
• Have the vehicles they want to ensure been involved in any questionable events? 
• Do their phone numbers, email addresses and places of employment match up?

Obviously the less truthful a potential customer is during the application process, the more likely they will not be truthful when there is an issue or a claim.

Social Network Filter
But a customer could complete an application with information that checks out perfectly and still be intending fraud. Indeed the more sophisticated fraud efforts typically use “clean” identities. This is where Social Network or “Bad Crowd” analytics come into play. Typically even clean identities have connections – be they email addresses, phone numbers, employment or car repair addresses that are connected to other fraud attempts by other identities. The system analyzes these identity markers and maps them up against other known fraud nodes. The closer the connections between the applicant and various markers of know fraud, the higher the “Bad Crowd” index and the greater probability that this identity intends fraudulent activity.

By applying both the Veracity and Social Network screens in real time during the on line application process VeracityID is able to give its Insurance partners a much clearer picture of the fraud risk associated with a given application. This allows them to make a fact based decision both on whether to underwrite the risk and what price offer to make. By doing so, our Insurer partners avoid undertaking unreasonable risks, allowing their competitors who don’t use VeracityID to have them.

There is power in the data
As VeracityID signs up more Insurer partners in a given segment like personal lines automotive, we will be able to use the blinded data used in our process to provide Insurers with statistically valid models that they can use to assess and predict the likelihood of fraud based upon the overall veracity of the application and the ‘relational connectedness’ of the identity presented. This capability will pay significant dividends to early adopters but should eventually yield industry insights that hopefully will result in substantially less fraud for everyone, improving the insurance value proposition for insurers and customers alike.