Introducing Identity Check Confidence Score

At Whitepages Pro, we are always looking for innovative ways to help our customers fight fraud and more confidently identify legitimate users. Market feedback is an extremely important driver in our product roadmap, and customers have told us that they want an easier way to integrate and utilize the sheer breadth of data we provide with Identity Check, our 5-in-1 global identity verification solution that returns up to seven match statuses and the data details of the five key consumer attributes of name, address, email, phone and IP.

In a typical scenario, our customers employ automated rules to decide what to do with a transaction. A rule is simply a conditional statement such as “IF billing phone matches billing name, then do X” or “IF IP is a proxy, then do Y”with decisions falling into one of three categories: approve, review, or reject. As our customers grow their businesses, build more sophisticated workflows, and start leveraging more of our Identity Check attributes, it can become extremely tedious to evaluate, code, and maintain a large set of these rules.

We’ve been working hard to address this issue, and today, we are excited to announce the launch of our new Identity Check Confidence Score, which not only solves this problem, but provides several additional benefits.

The Confidence Score
The Confidence Score provides a comprehensive assessment of a transaction by leveraging the millions of transactional patterns across our network and the power of Identity Check’s 70+ data elements in a single, actionable score—made possible by sophisticated data analysis and machine learning models. The core strength of the score is that it reflects real life transaction outcomes based on the feedback information we get from our customers. We also see signals across our network and calculate multiple proprietary inputs such as identity element velocities, transactional frequencies, and linkage histories. This is all done in less than a second—so that our customers can make instant decisions, whether it is approving an eCommerce order, reviewing a loan, creating a new account, or transferring money.

The score ranges from 0 to 500. In general, a higher number score indicates a riskier transaction, while a lower number indicates a good transaction. What this means is that our customers can use the score on both ends of the spectrum. For example, higher scores can be used to reduce chargebacks and fraud losses, while lower scores can be used to quickly identify and clear genuine users—saving manual review costs and allowing businesses to scale rapidly.

Advantages of the Confidence Score

  • Intelligence: Since the Confidence Score utilizes all of Identity Check’s 70+ data elements along with millions of transactional patterns across our network, it is a powerful signal to use for identity verification
  • Efficiency: Whether you use our data for rule building in a platform or in a machine learning model, you can now simply add the score as single attribute and avoid having to evaluate and integrate multiple data points for best results
  • Adaptive: Our machine learning model adapts over time and learns new patterns so you continuously get the best fraud fighting intelligence—without having to make any changes on your end

Accessing the Confidence Score

We are offering our customers three easy ways to access the Identity Check Confidence Score:

  1. Fraud platform: If you are on Accertify or CyberSource, the score is already available to be used for rule building
  2. Identity Check API: The Confidence Score is also available as an additional attribute in our Identity Check API. This is ideal for customers using a proprietary machine learning model or decisioning platform
  3. Identity Check Web: The Confidence Score is available as a beta feature in our web interface to help all the manual review teams using our product

Most importantly, we are not charging anything additional for using the Confidence Score! Click here to learn more about the Identity Check Confidence Score.

Thanks for reading! You might be interested in these posts, too: