John Ray, Enterprise Fraud & Identity Analytics Leader, presented “Evolution of Identity” at the Equifax Spark 2019 conference. For more information on the event, read our online Spark magazine.
Verifying a person’s identity has changed tremendously over time – and it continues to evolve. Identification of an individual is an age old practice that has been used to quickly understand a person’s social status, family ancestry, societal ranking and uniqueness. Did you know fingerprints were used as a form of identity verification during the Qin and Han Dynasties, which date back from 221 BC to 220 AD? While biometrics have been around for ages, there have been significant shifts in just the last few years.
Extensions of our Identity
In the U.S., we’ve relied heavily on name, address, date of birth and Social Security number to uniquely identify individuals. However, they are no longer enough in today’s digital era. In a world where customer experience drives market choices, consumers want less friction when doing businesses. This has led to the use of email addresses and phone numbers as extensions of our identity. But the problem is that much like name and address, phone numbers and emails can be changed at any point, shared with family and friends, or created with minimal identity proofing.
When it comes to identity verification, these changes trigger risk signals due to common velocity checks within your own data – and within vendor data. This results in compounded risk signals that are mostly false positives. When you see this sharing of touchpoints, it’s critical to find corroborative evidence that the user is in fact not a risky transaction. Things like geolocation, household level analysis and IP classification can be powerful false positive reducers. Navigating which signals are true signals and which may become noise is critical now more than ever.
Social Channels Can Increase Noise – and False Positives
Social media accounts are also used for authentication and verification. However, they can actually increase noise rather than provide clarity. The chart below shows the high number of social media accounts per age group, in contrast to credit hits versus any social media accounts.
Source: Internal Equifax study from March, 2019
The first thing to notice is the average number of social media accounts per age group, represented by the gray bars. For typical working age individuals, the average number of social media accounts can be over 25. Many of these accounts will have conflicting information, stale information, or simply have been abandoned. Despite the successes of Facebook, Twitter and LinkedIn, traditional financial profiles are still just as valuable as ever when it comes to knowing, verifying and finding suspicious patterns in identity assessments. This is evident as consumers hit their late 20’s, as they have a credit profile more often than a social media account.
While there is certainly value in both social media accounts and credit data, listening to the right signals at the right time becomes paramount in identity verification.
Mobile Usage Continues to Climb
Given most industries are using device-related signals, a logical point of reference can be mobile devices. They are used to interact with businesses, as well as verify and authenticate identities. By 2020, more people will have mobile phones than bank accounts, running water or a car. Of these, 44% will be getting a new device as soon as their contract allows. That usually occurs every two years. The challenge is to maintain user experience through device upgrades, downgrades and sharing. Again, signals of risk can easily become noise.
What Does the Future Hold?
As consumers continue to change devices, social media accounts or other extensions of their identity, verification solutions must evolve too. The future is the ability to churn through vast amounts of constantly aging data, while intelligently selecting signals from noise. That means businesses must recognize the liquid nature of consumer identity, or they will fall behind.