Risk Score tells you whether or not the input data on a specific user matches the data we have in our proprietary identity graph—and offers a risk score based on the match (or lack thereof).
Leverage the power of Risk Score and our broad, real-time identity graph, which spans both the digital and terrestrial ecosystems and provides insight into how and if personal and professional identities are linked. Risk Score API can accept a wide variety of inputs such as:
- Name & Address
- Phone
Based on the input provided in the user object, you can understand how what you have provided matches (or does not match) against our identity graph. In addition to the match characteristics, there is an associated risk score, which is a proxy for friction that may need to be applied downstream.
All this combined helps you identify risk on a form field, sign-up, onboarding process, or application.
The direct output of the provided information (at the contact element level) could potentially represent any of the following categories :
| Match Type | Data Type | Description |
|---|---|---|
| Tangled | string | Matches an input belonging to a different individual (no relationship). This is typically an indication of increased fraud risk; further downstream friction may need to be applied specific to that identifier. |
| Household | string | Matches an input field belonging to someone else inside the same household (spouse, young adult, etc). This is typically not an indication of fraud risk, however some friction may need to be applied. |
| Self | string | Matches an input field directly belonging and pointing to the most likely person requested. All outputs with 'self' must share metadata within our graph as this is the anchor in which all other fields are classified. |
| Unknown | string | Not enough information to qualify as any of the above classifications. This may or may not be an indication of fraud risk; further downstream friction may need to be applied. |
| Risk | double | Indicates the level of risk associated with the given inputs. This current risk score is based upon deterministic email related activity such as, first/last seen timestamps, number of sources and observations along with whether or not an observed linkage to a first party cookie exists. If the risk score is closer to 1.0, then the individual and associated inputs are deemed more risky. |