What's new: Aggregated traits provide consumer context even when individuals aren’t recognized

When Faraday evaluates a customer list or lead file, our first job is to recognize as many individuals as possible. And we’re very good at it. The Faraday Identity Graph (FIG) consistently delivers excellent match performance, confidently identifying individuals across channels and returning up to thousands of high‑quality appended datapoints per person. In fact, as we outline in our recent blog on the subject, in most consumer‑facing use cases, we surpass industry matching benchmarks, powering precise scoring, personalization, and segmentation.

But no identity resolution system (whether it’s Faraday, Experian, TransUnion, or anyone else) matches every individual with perfect 100% certainty. People use nicknames. They enter secondary homes instead of primary residences. Vendor files sometimes lack specific attributes. And occasionally the provided information simply isn’t complete enough to resolve a person at the individual level.

Historically, unmatched individuals came back with no appended context at all, a common limitation across identity resolution systems.

Aggregated traits change that.

What aggregated traits are

Aggregated traits provide best‑available context for individuals we cannot match at the individual level. Rather than returning nothing, Faraday looks at the surrounding population (using a structured waterfall) to infer reasonable values.

This is not data imputation for known people. We do not overwrite or fill missing attributes for individuals who match successfully in the Identity Graph. We only generate aggregated values for individuals we cannot match.

Aggregated traits exist to answer a simple question:

When a client sends us a person we can’t individually recognize, what’s the most useful information we can provide?

Identity matching at Faraday

Before getting into the specifics of aggregation, it helps to zoom out briefly. Faraday’s matching pipeline evaluates every record using multiple levels of identity evidence: person‑level identifiers, address recognition, and household context. When we confidently match an individual, we return precise appended attributes directly from FIG.

Aggregation only comes into play when that match isn’t possible.

How it works: the aggregation waterfall

When Faraday cannot match an individual, we progressively widen the lens of context. Each step uses only populations that are both (a) in the client‑provided list and (b) known to Faraday.

1. Cohabitant level
If Faraday can identify other individuals at the same address who share the same last name, we aggregate their traits. This is often highly accurate for household‑level attributes (e.g., property characteristics) and reasonably accurate for some person‑level attributes.

2. Address level
If no cohabitants exist, we look for any other recognizable individuals at the same address. We then aggregate their known traits.

3. Postcode level
If the address is new to us, we widen the scope to the postcode. We aggregate traits from recognizable individuals within that postcode who also appear in the client‑provided population.

4. Scope‑level
If no address or postcode matches exist, we return the aggregated value across the entire client‑provided scope population.

A few important notes:

• Aggregation only occurs when we cannot match the person.
• We do not fill in missing attributes for matched individuals (e.g., if a known person’s age is NULL, we return NULL).
• We aggregate across all data types (numerical, categorical, date‑based, etc.) each using appropriate rules.
• For "everybody" scopes built entirely from Faraday data, no aggregation is needed because every individual in that scope is already known in the Identity Graph.

Why aggregated traits matter

Aggregated traits make unmatched individuals actionable. Most clients still need to engage the people they can’t match, and without aggregation, those records would come back with no appended context at all, removing them from scoring, segmentation, routing, and personalization.

Instead of returning a blank row, aggregated traits provide structured, business‑relevant context based on the surrounding population you provided.

This mirrors how sales and marketing teams make decisions every day. When two leads look similar, but one lives in a higher‑value ZIP code or a certain property type, that context shapes how outreach is prioritized. Aggregated traits allow your automated systems to apply that same practical, common‑sense reasoning, even when a perfect individual match isn’t possible.

How this might show up in your industry

Aggregated traits for home services

Call centers and installers deal with messy lead files every day: typos, short forms, secondary addresses. Aggregated traits help teams spot homeowners in high‑value neighborhoods, recognize likely property types, and prioritize leads that fit install eligibility, even when the exact match isn’t possible.

Aggregated traits for retail & e‑commerce

Retailers often need to market to prospects who aren’t yet full customers. Aggregated traits help illuminate neighborhood affluence, household makeup, or lifestyle clues that make direct mail, audience building, and onsite personalization more effective, especially for lightly identified shoppers.

Aggregated traits for retail & e‑commerce credit unions & financial services

Membership and loan pipelines frequently include incomplete or inconsistent records. Aggregated traits help maintain a baseline understanding of risk, intent, and opportunity across the entire applicant list, supporting eligibility checks, prequalification workflows, and smarter outreach.

Across all verticals, the principle is the same: when the alternative is returning nothing, providing the best available context leads to better targeting, sharper prioritization, and more confident downstream decision‑making.

Transparency and control

Aggregated traits are currently enabled by default for client‑provided scope populations. If your team prefers not to use aggregated traits (or wants to discuss configuration options) your Faraday team can work with you directly.

To wrap things up

Aggregated traits provide useful, structured context when we cannot match an individual, all without modifying or imputing data for people we do know.

This feature ensures broader feature completeness, supports downstream scoring and modeling, and helps teams work with the full lead population they care about. If you'd like to learn more, reach out to your account executive!