If you’re using predictive models, the number one question you’ll eventually ask is: _Are these predictions actually working?_

At Faraday, we love that question. We’re not here to spin our wheels—we want to generate tangible value for each and every client. That ethos is built right into our sales process: before any client signs on the dotted line, we offer to run a custom [holdout](/blog/holdout-testing-evaluate-prediction-accuracy) analysis on their data to prove our models can accurately predict how their customers behave. That’s **transparency** and **trust** from the very beginning.

But after those lines are signed and once those models are live, it can sometimes become harder for clients to get that same level of visibility. Sure we have a variety of other [explainability and transparency](/features/ai-safety-and-responsible-ai) features but here at Faraday, we aren’t really ones for leaving well-enough alone.

Trust in AI isn’t built on promises, it’s built on proof. So that’s why we’ve released **Efficacy Reporting**, a new feature in the Faraday dashboard that gives you yet another layer of insight into how your models are operating.

## What it does

Just like a holdout test, Efficacy Reporting measures the correlation between Faraday’s predictions and what actually happens. Reports are autogenerated for pipelines that have an [Outcome](/docs/abstractions/outcomes) in the payload (which must have eligibility explicitly configured).

> 📌 Note: Efficacy Reports require a live, regularly updated dataset connection. One-off uploads won’t generate meaningful reports, since we need to verify outcomes over time.

Each report breaks down performance by decile—the top 10% of predicted scorers, the next 10%, and so on—so you can quickly see how outcomes line up against model expectations. A strong model will show the classic “lift” shape: higher attainment in the top deciles, tapering off toward the bottom.

[![Model efficacy report in Faraday dashboard](/images/blog/efficacy-reports-in-dashboard.png)](/images/blog/efficacy-reports-in-dashboard.png)

## Where to find it

Efficacy reports live in the [Pipeline tab](https://app.faraday.ai/pipelines) of the Dashboard for any qualifying pipeline. As they’re auto-generated, there’s no extra setup required, and they’re tied directly to the models they evaluate, so you’re always looking at performance in context.

## Who it’s for and how to use it

Efficacy reporting is designed for the teams and leaders who rely on predictive insights to make decisions.

- Marketing and growth teams can use it to build confidence that their campaigns are aligned with reality.
- Data teams can use it to validate that models are holding up over time without pulling ad‑hoc reports.
- Executives can use it to get clear, ongoing evidence that predictive AI is delivering ROI.

In practice, efficacy reports let you validate Faraday models over time without extra manual work, spot potential performance issues early, and build confidence with internal stakeholders that predictions are worth scaling.

## What’s next for transparency at Faraday?

Efficacy reporting is just one step toward making the performance of your predictions as visible and explainable as possible. We’ll keep expanding this framework so you can always see, in concrete terms, how Faraday is driving real outcomes.

Because at the end of the day, predictive AI should be more than a [black box](/blog/the-ai-blackbox-is-a-liability) — it should be something you can measure, trust, and build on.

Efficacy reporting lets you see how well your Faraday models are performing in the real world by comparing predicted outcomes to actual results.

What's new: Efficacy reporting makes modeling even more transparent

If you’ve spent five minutes in the identity resolution world, you’ve probably heard someone throw out a number like _“we match 99% of records, guaranteed\!”_

Sounds great, right? Until you make campaign projections for your CMO that … don’t entirely pan out.

The thing is, flashy numbers might close deals, but do they drive value? Only if they’re accurate.

So sure, Faraday can hit 99% matching in certain conditions (we’ll get to that later). But does that mean it’s the number you should plan your campaign around? Well, that depends: on what identifiers you’re looking for, and what you put in to find them.

[![Jack Black magic number gif - school of rock](/images/blog/magic-number.gif)](/images/blog/magic-number.gif)

## Match rates aren’t one-size-fits-all, and they’re not a magic number

We’ve written before about why [match rates matter](/blog/identity-matching-makes-or-breaks-predictions) (because they’re the foundation of enrichment and predictions) and why [transparency and accuracy matter even more](/blog/is-your-id-match-rate-real) (because vanity numbers break downstream models). But here’s the thing: when it comes to match rates, there’s no universal “magic number.” Anyone selling a consistent 90%+ is probably doing one of three things:

- Counting duplicates (two bad emails for the price of one\!)
- Leaning on easy identifiers like household address and presenting that rate as always true
- Or—a personal favorite—straight-up inflating numbers.

In reality, match rates depend almost entirely on **what identifiers you hand us**. Think of it like cooking: bring fresh ingredients, and you’ll get a great meal. Show up with a half-empty pantry and a can of expired beans, and we can only do so much.

Moreover, it’s not just about whether something matches—it’s about _at what level_ they match. Faraday can resolve identities at the person or household level, and in some cases at a broader level like ZIP code or metro area. Which one matters depends entirely on your use case. Sending direct mail? Household-level matches are usually perfect. Calling or emailing individuals? Then person-level accuracy is what you need.

## So what can you actually expect?

With all of that context in mind, here’s what you can actually expect with Faraday:

- **Email-only datasets** → **\~60%+** individual match rate. Not bad, considering how many emails are throwaways, duplicates, or churned.
- **Add phone numbers** → things stabilize. Old numbers still get recycled, but you’re closer to **\~65%+**.
- **Bring addresses into the mix** → now we’re cooking. Household match rates can climb into the **80–99%**+ range depending on data quality, with **70–80%** at the individual level.

**And here’s an important point**: yes, on a clean dataset with addresses, we can technically match **99% of residences**. But do we advertise a 99% match rate?

No. Because that would be misleading. Of course we can recognize residence level matches on that dataset—we literally have the addresses—but what we’re actually proud of in that example is that we can match **\~80%** of _individuals_ on the same dataset. Is that less flashy? Sure. But is it accurate, actionable, and potentially value driving? Absolutely.

And that means you won’t be backtracking later when projects built on inflated numbers don’t pan out.

## Why this matters for CMOs

If you’re a CMO you already know where this is headed. Inflated numbers might look good in a pitch deck, but they don’t hold up when you’re presenting results to the board. Every padded “match” that isn’t real will:

- Waste spend in acquisition
- Poison your enrichment
- Undermine predictive models you rely on

That’s why Faraday takes the harder road: we dedupe, we verify, and we only count matches that are actually usable in production. For you, that means you can confidently plan campaigns and allocate budget knowing the foundation is solid—no backtracking when the numbers don’t pan out.

## So what’s the punchline?

At the end of the day, match rate isn’t about chasing the biggest number—it’s about trusting the one that will actually power your campaigns. Faraday’s approach gives you the clarity and reliability you need to move forward with confidence.

Stop planning campaigns on vanity numbers. Let's see what _actual_ match rates look like with your data.

Faraday delivers real, verifiable match rates—not inflated vanity numbers—so CMOs can confidently plan campaigns on accurate foundations that actually drive value.

Where's my 99% match rate!? Well, right here, but it’s complicated…

See our complete guide: [Predictive AI for e-commerce and retail brands](/blog/retail-ecommerce-complete-context-guide)

Consumer brands today run on data. From personalization to retention, every marketing move depends on how well you know your audience. But here’s the catch: even with mountains of first-party data, not all of it is actually usable.

Some data is locked behind privacy regulations, some is incomplete, and much of it simply isn’t actionable. Meanwhile, CMOs are under pressure to reduce CAC, prove marketing’s contribution to revenue, and differentiate in markets where competitors often have access to the same third-party sources. You can’t grow on data alone—you need data you can act on.

## Problem

And that was exactly where a leading DTC consumer eyewear brand we work with found themselves. They needed reliable customer intelligence to guide merchandising, lifecycle messaging, and geo activation. But much of their most informative first-party data—like prescriptions and exam info—qualifies as [Protected Health Information (PHI)](https://www.hhs.gov/hipaa/for-professionals/privacy/laws-regulations/index.html) under [HIPAA](https://www.hhs.gov/hipaa/for-professionals/privacy/laws-regulations/index.html).

HIPAA tightly governs how covered entities and their business associates handle PHI. Using PHI to segment, target, or personalize marketing requires explicit patient authorization or de-identification that meets strict standards. In practice, marketing teams are blocked from using this data to understand demographics, build segments, or activate campaigns. Even analytics teams face barriers when joining or sharing it downstream. The result: blind spots that inflate CAC, weaken segmentation, and slow campaign velocity.

This eyewear brand’s CMO and team had major gaps: their customer base appeared to skew much older than reality, household and location context was incomplete, and they lacked confidence to prioritize creatives, offers, and store-adjacent campaigns.

## Solution

But with Faraday, it doesn’t have to be this way. Even when first-party data is incomplete, restricted, or locked behind HIPAA, Faraday makes it actionable.

### Faraday’s data pillars

Faraday provides three complementary types of data that work together as a privacy-safe growth toolkit:

1. **Identity completion** – Filling in the missing fields that make a record usable (e.g., appending verified address, phone, or email to partial identifiers).
2. **Consumer data** – Rich demographic, lifestyle, property, and geo datapoints that unlock segmentation and analytics without relying on PHI.
3. **Custom predictive datapoints** – On-demand models like Likelihood-to-Convert, Best-First-Product, or Credit Score Proxy, built on top of the [Faraday Identity Graph](/features/faraday-identity-graph).

Unlike generic datasets available to any competitor, Faraday’s Identity Graph is proprietary—built from verified, diverse sources and continuously maintained for completeness. Together, these pillars help CMOs complete their records, enrich them with context, and predict what comes next.

### Why consumer data was the unlock

In this eyewear brand’s case, the **consumer data appends** pillar supplied the missing context—without touching PHI. Faraday’s U.S. identity graph linked public and commercially available consumer datapoints to their customer records, enriching them with:

- **Demographics** (age, income proxies, household composition)
- **Lifestyle attributes** (interests, shopping behaviors, property/home tenure)
- **Location context** (precise household-level geo, nearest store)

This enrichment enabled:

- A truer picture of the brand’s age mix and household dynamics.
- Weekly **closest-store** data powering one-to-one geo messaging (email, SMS, push).
- Append-driven cohorts aligned with survey-based attitudinal segments—without needing PHI.
- An identity layer that could be extended across martech partners via Faraday’s [Universal ID](/blog/universal-id).

With these foundations in place, the next step was to operationalize the data.

## How we rolled it out

The rollout followed a clear progression. [Matching](/blog/match-boost-improves-enrichment) is the foundation—without reliable matches, enrichment and predictions fall apart. Deterministic keys like email and name+address ensure that the right consumer datapoints attach to the right individual, building trust in the insights that follow. [Strong matching](/blog/is-your-id-match-rate-real) also reduces wasted spend, prevents duplicate outreach, and gives teams confidence that when a segment is defined, it truly represents the audience they want to reach. To cement that foundation, the team validated coverage by cohort, compared age accuracy against survey and date-of-birth panels, and tuned household logic with clear match-rate expectations.

With this groundwork in place, Faraday’s consumer data began to deliver tangible value. Weekly closest-store feeds powered geo-personalized campaigns, while enriched cohorts aligned cleanly with the brand’s attitudinal segments. These quick wins proved that non-PHI consumer data could fill critical gaps and unlock action right away, giving marketing teams the confidence to activate without regulatory risk.

The natural next step was to extend beyond enrichment into prediction. Faraday’s custom predictive datapoints drove repeat-purchase and product-interest segments, and surfaced growth opportunities through a lite [Market Opportunity Analysis](/docs/abstractions/MOA). By layering prediction on top of trusted enrichment, the brand gained a clear path from raw data to measurable outcomes: age distribution rebalanced to align with survey ground truth, store-adjacent messaging became a high-engagement staple, and campaign testing velocity increased without waiting on PHI approvals.

## Why CMOs choose Faraday consumer data

In a market crowded with data vendors making big promises, Faraday stands apart by focusing on what actually moves the needle for marketing leaders. Many providers chase inflated match rates or push generic attributes that look good in a deck but fail in the field. Faraday’s advantage lies in pairing a privacy-safe foundation with exclusive, activation-ready data tied to a proprietary identity graph.

- **Privacy-safe by design:** Non-PHI append workflow that respects HIPAA.
- **Action-ready:** Datapoints map cleanly to one-to-one channels, geo plays, and predictive models.
- **Proven impact:** Keys, QA, and repeatable playbooks turn enrichment into lower CAC, stronger pipeline contribution, and higher LTV—not dashboards.

## Conclusion

Consumer brands don’t need more raw data—they need the right data they can actually use. Faraday delivers exactly that: privacy-safe enrichment, predictive insight, and proven rollout. If you’re ready to turn non-useful records into measurable growth, [we should talk](/contact).

A leading DTC eyewear brand turned restricted and incomplete first-party data into growth by using Faraday’s consumer data to enrich records, enable geo-personalized campaigns, and drive measurable results.

DTC eyeglasses brand unlocks growth with Faraday's consumer data

**See our complete guide:** [Predictive AI for agencies](/blog/faraday-for-agencies).

---

Marketing agencies today are in a bind.

Their brand clients expect clearer data, sharper predictions, and measurable ROI—but most agencies are working with the same third-party datasets as their competitors. That makes it hard to differentiate, hard to prove ROI, and harder still to retain high-value accounts.

Agencies that can’t surface unique, predictive insights risk looking like commodities in an industry where brands are paying for strategic advantage, not generic analytics.

## The problem

One performance marketing agency faced this challenge firsthand. Their brand client’s repeat purchase rates had plateaued, and the team realized that their audience assumptions were largely untested.

They believed their “younger” online buyers were driving growth—but without deeper enrichment, they lacked visibility into who these customers actually were. Their internal dashboards showed the numbers, but not the people behind them.

## The solution

Using **Faraday’s demographic enrichment**, the agency reshaped how their client understood its core audience. The enriched data revealed that what the brand assumed were “younger” online buyers were actually **five to ten years older** than expected.

Armed with this new demographic clarity, the agency refined welcome flows, personalized product recommendations, and built messaging that reflected the audience’s true profile and motivations.

## The outcome

Within just five months, the client saw a **17% lift in repeat purchases** compared to the same period the year before. Beyond the lift itself, these refinements strengthened customer lifetime value and gave the marketing team newfound confidence in their audience strategy.

> “We always had these customer personas in our minds,” said the client’s marketing lead. “It was interesting to see how what we thought were younger online customers weren’t as young as we thought.”

By helping the agency uncover and act on these insights, Faraday enabled them to elevate from an execution partner to a **strategic growth driver**—delivering proprietary, data-backed understanding their competitors couldn’t match.

## What does this mean to you?

For agency CMOs, this is what defensible differentiation looks like: unique, predictive enrichment that drives measurable performance improvements and scales across clients without adding complexity.

With Faraday, agencies can move beyond generic reporting and offer clients proprietary insights that directly impact growth and retention.

## Ready to differentiate your agency?

Faraday helps agencies move beyond generic data to deliver proprietary predictive insights that win and retain high-value clients. Let’s talk about how we can power your dashboards with enrichment and predictions that set you apart.

Faraday enabled a performance marketing agency to boost repeat purchases by 17% by enriching their client’s customer data and revealing new insights that reshaped segmentation and messaging.

Performance marketing agency drives 17% lift in repeat purchases with Faraday’s demographic enrichment

**See our complete guide:** [Predictive AI for banks and credit unions](/blog/banks-credit-unions-complete-context-guide).

---

Credit unions exist to serve their members and strengthen their communities. That mission-driven focus is what sets them apart—but it also means they don’t always have access to the same marketing tools, resources, or budgets as megabanks and fintechs. The challenge today is clear: how can credit unions compete against larger institutions armed with endless resources and cutting-edge technology?

If you’re leading marketing or member experience at a credit union, you know this pressure firsthand. You’re competing with organizations that have bigger budgets, flashier tech, and armies of data scientists. Meanwhile, your team is lean, your resources are stretched, and the board wants clear proof that every marketing dollar is driving measurable growth.

The issue is not that you don’t have data — you’ve got mountains of member information. The challenge is turning it into something actionable. And too often, your campaigns default to generic blasts: everyone gets the same credit card offer, the same auto loan promo, the same mortgage refi pitch. Members tune out, conversions lag, and you’re left struggling to show ROI.

The truth is simple: **without predictive product recommendations, you’re leaving millions on the table every quarter.**

## The problem: your members don’t take your offers

If you’ve gotten this far into the blog, you probably already know some of this pain:

- **Spray-and-pray campaigns don’t resonate.** When every member gets the same offer, most of them see it as noise.
- **Lean teams can’t build predictive models in-house.** You don’t have the bandwidth to mine your data and run experiments the way megabanks do.
- **Boards expect measurable ROI.** It’s hard to prove marketing’s value when campaigns feel like guesswork.

This all results in missed opportunities to deepen relationships and grow wallet share. Every generic email is a missed chance to connect with the right member, at the right time, with the right product. That means money is left on the table and members don’t get the services and products they actually need and want.

## The solution: predictive datapoints to show customer intent

While this is a complicated problem, Faraday makes the answer simple.

By combining your first-party member data with the [Faraday Identity Graph](/features/faraday-identity-graph), which features 1,400+ consumer datapoints on 240M U.S. adults and their households, we can generate custom predictive signals like: likelihood-to-open-auto-loan, likelihood-to-refinance-mortgage, or likelihood-to-apply-for-credit-card.

Most importantly, we can generate a predictive datapoint [that ranks which of your products each member is most likely to choose](/templates/next-best-offer). That way, instead of blasting every member with the same offer, you’ll know exactly which product resonates. Imagine seeing: Jan is 90% likely to open an auto loan, but only 40% likely to refinance her mortgage. You’d probably send her the offer for the auto loan over the mortgage refi — and as a result, she’d be much more likely to convert.

Our data solves various core challenges:

- **More relevance, less noise.** No one likes ads that don’t resonate, this way members see offers that actually matter to them.
- **Smarter use of lean teams.** You don’t need an in-house data science department to compete with megabanks, Faraday adds a predictive data layer into your existing tech stack.
- **Clearer ROI.** Campaigns become more efficient and easier to measure, making it simple to show the board real growth impact.
- **Transparent and compliant.** Every model is also fully transparent and auditable, so you can stay aligned with fair lending requirements.

Once the recommenders are built, delivery is seamless. Predictive datapoints flow directly into your CRM or marketing automation platform via API — no heavy IT project required. And because every model is fully transparent, you can align campaigns with clarity and demonstrate to the board how marketing spend translates into measurable member growth.

## How this looks in practice

Take [Advia Credit Union as an example](/blog/advia-auto-loan-growth). Like many credit unions, Advia wanted to deepen relationships with existing members and grow wallet share, but generic campaigns weren’t moving the needle. After linking their first-party member data with the Faraday Identity Graph, they used predictive product recommendations to identify which members were most likely to open an auto loan. That meant they could target offers only to the people who were truly interested — no wasted touches, no guesswork.

The impact was immediate. Within 30 days, Advia had already generated \$1.1 million in new auto loans. After 90 days, the campaign had delivered \$2.7 million in auto loans with an application rate more than four times higher than their baseline. And because every model was transparent and bias-checked, Advia was able to stay aligned with fair lending requirements while still giving members the personalized offers they actually wanted.

## Ready to stop guessing and start predicting?

If you’re tired of generic campaigns that don’t move the needle \- and tired of the board breathing down your neck for ROI \- predictive datapoints from Faraday can make growth measurable, compliant, and easy.

Faraday is built for teams exactly like yours. Ready to see how predictive data can unlock your next $2.7M? [Reach out.](/contact)

In a competitive market, credit unions need to make sure to maximize the value of every account holder. Without predictive product recommendations, they’re leaving millions on the table every quarter.

Tired of credit union members not taking your offers? Fix it with predictive data

When marketers start looking into lead scoring and predictive data, they often hear about "real-time" delivery and get pretty excited. And we get it. It has that instant-gratification appeal: the idea that you can get a prediction in the blink of an eye and make a decision on the spot. But here’s the catch—real-time isn’t always the smartest path. In fact, for most businesses and use cases, **batch predictions actually deliver richer, more useful results.**

At Faraday, we offer both methods—batch and real-time—because they serve different needs. But understanding when to lean on one versus the other can save you money, improve campaign efficiency, and make sure you’re not chasing speed at the expense of insight.

So when is "real-time" actually what you're looking for? Let's find out 👇

## Why batch predictions are the default

After that setup, you might be not surprised to learn that real-time actually isn’t the default way Faraday delivers data. 😱

That’s because batch processing is the real workhorse behind most predictive data and modeling. Think of batch like a doctor running a full health checkup versus just taking a pulse. Sure, the pulse might tell you something critical in a hurry — like Charlie isn’t dead — but it doesn’t actually reveal much else about what’s going on. For the doctor, both the pulse and the full checkup are useful, but they don’t serve the same purpose and can’t replace each other.

Like the full checkup, Batch doesn’t have the immediacy of the pulse check. But what it gives you instead is significantly more nuance and detail.

At Faraday, batch is our default delivery method because most of the time, you need the nuance and the deep context more than the immediacy. And beyond just getting more info about an individual, batch allows you to score an entire audience (your CRM, acquisition list, or customer database) all at once.

Batch is our default because it gives you the _complete picture_ instead of just a quick, decisive signal. Here’s what that really means in practice:

- Score entire populations you already know (like your CRM or customer list)
- Incorporate first-party and third-party attributes into your predictions
- Add multiple datapoints (including predictive datapoints) at once, rather than one-off signals
- Deliver results in flexible formats—CSV, Snowflake, S3—that plug into your systems
- Run on a custom schedule (daily, weekly, monthly) so the data is always refreshed at a rate that matches your use case and needs
- Enrich identity data more deeply, returning additional emails, phones, and addresses

Together this means that you’re not just scoring people, you’re understanding them in context (and scoring them even more). Want to prioritize segments of your CRM contacts? Assign reps to the right accounts? Qualify appointments with [Credit Score Proxy](/blog/prioritize-with-credit-score-proxy)? Batch makes all of that possible, while giving you flexibility and scale.

> **TLDR**: Batch gives you depth, flexibility, and a much more complete picture of your customers and prospects. That’s why it’s the default choice for most of our clients.

## When real-time makes sense

So yeah, batch is pretty great.

But there are moments when speed really does matter. Imagine a call center that has seconds to decide whether to accept or reject an incoming lead instantly in a [ping/post](/blog/so-what-is-a-ping-post-anyway) environment. Or a sales rep standing on someone’s porch with a tablet, trying to decide whether to make the pitch. In those scenarios, waiting for batch processing isn’t an option, or necessary. You don’t need the deep context, like the pulse, you just need a clear signal in a hurry.

That’s where [Faraday’s Lookup API](https://faraday.ai/docs/features/lookup-api) comes in. By pulling from our 3rd party data and pre-scored national populations, we can give you an accurate prediction on the fly. It’s perfect for situations like:

- [Bidding in ping/post systems](https://faraday.ai/blog/what-happens-when-predictive-AI-meets-real-time-lead-filtering) like with [ActiveProspect's](https://activeprospect.com/) [LeadConduit](https://activeprospect.com/leadconduit/)
- [Scoring addresses during door-to-door sales](https://faraday.ai/blog/salesrabbit) with tools like [SalesRabbit](https://salesrabbit.com/)
- Assessing likelihood to sign up on a landing page in real time

**The key tradeoffs:**

- Data must be sent via POST request to a specific API endpoint
- Egress path is narrow: you hit us, and you get the datapoint back — no [Snowflake](https://www.snowflake.com/en/), [S3](https://aws.amazon.com/s3/), or warehouse delivery
- Requires careful configuration of scope populations in advance
- You can only get one datapoint at once — giving you a less complete picture of your target

> **TLDR**: Use the Lookup API when you need an instant signal about someone you just met — like a new lead, form fill, or sales visit. It’s fast, simple, and scoped — but limited to national data only.

## How theses look in the ✨real world✨

Both delivery methods generate real value for Faraday customers—but in very different ways.

Take real-time first: when a regional home-renovation company saw conversion rates slipping by \~21% during peak call‑center traffic, Faraday’s instant lead prioritization turned things around. By feeding a custom “likelihood‑to‑convert” datapoint straight into their CRM, reps focused on high‑quality leads right as they arrived. The impact? [A 13% lift in conversion during high‑traffic periods and a 22x monthly ROI](/stories/call-center).

Now contrast that with batch: [Bee’s Wrap](/stories/bees-wrap), a fast‑growing DTC brand, used Faraday to connect its [Shopify](https://www.shopify.com/) order data to the Faraday Identity Graph and run persona modeling at scale. The batch analysis surfaced detailed personas and geographic propensities that became the backbone of their Target retail pitch.

The outcome? Placement in 550+ stores and an initial six‑figure wholesale order.

Together, these examples show the complementary strengths: batch delivers the depth and context needed for strategic moves, while real‑time provides the immediacy required to win in the moment.

## So which is better?

Well here’s the thing, batch and real-time aren’t competitors — they’re complements. Two specialized tools to keep in your belt for different situations. Batch is about depth, flexibility, and context. Real-time is about speed and decisive action. Used together, they give you both the strategic insights to plan campaigns and the tactical edge to win in the moment.

At Faraday, we deliver both. Most of our clients rely on batch as their foundation, completing identities, enriching consumer data, and providing rich predictive datapoints for entire audiences on a schedule that keeps campaigns humming. Then, when a moment calls for immediacy, our Lookup API steps in to provide the instant signal. The combination ensures you’re never forced to choose between nuance and speed; you get both, working in concert.

Want to learn more about how these can work for you? [Come say hi](/talk-to-sales).

Faraday offers both batch and real-time predictions, helping brands balance the depth and context of large-scale scoring with the speed and decisiveness of instant lead insights to maximize ROI.

Real-time predictions vs. batch data enrichment: what you need to know

If you’ve ever sat through a pitch about data or identity resolution, you’ve probably heard at least one big shiny number: _“We can match 80% of your records\!”_

But here’s the catch: an 80% match rate is meaningless if those matches aren’t real people you can actually contact or use for modeling. False positives don’t just pad a report — they waste budget, erode trust with your customers, and poison downstream predictions and outreach. Worst of all, they mean you can’t plan effectively.

At Faraday, we’ve spent over 13 years refining a simple philosophy: every match should be a record you can confidently act on. And that philosophy isn’t just talk — it’s baked into how our system works day in and day out.

## Faraday delivers matches you can trust

Faraday’s identity resolution system is built to prioritize confidence over vanity stats that sound good on paper but fall apart in real campaigns. This results in a system designed not just to maximize match rates, but to ensure that when we say _“this is the person,”_ you can act with confidence.

Here are some steps we take to ensure this outcome:

**Deduplication and householding**  
We don’t just stitch data together — we actively dedupe records and recognize when multiple identifiers belong to the same household. That means one clean, accurate profile instead of duplicates inflating your numbers.

**Handling life changes**  
People move, change names, swap emails. Most systems struggle to keep up. But the [Faraday Identity Graph](https://faraday.ai/features/faraday-identity-graph) (FIG) is built on longitudinal data, so we can track those shifts and keep matches accurate over time.

**Built-in scale and enrichment**  
Every match taps into our FIG, which contains 1,400+ datapoints on 240M U.S. adults and their households. That means when we confirm a match, you’re not just identifying someone — you’re enriching them with trustworthy, actionable context for campaigns and predictions.

**Multiple paths to identity**  
Matches aren’t only about exact email or address. We account for nicknames, partial addresses, and alternative identifiers — always with strict rules to avoid false positives. And thanks to our newest update, we’ve added yet another path to identity. 👇

## Our latest update

As one of our incredible Data Engineers, Tom Caruso put it in our internal Slack announcement for this update:

_“This new update is cool because it uses the kind of context you can only get from dealing with national populations for years. The key insight we observed is that, beyond the traditional means by which matches are generated — like name \+ address or email \+ phone — many people in the U.S. have names that are actually unique within their geographic area. For instance, while there might be a dozen John Smiths in Los Angeles, there’s only one Tom Caruso in ZIP code 05405, and only one Ben Rose in all of Vermont.”_

By compiling these unique name \+ location pairs from the Faraday Identity Graph, we added a new matching source that kicks in when those other identifiers aren’t available. That means we can now match records confidently even when key fields like email, phone, or full address are missing.

This particularly benefits clients working with partial contact data — like name \+ email \+ state only — or with sparsely populated address fields, which are usually Faraday’s strongest component in identity resolution. By recognizing these unique combinations, Faraday can salvage records that would otherwise be lost, adding millions of verified people to campaigns. And because we only apply this method when a name-location pair is truly unique, the update strengthens our core promise: higher match rates without sacrificing confidence.

This is the critical distinction. Other vendors inflate match rates by accepting ambiguity — counting multiple possible matches as wins. Faraday’s approach adds coverage only when we’re certain it’s the right person.

**TLDR:** Faraday delivers identity matching you can trust. No noise, no padded numbers — just more real people you can actually market to.

## The proof in the numbers

Early testing of this update had already delivered major gains in match rates for multiple clients.

One DTC optics brand immediately saw an additional \+1.5 million matches on an existing dataset and another client, a performance marketing company, gained \+1 million immediately.

Every single match of these millions translates into fewer wasted dollars, more reliable models, and campaigns they could actually plan against with confidence.

For you, that’s the difference between “numbers on a slide” and ROI you can feel in the field and in the boardroom.

## Why marketing leaders need better matches

Identity resolution isn’t just about hitting a high match rate — it’s about whether you can trust those matches to guide your spend and strategy. That’s what sets Faraday apart.

- **Confidence at the core:** Every match is one you can confidently act on. We don’t inflate numbers with guesses or maybes.
- **Proven over time:** With 13+ years of experience and longitudinal data, we capture life changes and long-term patterns that competitors miss.
- **Built-in scale:** Our Identity Graph spans 240M U.S. adults and households, enriched with 1,400+ datapoints and continuously updated.
- **Innovation in motion:** Updates like unique name \+ location matching show that even after more than a decade, we’re still pushing for incremental lifts that flow straight to ROI.
- **Better outcomes downstream:** Trustworthy matches mean stronger predictions, more efficient campaigns, and marketing dollars that work harder.

## We’re always improving

This update is just the latest example of how Faraday treats identity resolution as a living system, not a one-and-done feature. Because of our longevity and ongoing innovation, our clients can trust their matches to be accurate today — and better tomorrow.

Identity resolution isn’t glamorous. But every percentage point of improvement falls directly to the ROI bottom line. And that’s what makes it worth obsessing over.

Stop guessing and start planning with confidence. [Get in touch with our team](/talk-to-sales).

Faraday delivers identity matches you can actually trust—clean, accurate, and enriched with real consumer context—so every record drives smarter predictions, campaigns, and ROI.

Is your identity match rate driving real value or just vanity?

**See our complete guide:** [Predictive AI for home services brands](/blog/home-services-complete-context-guide).

---

If you’re running marketing at a home services company, the pressures never let up. Your CFO is demanding lower CAC. Your sales team is tired of chasing unqualified appointments. Your installers are sitting idle after no-shows. And your call center is burning time on reschedules.

These pressures add up fast — in fact, for one national home services brand, they were burning more than $100,000 per month on junk leads. But by turning to Faraday, they cut that waste and freed their reps from chasing 13,000 bad prospects. Here’s how.

## The problem: you’re filling calendars with jobs that never happen

In home services, volume often wins the day. Marketing budgets flow into purchased leads, affiliates, and digital campaigns to keep the funnel full. But here’s the catch: industry research shows that [only about 12% of leads convert to customers](https://blog.gorizen.com/why-most-window-leads-dont-turn-into-sales).

- **Financing failures:** Many leads won’t qualify for credit with your financing partners, so your reps spend time setting appointments that never close.
- **No-shows:** Some prospects book, but when your crews arrive, no one’s home — burning time, fuel, and morale.
- **Last-minute cancellations:** Others cancel just before the appointment, wasting truck rolls and pushing up CAC and leaving installer schedules in chaos.

Every one of those unqualified or unreliable leads is money burned — in the form of call center hours, truck rolls, and lost opportunities.

## The solution: predictive lead suppression and qualification with Faraday

Faraday isn’t just a theory \- it’s a data platform built to plug into the exact systems you already use to buy and manage leads. That means we don’t just tell you what you _could_ do differently; we deliver the exact datapoints you need into your workflow (in batch or real time) so you stop wasting budget immediately.

But there are a range of ways Faraday’s data can become tangible ROI for your company. Here are just a few:

### Stage 1: Don’t buy the wrong leads

Most home services brands buy leads via [ping-post systems](https://faraday.ai/blog/so-what-is-a-ping-post-anyway). Here, you only get a partial “ping” (basic info like ZIP and homeowner status) before deciding whether to bid. Using [Faraday’s real-time Lookup API](https://faraday.ai/docs/features/lookup-api), you can request a custom predictive datapoint — like **Likelihood-to-Convert** — for each ping.

Through our integration with [ActiveProspect’s](https://faraday.ai/blog/dtc-homeservices-uses-credit-score-proxy) [LeadConduit](https://activeprospect.com/leadconduit/), low-scoring pings are automatically rejected before purchase \- so no money or time are wasted on them.

👉 Outcome: you never pay for the junk leads in the first place, saving thousands per week in wasted budget.

### Stage 2: Qualify appointments with financial signals

Even after rejecting bad leads, many home services CMOs face another pain: reps booking time with customers who won’t pass credit. Faraday helps here too by delivering [Credit Score Proxy](https://faraday.ai/blog/dtc-homeservices-uses-credit-score-proxy), one of 1,400+ prebuilt consumer datapoints from the [Faraday Identity Graph (FIG)](https://faraday.ai/features/faraday-identity-graph). Credit Score Proxy ranks U.S. households by purchasing power and conversion potential, allowing you to deprioritize or flag likely credit fails before time is put into a meeting, only to disqualify them.

👉 Outcome: fewer wasted appointments, less frustration in your sales org, and more predictable CAC.

### Stage 3: Enrich and prioritize at scale

Beyond suppression and credit qualification, Faraday’s platform can continually enrich your lead and customer records with new datapoints — both prebuilt consumer data and custom predictive models trained on your outcomes. That means your call center can prioritize its daily queue around the most winnable prospects, and your marketing team can target high-propensity households for direct mail or digital campaigns.

👉 Outcome: a cleaner funnel from start to finish, with fewer bad appointments and more closed deals.

## The business impact

When CMOs and marketing teams use Faraday to suppress and qualify leads, the results show up fast across the funnel. Here’s how:

- **Lower CAC at the top of the funnel:** Real-time suppression through the Lookup API and LeadConduit integration means you stop paying for the bulk of leads that won’t convert. That’s not abstract savings — it’s marketing dollars preserved before your reps even touch the phone.
- **Higher efficiency in the middle of the funnel:** Credit Score Proxy flags households unlikely to pass financing. The result is fewer wasted appointments, fewer dead-end truck rolls, and a sales team that spends more time closing deals.
- **Smarter outreach at the bottom of the funnel:** Enrichment with predictive and consumer datapoints powers cleaner call center queues, more targeted direct mail, and more effective digital audiences. Every touchpoint aligns with real conversion potential, not just guesses.

This isn't theoretical AI. [One national home services brand put this process into practice.](https://faraday.ai/blog/dtc-homeservices-uses-credit-score-proxy) By scoring every ping in real time with Faraday’s Lookup API and automatically rejecting low-quality leads in LeadConduit, they cut more than 13,000 junk leads in just five months — saving over $500,000 in wasted spend in just the first 5 months. Building on that success, they layered in Credit Score Proxy at the appointment stage, which allowed them to flag low-likelihood households before scheduling in-home visits. The results were clear: fewer wasted appointments and a sales team free to focus on winnable prospects.

As their Sr. Director of Marketing put it:

“We now prioritize calls based on Faraday scores, kick back low-quality leads, and save thousands of dollars every week. A high Faraday score goes to the front of the line. It’s been a complete shift.”

## Why this matters now

For home services CMOs and VPs of Marketing, the mandate is clear: deliver more installs with less waste. Faraday helps you do exactly that by combining consumer enrichment and predictive scoring into a seamless data platform. From rejecting junk leads before purchase, to filtering out low-credit households before appointments, to prioritizing outreach across the funnel, Faraday ensures every marketing dollar works harder. That’s why leading brands now treat Faraday as the centerpiece of their go-to-market strategy — and why your call center doesn’t have to burn money on bad leads anymore.

If you’re tired of paying for leads your crews never install, let’s talk. Or see how Faraday plugs directly into your lead buying and call center workflows on our [home services landing page](https://faraday.ai/landing/homeservices).

Faraday helps home services brands cut wasted spend by rejecting junk leads, qualifying appointments with financial signals, and prioritizing call center queues using predictive data.

Home services call centers are burning money without lead scoring

Imagine this: You’re a brand with a list of leads, but no real way to tell which ones are actually in a financial position to become high-value customers. So you treat everyone the same. That means wasted spend, low conversion rates, and missed opportunities when your best reps chase the wrong people—or when a strong prospect drops out late in the funnel due to credit issues.

Now imagine you could suppress just the bottom 10%—the folks least likely to convert, based on financial behavior and purchasing power—without pulling a single credit score.

That’s where **Credit Score Proxy** comes in.

## So what is Credit Score Proxy?

Credit Score Proxy is a relative marketing value ranking. It segments 120 million U.S. households into 12 rank-ordered categories (A1 to E3), using behavioral and lifestyle data—like age, property characteristics, aggregate credit indicators, and more (we also offer a variant of this datapoint in which the 12 categories are collapsed to 5 more-manageable categories).

It can function as a predictive proxy for financial reliability, offering many of the same benefits as a **credit score** for marketers, without using actual sensitive credit data or requiring a credit pull.

This table shows how Credit Score Proxy tiers typically align with traditional credit score ranges (in the risk score column) but please note: these are illustrative estimates, not actual credit scores. Credit Score Proxy doesn’t use credit data or generate credit scores; these risk scores are included to help marketers understand how each tier relates to familiar financial benchmarks.

| Credit Score Proxy Rank | Illustrative Credit Score Range | % of Households in FIG\* | Marketing Profitability |
| ----------------------- | ------------------------------- | ------------------------ | ----------------------- |
| 1.0                     | 760                             | 17%                      | **Best**                |
| 2.0                     | 740                             | 10%                      | **Best**                |
| 3.0                     | 725                             | 10%                      | Above average           |
| 4.0                     | 710                             | 11%                      | Above average           |
| 5.0                     | 690                             | 11%                      | Average                 |
| 6.0                     | 675                             | 11%                      | Average                 |
| 7.0                     | 660                             | 10%                      | Below average           |
| 8.0                     | 650                             | 4%                       | Below average           |
| 9.0                     | 640                             | 4%                       | Below average           |
| 10.0                    | 625                             | 4%                       | **Poor**                |
| 11.0                    | 610                             | 4%                       | **Poor**                |
| 12.0                    | 590                             | 4%                       | **Poor**                |

**\* Note on distribution:** While the above table reflects approximate national distributions represented in FIG, the Faraday Identity Graph (FIG) shows a slight overrepresentation of higher-credit-score households and an underrepresentation of low-credit or underbanked individuals. This means your enriched dataset may lean more toward high-value prospects—helpful when prioritizing outreach, but worth keeping in mind for broad reach strategies.

So how should you think about using Credit Score Proxy in practice?

## What Credit Score Proxy isn’t

It’s easy to think of Credit Score Proxy as essentially a predictive credit proxy—and that’s not entirely wrong. But it’s also not quite right.

Unlike FICO or other regulated credit scores, Credit Score Proxy wasn’t built for lenders or compliance-heavy decisions. It’s designed for marketers who want to prioritize outreach based on a household’s likely value—not just financial risk.

It uses lifestyle, demographic, and inferred financial behavior to segment households by marketing profitability. No tradeline data, no credit bureau pulls—just smart signals modeled for outreach.

In short: Credit Score Proxy is inspired by credit scores, but optimized for marketing. It’s a broader, privacy-safe alternative that aligns with overall consumer value, not just credit risk.

## How Credit Score Proxy fits into Faraday

Faraday is a modern data platform that helps consumer brands grow by enriching their first-party records with **verified identity data**, curated **consumer datapoints**, and **custom AI-powered predictions**. At the heart of this is the [Faraday Identity Graph (FIG)](https://faraday.ai/features/faraday-identity-graph), our rich database that contains 1400+ unique datapoints on over 240 million U.S. adults and their households. This data is prediction ready, privacy safe, and sourced from some of the best providers in the industry.

We use the FIG to enrich your customer data by appending new information, giving you a deeper understanding of your audience and how best to engage with them. These enriched datapoints generally take three forms:

- **Verified identity data** — Fill in the blanks: name, email, phone, address. These verified identities form the foundation of every enriched record.  
  **Consumer datapoints** — Curated, industry-relevant traits like income, homeownership, household size, or age—plus prebuilt predictive scores such as likelihood to buy luxury goods or subscribe to streaming services. These datapoints make segmentation, targeting, and prioritization easy—even before layering on custom predictions.
- **Custom predictions** — AI-powered scores trained on your historical outcomes, like Likelihood to Buy, Best Product Recommendation, or Persona assignment. Built specifically for your brand, these models use your first-party data alongside the 1,400+ behavioral datapoints in the FIG. Unlike off-the-shelf predictive insights, they’re tailored to your goals and go-to-market motion.

**Credit Score Proxy** is a textbook example of a prebuilt predictive **consumer datapoint**, and like any other datapoint from the FIG, it’s added as a new column to your dataset—ready to drive smarter targeting, prioritization, and ROI.

With Credit Score Proxy in place, you can:

- Build audiences by selecting only the highest-value tiers
- Filter out low-likelihood individuals for cost-efficient targeting
- Route leads or assign reps based on expected value

It’s one of the fastest ways to improve performance without adding complexity—and without pulling credit.

## Credit Score Proxy in action

For example, here’s how [a leading direct-to-consumer home services brand used Credit Score Proxy to sharpen its acquisition funnel.](/blog/dtc-homeservices-uses-credit-score-proxy)

- **At the lead purchase stage**, they paired Credit Score Proxy with a custom predictive score to reject low-quality leads in real time—saving over **$500,000 in wasted spend** within five months.
- **At the appointment stage**, they used Credit Score Proxy tiers to flag households likely to fail credit checks, which were failing about **75% of the time** in lower tiers. By suppressing these leads, they freed up their reps to focus on high-value prospects.

The result: fewer wasted appointments, lower acquisition costs, and a more efficient sales funnel overall.

## Why it matters to you

Many of our clients need a way to infer financial standing without crossing regulatory lines. Credit Score Proxy gives you a privacy-friendly path to do just that—especially useful in industries like:

- Home services
- Solar and energy
- Education
- Financial products (without lending)

It’s one of the most effective ways to suppress low-value outreach and drive higher ROI on every campaign.

Want to learn how Credit Score Proxy works inside your workflow? [Let’s talk.](https://faraday.ai/contact)

Credit Score Proxy helps brands prioritize high-value prospects by ranking U.S. households into predictive tiers of marketing value—without using sensitive credit data.

Prioritize your most profitable prospects with predictive datapoints like Credit Score Proxy

See our complete guide: [How predictive lead scoring works — and why it outperforms traditional methods](/blog/predictive-lead-scoring-b2c)

Every marketer knows the feeling: you’ve invested heavily to drive demand, nurtured warm leads, and handed them off to the call center — only to watch conversion rates flatline.

The call center’s job is simple in theory: take warm leads and turn them into real dollars. But too often, reps are churning through lists, treating every lead the same, and burning time on the wrong prospects. The result? **Terrible conversion rates, frustrated teams, and wasted budget.**

## The real problem isn’t marketing — it’s targeting

Before you point the finger back at your CMO, it’s worth asking what really happens once a lead hits the call center.

The challenge might not be the number of leads or the skill of the reps — it’s whether the right prospects are even getting attention in the first place. Without a way to distinguish high-intent buyers from tire-kickers, reps waste valuable time chasing the wrong people while genuine opportunities slip away.

And the unfortunate reality today is that many call centers still operate on instinct and raw volume rather than intelligence, leaving conversion potential untapped.

## How predictive scoring fixes the problem

But it doesn't have to be this way anymore\! [Custom predictive datapoints](/blog/what-data-do-you-need-for-marketing) like **Likelihood-to-Convert** allow you to [score leads in real time](/blog/real-time-vs-batch-which-drives-roi), separating serious buyers from casual interest before they ever reach a rep.

Take [a national DTC home services retailer](/blog/dtc-homeservices-uses-credit-score-proxy) we work with as an example. They were seeing thousands of leads pour in, but their call center was overwhelmed and conversions lagged. Instead of asking their reps to just "work harder," they came to Faraday.

Together, we built a custom predictive datapoint using their own first-party-data as well as the [Faraday Identity Graph](/features/faraday-identity-graph) (FIG) which contains 1400+ datapoints on 240M+ US consumers and their households. This datapoint identified which households were most likely to purchase.

This allowed every new lead that entered the center to be automatically given a score on their likelihood to buy and the call center could focus its most experienced reps on those high-scoring prospects while still nurturing the rest with lower-cost resources.

The difference was immediate:

- **In 2 days:** 2% lift in contact rates
- **In 1 month:** 4% lift in conversions
- **In 3 months:** Predictive scoring became the nucleus of their tech stack — saving over $100K/month in wasted effort!

[![Diagram showing set rate vs. date](/images/blog/american-standard-chart.png)](/images/blog/american-standard-chart.png)

## Why it matters to you

For marketing executives at consumer brands, especially in high-consideration sectors like financial services, solar, and home improvement, this isn’t just a call center optimization trick.

It’s a way to:

- Protect marketing ROI by ensuring warm leads don’t go cold in the call center.
- Improve conversion efficiency without increasing ad spend.
- Build sustainable competitive advantage with predictive intelligence your competitors can’t replicate.

## Conclusion

Call centers don’t have to be a black hole where good leads go to die. By pairing your first-party data with predictive intelligence, you can ensure that your marketing investment actually turns into revenue. When you know which prospects are most likely to convert, you can transform your call center from a cost center into a growth engine. Smarter targeting protects ROI, boosts conversions, and builds lasting competitive advantage.

Find out the difference smarter lead scoring can make today. [Come say hi.](/talk-to-sales)

Predictive lead scoring transforms call centers from cost sinks into growth engines by ensuring reps focus on the prospects most likely to convert.

All posts