[Dharmesh Shah](https://www.linkedin.com/in/dharmesh/) ([HubSpot](https://www.hubspot.com/)'s CTO) just [announced](https://www.linkedin.com/posts/dharmesh_breaking-news-hubspot-is-the-first-crm-to-activity-7336128750297735169-3O1-/?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAtKS-0BhomXJQm0PPBSMetJRw03HNXFZ2M) that HubSpot is now the first CRM natively connected to ChatGPT. 

This marks a major shift in how people will interact with their business data. Instead of clicking through dashboards or waiting on reports, users can now ask their CRM complex questions in plain English—and get instant answers. It’s a breakthrough in accessibility, speed, and strategy.

But while the interface is changing, the fundamentals aren’t. AI can only work with the data it’s given. So if your CRM is full of low-quality, outdated, or incomplete records, your shiny new chatbot will still give you shallow, outdated, incomplete answers.

## Your CRM can't predict the future if it's stuck with bad data

Imagine you're a CEO checking in on your sales pipeline. You fire up your shiny new CRM chatbot and ask, "Who are our warmest leads right now?" It spits out a list of accounts that sort of resemble last year’s ICP—based on a handful of clicks, a webinar signup, maybe a stale lead score from Q4. No context, no nuance, and no real signal. It’s not business intelligence, it’s data theater.

On the flip side, if your CRM is enriched with high-quality behavioral data and real-time propensity scores (delivered as custom datapoints) from a platform like Faraday, the response gets sharper: "These households show high intent for solar based on how similar they are to the last 5 deals you sold—high interest in environmental causes, EV ownership, and similar income levels. Prioritize them via SMS, and afternoons work best." Now that’s useful.

## Good AI needs great data

Everyone’s racing to integrate ChatGPT into their stack. But very few are thinking beyond the UI. The big unlock isn’t just asking better questions, it’s giving AI better context.

That’s where platforms like Faraday come in. We bring:

* Thousands of predictive datapoints from the [Faraday Identity Graph (FIG)](https://faraday.ai/features/faraday-identity-graph), including behavioral signals like payment habits, shopping styles, and lifestyle indicators.  
* Campaign-ready propensity scores that predict conversion, churn, upsell potential, and more.  
* Powerful consumer segmentation and personas that help you tailor messaging, creative, and channel strategy to match how real people behave—not just who they are on paper.  
* Privacy-safe enrichment that keeps your customer data siloed and secure.

## Better inputs, better outcomes

So yes, talk to your CRM. Ask it for trends, next steps, and hot leads. But if you want answers that are actually actionable, make sure your data is ready to support them.

Better interfaces are exciting. But better inputs are what make them work.

Want to see how smarter data makes AI work harder? [Let’s talk.](https://faraday.ai)

Smarter CRMs need smarter data—because even the best AI can’t deliver useful insights if it’s working with outdated or incomplete records.

Why smarter CRM experiences need smarter data

[![predictor leaderboard](/images/blog/predictor-leaderboard.png)](/images/blog/predictor-leaderboard.png)

We just launched something we’re really excited about: the [Faraday Predictor Leaderboard](/leaderboard). It’s a live look at the most consistently predictive traits in our platform, based on thousands of AI models and billions of real-world events.

You can filter by industry and use case, so whether you're in solar or e-commerce, doing lead scoring or churn prevention, you see what relevant-to-you traits are driving results, today. Things like “recent mortgage inquiry” or “owns a pool” aren’t just trivia—they’re signals that real models are using to generate lift.

In other words: the leaderboard gives you a peek at what’s driving value across the Faraday predictive universe, in real time.

And the takeaway? You don’t have to guess what traits matter anymore. The leaderboard shows you what’s working *right now*—so you can shortcut the discovery process and zero in on the signals that are already driving lift across industries like yours.

## How the Signal Index works

When you look at the leaderboard, the first thing you likely notice is the signal index. After all, it is the first column, but what exactly does that mean? 

Each trait on the leaderboard is assigned a signal index—a relative measure of how often it appears as an important predictor across thousands of live Faraday models. It’s not a raw number or percentage, but a way of showing which signals are standing out right now compared to others. This approach, sometimes called "indexing," is common in benchmarking and lets us highlight what’s rising to the top across industries and use cases.

## It all starts with FIG

Behind the scenes of the Predictor Leaderboard is one of the most powerful engines at Faraday: the [**Faraday Identity Graph (FIG)**](/features/faraday-identity-graph). FIG is our proprietary graph of nearly the entire U.S. adult population, connecting thousands of behavioral, demographic, and life event traits to real individuals. It’s how we detect stable, predictive signals that drive meaningful outcomes—like improving retention in financial services, increasing acquisition in solar, or boosting reactivation in e-commerce.

## So what exactly is a "trait"?

At Faraday, a trait is a specific detail about a person that helps explain their behavior—things like whether they’ve recently moved, have kids at home, or use a certain kind of credit card. When our models spot consistent patterns between these traits and real-world outcomes, they become powerful predictors of what someone might do next.

The traits on the leaderboard aren’t just interesting facts, they’re there because they’ve proven, across live campaigns, to consistently drive results. They help our models understand not just what someone did, but who they are and what matters to them.

So while the leaderboard gives you a snapshot of what’s working today, it also reveals the deeper, often surprising, behavioral signals powering our predictions.

## How to use the leaderboard

Think of the Predictor Leaderboard as a window into how Faraday models actually work.

It’s not a tool for tweaking models or choosing traits—in fact, all Faraday deployments already include built-in explainability, so users can see what’s driving predictions in their own campaigns. Instead, the leaderboard is a live demonstration of the kinds of traits that consistently drive predictive lift across industries and use cases.

If you're considering using Faraday, it gives you a real-world preview of the kinds of signals our models uncover—like how “dwelling type” shows up in solar acquisition, or how “shopping style” influences reactivation in e-commerce.

You’re not picking traits. But you are seeing what matters—and that insight can inform how you think about your audience, your positioning, and the outcomes you want to drive.

## So what’s trending this week?

Here are the top 5 traits currently driving predictions across Faraday:

1. **Shopping styles** 🛍 How a household prefers to shop (online, in-store, catalog, etc.). This trait is dominating campaigns that aim to personalize outreach based on behavior—especially in e-commerce and omni-channel retail.

2. **Age** 🎂 Simple but powerful, age is still one of the strongest predictors of conversion, churn, and reactivation, across almost every industry.

3. **Dwelling type** 🏡 Is it a single-family home? An apartment? A duplex? This trait is crucial for verticals like solar and insurance, where property type impacts eligibility and likelihood to act.

4. **Gender** ⚧ When used thoughtfully, this trait helps tailor messaging and product recommendations, especially in health, wellness, and fashion.

5. **Online household** 🌐 Does the household have a reliable internet connection? This trait is extremely predictive for digital product adoption and campaign engagement in rural and suburban markets.

Right now, these traits are being used to improve conversion rates, suppress churn, and drive smarter acquisition. That’s what the leaderboard shows: what’s working, right now, across real businesses and real campaigns.

## Iteration is the key

The leaderboard isn’t static—and that’s the point. Traits rise and fall, and we track those shifts obsessively. Not just because it’s interesting, but because it’s essential.

Getting prediction right isn’t about one great model—it’s about improving with every pass. That’s why we monitor trends over time, ask what’s changing and why, and use those insights to refine how our models learn.

This is core to how Faraday works: persistent observation, rigorous iteration, constant tuning. The result? Smarter campaigns, sharper targeting, and better results for our customers—week over week.

Want to see what traits are trending this week? [Check out the leaderboard](/leaderboard)\!

The Faraday Predictor Leaderboard offers a live, filterable view into the traits driving the most predictive lift across industries—giving you a real-time peek into what’s working in AI-powered marketing today.

Just launched: the Faraday Predictor Leaderboard

Sharing data for testing is a big step, and we know that sending anyone your closely guarded trove of customer information can be a little scary. That’s why, at Faraday, we take clear, deliberate steps to eliminate any risk to your records.

Here’s what to know about how we handle your data, in plain terms.

## NDA: At a glance

To start, we make sure the rules are clear on both sides: we’re bound by a mutual NDA, meaning confidentiality goes both ways. 

Here are some key points that keep you secure:

* As a data processor under data privacy laws, any data a client or prospect shares with us is used solely for providing our services or running agreed-upon tests. So when you send us your data, you can feel secure in knowing exactly what it will be used for and when.  
* Data security is our top priority. We’re [SOC 2 Type II audited](https://www.imperva.com/learn/data-security/soc-2-compliance/), read the next section to learn more.  
* Our mutual NDA ensures that any data you share is protected by strict confidentiality obligations.

## Security: We've got the receipts

At Faraday, we take data security seriously. That’s not just talk—we’re independently audited, fully encrypted, and built to comply with modern privacy laws. Whether you're testing data or running production campaigns, your information stays safe, confidential, and under your control.

Here’s how we do it:

* **SOC 2 Type II audited:** We undergo regular third-party audits to ensure our security controls are working as designed. This certification covers data protection, availability, and confidentiality across our systems.  
* **Compliant with major privacy laws:** Faraday operates as a data processor/service provider under regulations like GDPR and CCPA. That means we only use your data for the agreed-upon purposes—and never for our own gain.  
* **Fully encrypted:** All customer data is encrypted both at rest and in transit using industry-standard protocols. No shortcuts.  
* **Minimum internal access:** Access to data is tightly restricted, even within your own brand’s workspace. Only the team members directly responsible for running your project can view your data. (For example, the person writing this post—a marketer—has never seen a client record.)  
* **Siloed accounts:** Your data is never pooled with other clients’ data for modeling or analysis. Each account is kept fully separate, and customer-level data is only ever used to train models for your specific use case.

Want to dig deeper? Take a look at [our security page](/security).

## Validation: See it before you believe it

And why do we ask for your data before you buy? 

Transparency and evidence matter. At Faraday, we don’t believe in taking on clients we can’t generate value for. That’s why we conduct a complimentary holdout test by default, before you ever commit. This test only shows how well we can model on your data, with your data, so it requires an upload prior to signing up.

* We don’t just tell you our AI/ML models are working, we show you. To accomplish this, we run secure, real-data tests to validate Faraday’s performance before any commitment is made.  
* With these initial models, you’ll see the impact of our predictions on your specific use case, with your individual data (enriched with the Faraday Identity Graph), not just a generic demo.  
* No [black box](/blog/the-ai-blackbox-is-a-liability), no sales fluff, just results you can evaluate firsthand.  
* The results are yours to explore—you’ll see performance before making any commitment.  

Want to know how that works? Take a look at this blog about how [Faraday validates our models using your unique data](/blog/holdout-testing-with-faraday).

## In conclusion

Protecting your data is one of the most important things we do at Faraday. Our platform doesn’t work unless your data is safe—and we take that responsibility seriously. Hundreds of brands have trusted us with their data, and we’re proud to say this is one area where we consistently shine.

If you still have questions or concerns, don’t hesitate to reach out. We’re here to help.

Faraday takes a variety of measures to protect your data during pre-purchase model validation, here's the TLDR.

How Faraday protects your data during testing

Data appends—a process by which we add third-party data to what you already know about your customers—can ramp up your targeting, personalization, and market research. It’s one of the most effective ways to give your campaigns more context and reach. We’ve written before about [how Faraday handles appends](/blog/data-appends-blog), but there’s a follow-up question worth asking:

We’re not the only company offering these services and products. So what makes our brand different?

Here’s how we stand out:

### 1\. Best user experience—for developers and the rest of us

Whether you want to enrich a few records or run appends across millions, our modern [API](/docs/reference) and clear [documentation](/docs) make it easy to get started. You don’t have to worry about sending files by SFTP and waiting days for fulfillment. We’ve designed our platform to be fast, fully automated, transparent, and developer-friendly.

### 2\. High-quality data, ready for modeling

We source data from top-tier providers with robust coverage of U.S. adults, households, and consumer behavior. But we don’t stop there. We also clean, standardize, and organize every field to make it useful for AI and ready for immediate predictive modeling. You get reliable data that’s immediately actionable.

### 3\. Expert support to make it easy

When you hit a roadblock, we’re here to help. Our expert support team works closely with your team to ensure onboarding is smooth and your integration runs without a hitch. Moreover, our data science team is “forward-deployed,” ensuring that whatever help you need will be accessible, whether you’re troubleshooting an API call or just need guidance on best practices, we’re here to back you up.

## Less time managing data, more time using it

No one gets into marketing to build pipelines and dedupe records all day. From sourcing to setup to syncing, Faraday handles the grunt work so you don’t have to. The result? Less time spent fiddling with infrastructure, more time acting on insight.

Want to see what our data append services can do for you? [Let’s talk.](/contact)

Faraday’s data append services stand out by combining a seamless user experience, AI-ready high-quality data, and hands-on expert support—so you spend less time managing data and more time using it.

What sets Faraday’s data services apart?

If you read our last blog, [What’s in a deployment? How to read your Faraday outputs](how-to-interpret-your-outputs), you’ve already figured out what comes out of Faraday; now it’s time to understand how to shape it. While that post broke down the different types of payloads you can get from Faraday, this time, we’re going one level deeper: into the deployment settings themselves. 

Think of this as a user-friendly guide to all the options you might see when setting up a deployment—from filters to formats, geography to structure, and beyond. These settings help ensure that your predictions don’t just look good in a spreadsheet—they’re ready to plug into your actual tools, from CRMs to ad platforms.

If you've ever wondered what settings like "representation format," "add geometry," or “deployment analysis” actually do—or how they impact your final output—this blog is for you.

## But first, what’s a deployment again?

[Deployments](../docs/abstractions/deployments) (called “targets” in the API) are how you tell Faraday where to send your predictions and what those predictions should look like when they arrive. They’re created within a pipeline and allow for detailed customization to make sure the output fits your intended use case—whether you’re running a digital ad campaign, syncing data back into a CRM, or creating a list for direct mail.

### 1\. Where your predictions go

The first decision you make when creating a deployment is choosing its destination. This could be a Faraday-managed connection (like your CRM, ESP, or ad platform), a location in your own infrastructure (like a cloud bucket or database), or just a downloadable CSV. This is the foundation—everything else shapes what gets delivered. If you’d like to see a CSV broken down, check out that [last blog](how-to-interpret-your-outputs) we mentioned\!

### 2\. What your predictions look like

#### Representation format

In the app, we define this as the following: “Choose how you want the pipeline's output to be formatted for this target."

Think of representation format as the style or privacy level of your output—whether it’s raw, anonymized, or summarized. Deployments can be configured in a few different formats depending on how you plan to use the data:

* **Hashed**: This is the default format, typically used for ad platforms. PII is hashed to protect identity while allowing for matching.  
* **Referenced**: This format lets you use your own internal identifier to merge the data back into your system.  
* **Aggregated**: Useful for geotargeted ads or planning, canvassing, direct mail and more, this format summarizes how many individuals fall into key groups (like a persona or outcome) per geographic area. You’ll get a row per geographic area, with counts of people per persona or outcome. In this row, data can be aggregated into the following categories: ZIP code, census tract, carrier route, county, metro area, state, etc.  
* **Identified:** This format includes direct personal identifiers like name and address, allowing you to activate immediately through mailers, call centers, or CRM integrations. Because this format exposes PII, it’s only available in secure workflows with appropriate safeguards and permissions in place.

#### Column headers

You also get to choose how the column headers are labeled:

* **Machine-friendly** headers are concise and consistent, ideal for automated ingestion. Useful especially for our API users.   
* **Human-friendly** headers spell things out a bit more, which can make it easier for marketers and other stakeholders to work with the data directly, especially in tools like CRMs or email platforms. For example, “`fdy_persona_set_{UUID}_persona_name`” could appear as “`fdy_persona_set_customers_persona_name`” in human-readable mode.

### 3\. Filters, limits, and structure tweaks

#### Filters

Filters allow you to narrow your deployment to just the segments of interest. You can filter by personas, outcomes (via percentile or probability), or even recommender results (such as top product matches). This is how you target only the individuals who matter most for a given campaign. For example, give me the top 10% of scorers for my outcome. 

#### Limits

Need to limit the number of people you're deploying to? You can set a top or bottom row limit—useful for campaign budgeting, prioritization, or suppression. 

**📘Note:** in preview mode, deployments are capped at 1,000 rows. We break down the difference between preview mode and a deployment later in this blog\!

#### Structure

You can also rename or reorder columns to better match the formatting requirements of your destination platform. For example, if you're sending a list to LinkedIn or Facebook, you can use one of our pre-formatted options for easy compatibility.

### 4\. Geography & “Add geometry”

When building a geotargeted campaign, geographic options and the “add geometry” setting come into play. You can define the geographic granularity—ZIP code, metro area, census tract, etc.—and include shape data for use in map-based planning or delivery platforms. This setting only applies to aggregated deployments. 

If you’d like more information about how to read these outputs, just look at [our recent blog](how-to-interpret-your-outputs) which broke down how to read the labels\!

[![Example of what an add geometry output might look like](/images/blog/add-geometry-output.png)](/images/blog/add-geometry-output.png)

### 5\. Deployment analysis

Deployment analysis gives you a snapshot of what’s inside your deployment before you send it off. It includes summary stats on key traits and geographic distributions, delivered as a downloadable PDF or accessible via the API. This can help you validate that the list aligns with your goals—for example, checking that you’re really hitting the right income bands or urban areas before launching a campaign.

📘 **Note:** Access to traits in deployment analysis depends on your account permissions. If you’re not seeing the traits you need, reach out to your account manager to learn more about available options.

### 6\. Score explainability

Finally, if you've selected prediction explanations during pipeline setup, your deployment can also include score explainability. This details which traits had the greatest influence on an individual’s predicted score. It’s useful for internal transparency, model evaluation, or even surfacing to customer-facing teams.

If you're not sure whether you've enabled or don’t know how to read it, reach out to your Faraday team, we’ll confirm or help turn it on.

### 7\. Preview mode vs. full deployment

Before launching a full deployment, it’s helpful to run a preview—and it’s important to understand how previews behave differently. In preview mode, your deployment is capped at 1,000 rows, no matter what filters or limits you’ve set. This cap exists so you can sanity-check your configuration without triggering a full export. Once you’re ready to go live, the full deployment will reflect your entire population and apply all settings as specified—no artificial row limits included.

## Wrap-up

Deployments are more than just the final step in your predictive workflow—they’re the bridge between insight and action. With the right configuration, you ensure your models aren’t just accurate, but actually useful—delivering the right predictions, in the right format, to the right destination.

A practical guide to customizing Faraday deployments—from filters and formats to geography and output structure—so your predictions arrive ready for action.

How to shape your deployments: a guide to filters, formats, and more

If you’re a marketer asking how your team can qualify and prioritize leads more effectively, you’ve probably heard about predictive lead scoring. But what actually makes it different from traditional lead scoring, and is this ‘upgrade’ worth it?

Well, it’s 2025, and nobody’s got time for rhetorical suspense: yes, predictive lead scoring is a major upgrade, and yes, it’s worth it. Let’s break down why. 

## Before we dive in

Our CTO, Seamus Abshere, recently published a deeper technical breakdown of why simple scoring systems like RFM and first-party data alone aren’t enough to guide modern marketing strategies. That blog walks through the data science behind predictive scoring and why it outperforms old-school heuristics.

This blog aims to take that same content and break it down for non-technical readers, without the code samples or statistical deep dives. But if you’re feeling extra brainy today, or just want to try to understand the concepts like an expert data scientist would, [you can find Seamus's original post, “Why first party data (and RFM) is not enough”, here](/blog/why-first-party-data-and-rfm-is-not-enough).

## What is lead scoring?

Lead scoring is the process of ranking leads to determine how likely they are to convert into customers. This helps sales and marketing teams focus their time and resources on the people most likely to say “yes.”

There are two main approaches: traditional and predictive. Each aims to support conversion efforts, but they differ fundamentally in how they evaluate leads.

### Traditional vs. predictive lead scoring: A quick comparison

To understand the evolution from traditional to predictive lead scoring, it's helpful to see the differences laid out side by side:

| Feature | Traditional Lead Scoring | Predictive Lead Scoring |
| ----- | ----- | ----- |
| **Method** | Rule-based (manual point systems) | Machine learning trained on past outcomes |
| **Data used** | Basic firmographics, behaviors, form fills | Hundreds of behavioral, demographic, and contextual signals |
| **Scalability** | Manual tuning required per campaign | Fully automated, adaptable, and always learning |
| **Accuracy** | Assumptions-based, subjective | Outcome-driven, empirically validated |
| **Maintenance** | Static, often neglected | Continuously improved via retraining |
| **Transparency** | Clear but simplistic | Advanced but explainable (e.g., model cards) |

This comparison highlights how predictive lead scoring offers deeper insights, broader data coverage, and greater efficiency—without the manual overhead.

## Why smart CMOs are shifting towards predictive

With the complexity of today’s customer journeys and an explosion of marketing channels, relying on static, point-based scoring just doesn’t cut it anymore.

Traditional scoring is great—if it’s 2008\. But in 2025, teams are managing thousands of signals across channels, and the limitations of hand-crafted scoring systems are showing:

* **They can’t scale**—manually assigning points doesn’t work across product lines or geos.  
* **They’re biased**—they reflect gut instinct, not real data.  
* **They’re static**—the model that “worked” last quarter might be off-base today.

Predictive lead scoring flips that script. By analyzing your historical data—who actually converted, when, and why—machine learning models generate scores for new leads that reflect real patterns, not assumptions.

This shift enables marketing and sales teams to act faster and with greater confidence, knowing their lead rankings are grounded in evidence.

## How Faraday does predictive lead scoring differently

Of course, not all predictive models are created equal. At Faraday, predictive lead scoring isn’t just a model—it’s an integrated strategy designed for accuracy, usability, and transparency.

We help brands:

* **Enrich leads with real-world context** using the Faraday Identity Graph (240M+ U.S. adults)  
* **Train custom conversion models** based on your actual outcomes—not someone else’s benchmarks  
* **Deploy scores in real time** into CRMs, ad platforms, and lead routing systems  
* **Understand what’s working** with model transparency and segment-level insights

The goal is not just to predict who is likely to convert, but to help your team understand why they’re likely to convert—and how you can act on that knowledge.

## Common questions (and misconceptions)

As predictive scoring becomes more popular, a few key questions tend to come up. Here’s how we typically address them—drawing from our own platform and our CTO Seamus Abshere’s deeper technical dive on the topic.

#### “Isn’t predictive AI just a black box?”

This is one of the most common concerns we hear, and it’s a fair one. Many predictive tools offer opaque outputs with no clear path to explain how a lead was scored. At Faraday, we’ve made transparency a core design principle. Every model we build includes:

* Model explainability breakouts that document the data used, what the model was trained to predict, and how well it performed  
* Feature importance scores that show which attributes are driving conversion likelihood  
* Segment insights that break down what’s working and why in human-readable terms

If you want to dig into the technical mechanics of explainability, [this post from Seamus](blog/why-first-party-data-and-rfm-is-not-enough) goes deep on how we make models transparent and actionable, not just accurate.

#### “What if I don’t have enough data?”

Another myth. The truth is, most brands have more data than they think, they just need help unlocking it. And when first-party data really is sparse, Faraday fills in the gaps with [enrichment](/blog/identity-matching-makes-or-breaks-predictions).

We tap into our proprietary [Faraday Identity Graph](/features/faraday-identity-graph) covering 240+ million U.S. adults, enabling your models to learn from behavioral and contextual signals beyond your CRM. This lets you:

* Model outcomes even with limited internal history  
* Boost accuracy by adding real-world context  
* Reach prospects with higher confidence

Predictive lead scoring doesn’t require perfection—it just needs patterns. And sometimes, those patterns are hiding outside your own CRM.

Take one client, for example: Bee’s Wrap.

They were able to take their targeting to the next level by tapping into third-party data through Faraday.

## How Bee’s Wrap went beyond demographics

[Bee’s Wrap](https://www.beeswrap.com/), a sustainable food storage brand, needed help identifying new retail partners—but traditional targeting methods like household income or zip code weren’t cutting it. They turned to Faraday to build predictive personas based on actual behavior, not just surface-level traits.

Using third-party behavioral data, Faraday identified nuanced lifestyle patterns—like eco-consciousness and repeat purchase behavior—that wouldn’t show up in first-party data alone. That insight helped Bee’s Wrap secure distribution in over 550 locations of a major national retailer - you might've heard of them, they're called Target.

It’s a perfect example of why modern scoring needs to go deeper than form fields or simple demographic filters. 👉 [Read the full story](/blog/bees-wrap-persona-analysis)

## Want to see how your leads would score?

If you’re curious how predictive lead scoring could work with your data, we’d love to show you. Reach out to see what it looks like in practice—and how it can plug directly into your workflow.

A non-technical guide to why predictive lead scoring outperforms traditional methods—and how Faraday makes it transparent, scalable, and ready to use.

Predictive vs. traditional lead scoring, for dummies

At Faraday, transparency and explainability aren’t just ideals, they’re a core and ever-present aspect of our product strategy. We believe that understanding how and why predictions are made is critical to building trust in AI. That’s why we’re continually expanding what you can see and do inside the platform.

In a [previous post,](https://faraday.ai/blog/what-is-explainable-ai-why-is-it-important) we outlined how we expose the various steps we undertake—through explainability reports and other transparency measures—to help you understand the what and the why of your predictions. But now, we’re going a step further.

## Predictive modeling isn't magic, it's strategy

For every [Outcome](https://faraday.ai/docs/abstractions/outcomes) you define in Faraday, we build hundreds of models using a variety of strategies. These strategies define how we create candidate models, and after rigorous cross-validation, the best performers are promoted to production. 

In Faraday’s API, a *strategy* is a specific configuration or approach for building a predictive model. It may define which data sources to use, how to segment the population, and what modeling technique to apply. Strategies are automatically applied and tested behind the scenes as part of our modeling pipeline.

While this information was always available, until now, it appeared only in the technical report, buried, and inaccessible to machines. But with our latest release, that changes.

## What’s new

### API

The API for `GET /outcomes/:id/analysis` now includes a `strategies` object that lists all strategies used in the Outcome. Each includes:

* A unique ID and slug  
* Strategy name and description  
* Providers (e.g. Faraday Identity Graph, self-provided)  
* Experimental flag 

If you're inspecting results for specific individuals, you can now trace exactly which strategy applied to them.

### UI

This information now also shows up in the Dashboard. Choose one of your Outcomes, then find a new “Strategies” tab in the performance reporting.

📘 *Note:* Strategy metadata will only appear for newly generated Outcomes. Older Outcomes may not include this information unless reprocessed.

## Here’s why it matters

This update makes model strategy metadata available to humans and automations alike, enabling more transparent audits, better debugging, and smarter automation for advanced users.

It also surfaces a key part of the Faraday platform that's typically invisible: the structured experimentation that goes into every prediction, and that’s so central to our work, it’s literally where we got our name (but enough about [Michael Faraday](https://en.wikipedia.org/wiki/Michael_Faraday) and his experiments)\!

If you're building automation on top of Faraday or just want to better understand what goes into your Outcomes, this update is for you.

And if you want to learn more, ask your account manager, or visit [the Outcome API documentation](https://faraday.ai/docs/reference/getoutcomeanalysis)\!

Faraday now surfaces the modeling strategies used to build Outcomes—via API and dashboard—bringing deeper transparency and auditability to how predictions are made.

More visibility into how your models are made: strategies now available in the API and Dashboard

Welcome to **Faraday Delivers**, a new blog series where we respond directly to insights shared on [USPSDelivers.com](https://www.uspsdelivers.com/). USPS has long been a trusted voice in marketing strategy, particularly around direct mail and customer engagement. In this series, we’ll explore their key themes and offer a Faraday perspective on how predictive AI and data science can take those strategies even further.

Today, we’re responding to ["Why Personalization May Help You Win with Millennials"](https://www.uspsdelivers.com/why-personalization-may-help-you-win-with-millennials/?).

## Why this article stood out to us

USPS makes a compelling case: Millennials value personalization. According to the article, 46% of millennials are willing to share personal data in exchange for more relevant brand experiences. When done well, personalization builds loyalty, drives purchases, and helps brands stand out in an increasingly noisy market.

But to us, that stat isn’t just interesting, it’s a launchpad for smarter strategy. At Faraday, we work with brands every day to help them reach millennial consumers through relevance—and one of the best ways we’ve found to do that is through predictive personalization.

## From personal to predictive

The USPS talks about making marketing feel personal. But here’s the thing: without prediction, is it really personal? Calling someone by name or referencing their last purchase isn’t true personalization. It’s decoration. To deliver real relevance, you need to anticipate what they actually want next.

That’s where Faraday comes in. Our predictive models don’t just analyze who your customers are, they forecast what they’re likely to do next. That lets you personalize with purpose, not just polish.

For example, our models can:

* Identify which customers are most likely to convert soon  
* Forecast repeat purchases or churn risk  
* Recommend channels and messaging based on predicted behavior

And yes, that includes direct mail. Our predictive models help you decide:

* **Who** should receive a mailer  
* **Which** personas live at that address  
* **What** products are most likely to resonate

All based on real, measurable likelihood to act, not just surface-level datapoints.

## A (hypothetical but realistic) example

Let’s say a debt consolidator brand wants to re-engage lapsed millennial customers. USPS suggests a personalized postcard could do the trick.

Faraday would build on that approach:

* Use historical data to identify which customers are most likely to buy in the next 30 days  
* Segment that audience by behavioral persona (e.g. student debt holders vs. big credit-card spenders)  
* Optimize the mailer’s creative and timing for each segment

The result? Higher response rates, lower acquisition costs, and a better customer experience—because the message feels right and arrives at the right time.

## Why this matters more than ever

Millennials are tech-savvy, budget-conscious, and, most importantly, inundated with marketing. They expect brands to get them. Predictive marketing helps ensure you meet that expectation without wasting spend on low-likelihood segments.

Direct mail is powerful—when it’s targeted. And for good targeting, personalization is critical—but prediction is the key unlock.

## Stay tuned

This is the first in our "Faraday Delivers" series, where we dive into USPSDelivers.com blogs and share how predictive AI can elevate the strategies they spotlight. We’re big fans of USPS’s work, and we think there’s a lot of room to build on it with the tools we use every day.

Want to learn more about predictive personalization for your millennial audience? Let’s talk.

This blog argues that personalization alone isn’t enough—brands need predictive AI to truly engage millennials with timely, relevant outreach that anticipates their next move.

Faraday delivers: Using predictive AI to engage millennials more effectively

“So what do we actually get from Faraday? Is there some kind of Excel file I can open, with all the data points laid out in plain English?” \- Your CEO in your 1-on-1\.

Sound familiar? In this post we want to help you communicate to your leadership what they can expect the Faraday system to provide them, enabling the marketing and product workflows of their dreams.

Faraday is often used to optimize outreach (through use cases like [lead scoring](https://faraday.ai/templates/lead-rejection) and [personalization](https://faraday.ai/templates/thematic-personalization)), helping teams identify which people are most likely to take meaningful action and tailor their communication to them. But the output you get includes much more than just a score. It’s a full snapshot of each person or household in your audience: scores, traits, behavioral segments, and cohort tags.

This post breaks down exactly what you'll find in a Faraday [deployment](https://faraday.ai/docs/abstractions/deployments), and how to use each piece to drive smarter strategy and communication across the funnel.

## Outcome scores: the two numbers that tell the story

This section refers specifically to [outcome models](https://faraday.ai/docs/abstractions/outcomes)—predictions like “which of my leads are most likely to purchase,” “who will respond to outreach,” or “who’s at risk of churn.” These are the most common prediction types in Faraday, and they return two scores per person.

Every outcome prediction in Faraday includes:

* [Propensity percentile](https://faraday.ai/blog/propensity-vs-probability-predictions): how this person ranks compared to everyone else in your prediction population. For example, a 95th percentile means they’re more likely to convert than 95% of people in the group.
* [Propensity probability](https://faraday.ai/blog/propensity-vs-probability-predictions): the actual chance this person will take the action you care about. For instance, a 0.72 means a 72% likelihood of conversion.

If the percentile helps you prioritize, the probability helps you forecast. Together, they give you both a ranked list and a sense of real-world likelihood.

In this post, we’re using a CSV file as the example format for clarity, but keep in mind, the structure and content of a Faraday deployment is the same no matter how you receive it—whether via data warehouse, API, or direct integrations. 

But regardless of where you get this information, we know that all this information can feel a bit overwhelming at first. While our support team is always here to help you make sense of your results, we’ve broken things down below to give you a head start.

Below is an example of an output that you could expect to see when you request a probability or propensity outcome. These returns can look a little intimidating if you haven’t read one before so let's break down what you’re looking at.

[![Sample CSV output showing outcome score](/images/blog/how-to-interpret-outputs/prob-vs-prop-returncsv.png)](/images/blog/how-to-interpret-outputs/prob-vs-prop-returncsv.png)

A much more detailed breakdown of propensity vs. probability is explained in our previous blog, [Propensity vs probability: Understanding the difference between raw scores and probabilities](https://faraday.ai/blog/propensity-vs-probability-predictions), written by one of our incredible data scientists, [Thibault Dody](https://www.linkedin.com/in/thibault-dody/). If you'd like a more detailed explanation of how these scores are calculated and when to use them, check that one out\!

## Recommender outputs: ranked predictions across many options

While outcome models predict whether someone will take a specific action, [recommender models](https://faraday.ai/docs/abstractions/recommenders) answer a different kind of question: “Which of these options is best for this person?” Rather than producing percentiles or probabilities, they focus on relative ranking, helping you surface the best match for each person from a set of options.

For example, a recommender might tell you which product a customer is most likely to purchase, or which offer will resonate most. Instead of returning a single score, these models return a ranked list of options for each person, based on predicted likelihood.

In a typical recommender output, you’ll see:

* A **list of recommendations** (e.g., products, offers, messages)
* A **rank** indicating preference order
* A **relevance score** (optional), showing how strong the model’s confidence is

You can use these outputs to power personalized product recommendations in email or on your website, route individuals into different messaging tracks based on their attributes or behavior, and dynamically customize the content or offers they see to create a more relevant and engaging experience. This kind of real-time personalization can significantly boost conversions and improve lifetime value\!

Here, we get to see an example of a recommender model return:

[![Sample CSV output showing recommenders](/images/blog/how-to-interpret-outputs/rec-returncsv.png)](/images/blog/how-to-interpret-outputs/rec-returncsv.png)

## Personas: behavioral context you can act on

Faraday [personas](https://faraday.ai/docs/abstractions/persona_sets) are behavioral segments created by clustering your customers based on hundreds of data points. They help you go beyond just knowing who’s likely to act—you get a sense of why they might act, and how to engage them. We build these personas using a k-means clustering process, which groups similar individuals together based on shared traits and behaviors, you can learn more about how it works [here](https://faraday.ai/blog/k-means-clustering-customer-personas).

You can think of a persona as archetypes. An “Eco Enthusiast” might respond well to sustainability-focused messaging, while a “Status Seeker” may prefer premium positioning.

Here’s how you might expect a persona return to look.

[![Sample CSV output showing personas](/images/blog/how-to-interpret-outputs/persona-returncsv.png)](/images/blog/how-to-interpret-outputs/persona-returncsv.png)

Personas also help you:

* **Tailor messaging and offers** to match motivation
* **Pick the right channels**, whether that’s social, email, or direct mail
* **Inform product and UX decisions** based on lifestyle or priorities (e.g. “Busy Parents” value speed and simplicity)

Want to go deeper? Our CEO, [Andy](https://www.linkedin.com/in/rossmeissl/), just wrote a whole [blog](https://faraday.ai/blog/when-and-why-to-update-personas) about how personas work, and why they’re the foundation of various other predictive tools.

## Traits: the signals behind the scenes

[Traits](https://faraday.ai/docs/abstractions/traits) are the demographic, behavioral, and lifestyle characteristics that describe each individual in your prediction results—attributes like age range, income bracket, homeownership status, household size, and more. At Faraday, these traits are powered by our proprietary [Faraday Identity Graph](https://faraday.ai/features/faraday-identity-graph) (FIG)—a comprehensive, privacy-conscious dataset that includes rich, consent-based attributes on over 240 million U.S. adults and their households. FIG combines data from hundreds of public and commercial sources into a unified identity framework, allowing us to surface actionable insights without requiring you to collect or manage sensitive personal data yourself.

[![Sample CSV output showing traits](/images/blog/how-to-interpret-outputs/trait-returncsv.png)](/images/blog/how-to-interpret-outputs/trait-returncsv.png)

These are the features the model uses to make its predictions. They also help you:

* Filter or sort your results
* Build secondary audiences
* Spot unusual patterns (e.g., "why are so many top scorers renters?")

In short, traits let you peek under the hood. They can explain why someone scored highly—or give you clues about where to dig deeper. 

If you’d like to explore the 1,500+ rich consumer traits included by default with every Faraday deployment, take a look [here](/features/faraday-identity-graph)\!  


## Cohort membership: groups you define

Faraday also returns [cohort](/docs/abstractions/cohorts) indicators that show whether someone belongs to one of the custom segments you’ve defined—like "Past purchasers," "Subscribers," or "Churn risks."

You can also create cohorts using the data already available within Faraday. For example, you could define a cohort for men between the ages of 21–45 who live in California, Utah, or Nevada. Once created, each person in your population is tagged with a simple true/false value—do they match this cohort or not? That makes these cohorts incredibly useful as [boolean filters](https://www.scribbr.com/working-with-sources/boolean-operators/): you can use them to include or exclude groups from predictions, targeting, or reporting.

Cohort membership helps you:

* Exclude irrelevant groups (like existing customers)  
* Compare segments side-by-side  
* Layer in your own business logic  
* Understand whether a person fits within an important audience segment

If that sounds similar to Personas, there are some critical differences worth calling out:

* **Personas are discovered.** Faraday generates them by identifying patterns in behavior and demographics across your customer base.  
* **Cohorts are defined.** You create them based on your business rules and goals—like engagement history, purchase frequency, or lifecycle stage.

In other words, Personas reveal who people are; cohorts reflect what they’ve done, who they are, or how your business thinks about them.

## Prediction explanations: why someone scored the way they did

Sometimes, knowing what someone scored isn’t enough—[you want to understand why](https://faraday.ai/blog/what-is-explainable-ai-why-is-it-important). That’s where prediction explanations come in.

When enabled, explanations provide a breakdown of which traits had the biggest impact on a person’s prediction. You might see that homeownership, household size, or a specific behavioral indicator pushed a score up or down. 

[![Sample CSV output showing explainability](/images/blog/how-to-interpret-outputs/explainability-returncsv.png)](/images/blog/how-to-interpret-outputs/explainability-returncsv.png)

These explanations can help your team:

* Sense-check the model's logic  
* Find patterns across high or low scorers  
* Build trust and transparency into outreach decisions

In short, explanations turn the model from a black box into something you can interrogate and understand. For more detail on how to interpret this output, see the documentation on [deployments](https://faraday.ai/docs/abstractions/deployments#understanding-score-explainability). 

## Putting it all together

A raw score can only tell you so much.

Let’s say you know Fred is **40% likely to convert**. That’s useful. But it’s limited. Without context, you don’t know who Fred is, what kind of message might resonate with him, or whether he even fits your audience strategy.

Now imagine you have a full output:

* Fred is in the **40th percentile**, which means he ranks higher than 40% of your prediction population—and he has a **0.40 probability** of converting.  
* He belongs to your **Newsletter Signups** cohort  
* He’s assigned to the **"Value Shoppers"** persona  
* He lives in a **three-person household** earning **$50-75k**

Suddenly, you’re not just guessing—you’re seeing the full picture. Maybe he’s not your top priority, but his persona suggests he’s responsive to discounts, and he’s already shown interest by signing up for your emails. That’s enough to make a targeted, budget-conscious offer worth testing.

Understanding your outputs turns a single prediction into a story. And stories are what make strategy work. Want to see a real example? 

## In conclusion

Prediction results from Faraday aren’t meant to sit in a spreadsheet. They’re designed to drive action: better segmentation, sharper messaging, and smarter growth. When you understand what each piece means—scores, personas, traits, cohorts—you can put the full power of predictive AI to work.

Want help interpreting your outputs? Reach out to our team or explore more in the Faraday docs.

This guide walks through every component of a Faraday deployment—scores, personas, traits, cohorts, and more—so you can turn raw predictive outputs into strategic, actionable insights.

What’s in a deployment? How to read your Faraday outputs

We recently shared a blog about our new [AI-optimized Every Door Direct Mail (EDDM)](https://faraday.ai/blog/AI-optimized-EDDM) offering, which is a cost-effective way to reach a wide audience. As acquisition costs rise and ad targeting regulations tighten—especially in finance, energy, and healthcare—marketers need options that balance scale, precision, and compliance. One of those options is Saturation Mailing. But direct mail isn’t about finding one universal approach and tripling down. It’s about building a flexible toolkit and adapting your strategy based on your audience, geography, budget, and goals.

That’s where [Saturation Mailing](https://www.growmail.com/blog/eddm-vs-usps-saturation-mailing/) comes in. It’s another powerful tool in the direct mail playbook—especially when you’re trying to reach as many qualified households as possible, while still maintaining control and strategic nuance. Let’s dig into how it works and why we think it deserves a spot in your toolkit.

## What is Saturation Mailing?

Saturation Mailing is a type of bulk mailing that allows you to qualify for discounted postage rates by delivering to a high percentage of addresses within a postal route—at least 90% of all *residential* addresses or 75% of all *total* addresses within that route. It's designed for reaching entire neighborhoods or regions, without requiring address-level targeting for each individual recipient.

It’s an efficient middle ground: you don’t need to build a fully personalized list, but you also don’t need to send indiscriminately. Instead, you focus on a geographic area with broad appeal and meet the threshold for postage discounts by sending to most—but not necessarily all—households.

This approach is especially useful when your product or service has broad local appeal—like community solar, internet service, or home improvement offerings. It helps you scale up affordably while keeping some strategic nuance, especially when paired with smart suppression and geographic targeting tools.

Thanks to USPS rules around Saturation Mailing, you can get smarter by suppressing addresses that don’t make sense for your campaign—without losing your discount, as long as you stay above the required delivery threshold. This flexibility lets you skip over existing customers, low-propensity households, or any other segments that aren’t a fit—without resorting to fully targeted mail.

## The role of suppression—and how Faraday helps

One of the most powerful parts of Saturation Mailing is suppression. You don’t need to mail every single address in a route, you just need to hit the threshold. Faraday helps you suppress existing customers, low-scoring leads, and households that don’t match your ideal personas.

Our platform uses behavioral and demographic data to score every household in the U.S., so you can make informed decisions about where to send—and where to hold back.

This is also a natural extension of other Faraday services, like persona modeling to define your most valuable segments, predictive scoring to prioritize high-intent households, and deployment planning to map out the smartest execution path.

## How does it compare to EDDM and fully targeted mail?

Every Door Direct Mail (EDDM) is another bulk mailing program offered by USPS. And while it is similar to Saturation Mailing, it comes with some specific limitations. It’s limited to flat-sized mailpieces (like postcards or bifolds), and doesn’t support letter-size mail in envelopes. It also doesn’t allow you to suppress addresses or use named recipients within a route. Accordingly, targeting services like our [AI-optimized EDDM](https://faraday.ai/blog/AI-optimized-EDDM) help you choose which routes will perform best in their totality, not how to target within individual routes. 

It’s great for broad awareness plays where cost is the primary concern and response rate isn’t as critical. Although EDDM is actually between 10 and 15% more expensive than Saturation Mailing in terms of postage, it often results in lower overall campaign costs due to simplified prep and no list requirements. Accordingly, it remains a popular choice for small businesses because of its simplicity. With no need for mailing or suppression lists or complex prep, EDDM makes it easy to get a campaign out the door quickly—especially for hyperlocal efforts like restaurant promotions, political flyers, or community event announcements. But if you’re aiming for higher ROI, more precise targeting, or the ability to suppress known non-converters, Saturation Mailing is almost always the smarter long-term investment.

Fully targeted mail, on the other hand, gives you total control. With Faraday, that means using predictive models to select individual households based on their likelihood to convert, lifetime value potential, or alignment with your brand’s key personas. You can personalize messaging, time campaigns to buying cycles, and avoid waste almost entirely. It’s the most precise option in your toolkit—but it also comes with higher postage costs and more logistical complexity, which is why it’s often used for high-value offers or audiences where every conversion counts.

Saturation Mailing lands in the middle. It offers cost benefits through USPS discounts and more flexibility than EDDM—especially when it comes to address suppression and mail formats. While EDDM typically restricts you to postcards or flats, Saturation Mailing allows a wider range of formats, including letters, as long as they meet USPS requirements. It’s a great option when you want efficiency without the overhead of fully targeted campaigns.

Each of these options has a time and place. The smartest marketers know when to lean into each—and how to mix them when the moment calls for it.

## When to use it

Saturation Mailing is ideal when you’re targeting neighborhoods with broad appeal, want to blend efficiency with some degree of targeting, and need to use mail formats (like letters) that drive better response rates. It's also a great option when your audience is geographically clustered but not defined by narrow demographics.

Because you can suppress part of the list—like current customers or low-propensity leads—you get more relevance and less waste. That’s where Faraday shines.

## Faraday is more than a mail vendor—we’re a strategic partner

Faraday isn’t in the business of selling templates or one-size-fits-all strategies. We offer a range of predictive tools and mailing options—from EDDM to household-level targeting to Saturation Mailing—because we know that every campaign has unique needs.

What makes us different isn’t just what we offer. It’s how we help you use those tools together to maximize results. We don’t just help you get in the mailbox—we help you get there smarter.

## Conclusion

In direct mail, versatility is everything. EDDM, Saturation, and fully targeted delivery all have a role to play. The key is knowing when and how to use each.

With Faraday, you’re not just choosing a tool—you’re gaining a strategic partner who helps you adapt, refine, and win. Let’s build campaigns where every piece counts.

This blog breaks down how Saturation Mailing offers a strategic middle ground between EDDM and fully targeted mail—helping marketers reach the right households at scale while remaining cost-effective.

Product