Let's take a look at some of the latest features in Faraday.

## Archiving resources

We’ve found that Faraday users love to experiment — implementing pipelines in Faraday to predict all kinds of behaviors in different ways. An unfortunate side effect of this is that your Dashboard can get pretty cluttered. Of course, you could always delete your resources, but what if you actually want them back someday?

For this reason, we’ve added the ability for you to archive resources you’ve created in Faraday.

[![Screenshot of the resource archival screen](/images/blog/product/archive-resource.png)](/images/blog/product/archive-resource.png)

To archive a resource in the Dashboard, simply click the … button on the resource, and click archive. If the resource you’re archiving has downstream dependencies–like an outcome that uses a cohort you’re trying to archive—those will be archived as well. Resources that you've archived can be found using the _archived_ filter in the list view of that resource type. To unarchive a resource, click the unarchive button in the same menu.

For API users, each resource has a POST request specifically for archiving and unarchiving, e.g. [archive cohort](/docs/reference/archivecohort).

## Deployments: new look & the ability to rename

We’ve given the Dashboard's [pipeline](/docs/abstractions/pipelines) deployment card a facelift so that it’s easier to tell, at-a-glance, what any given deployment’s purpose is–and added the ability to rename them, too.

[![Screenshot of the deployment screen](/images/blog/product/new-deployment-card.png)](/images/blog/product/new-deployment-card.png)

The new deployment card now only displays the most critical info by default: [connection](/docs/abstractions/connections) you’re deploying to, representation format (hashed, identified, etc), and any filters you’ve added (e.g. top 20%). Additional info can be found inside "more details."

By default, a new deployment’s name will be assigned as “Deploy to [connection_name],” but you can now rename them via the ... menu in the upper right of the deployment card. Additionally, you’ll find an indicator for when [preview mode](/docs/abstractions/pipelines#preview-mode) is enabled in a deployment card.

For API users, deployments (targets) can be renamed via the PATCH request [update a target](/docs/reference/updatetarget)'s `name` option.

---

Check out our [changelog](/changelog) for a dated list of updates.

Explore the latest Faraday updates with the ability to archive resources and a new look for pipeline deployments.

What's new: archiving resources & a new look for deployments

**FOR IMMEDIATE RELEASE**

BURLINGTON, VT — Faraday, the leading infrastructure provider for customer behavior prediction technology, has announced the beta release of its newest prediction: Recommendations.

Users can employ this new prediction to determine which action among several each customer is most likely to take. For example, a retail brand could use Faraday Recommendations to personalize an email campaign, highlighting the “next best offer” each individual customer is most likely to buy. An insurance carrier could use this new technology to predict which insurance line each customer is most likely to adopt next in order to shape a more responsive customer journey.

Recommendation joins Faraday’s other two generally-available predictions—Outcomes, which gauge the probability of each individual exhibiting or not exhibiting a certain behavior, and Personas, which organize a group of customers or other individuals into thematically coherent subgroups—to round out its platform. A fourth, Forecast, which predicts how much an individual will spend with a given brand (also known as lifetime value or LTV), is available in “alpha” with a broader release aimed for later this year.

As with all Faraday predictions, Recommendations are powered by best-in-class AI algorithms combined with panoramic built-in consumer data—over 1,500 attributes illuminating nearly every adult consumer.

“At Faraday we believe that every customer experience should and will be predictive,” said Andy Rossmeissl, the company’s CEO. “Our mission is to predict every customer behavior, responsibly, and our new Recommendations product is a major step in delivering that mission.”

Bespoke Post, a leading e-commerce subscription brand, used an alpha release of Faraday’s Recommendations offering to improve the “add-on” bundling experience at checkout. When this feature was used, it delivered a 10% conversion rate lift compared to a control group.

Meriwest, a large credit union operating in the San Francisco Bay and Silicon Valley areas, also used Recommendations in alpha, this time to predict which financial service each member is likely to adopt next. The credit union plans to use these predictions to personalize email campaigns, with the goal of boosting conversion rates.

At the technical level, Faraday’s Recommendations product employs a variety of time-tested algorithms which work in concert to deliver high-confidence predictions. Building a recommendation “model” involves two stages. First, the system uses matrix factorization to implement a “collaborative filtering“ approach. Next, the system uses findings from the collaborative filtering model to inform a decision tree-based approach (random forest ensemble). This technique allows Faraday to make predictions about existing customers as well as brand-new leads or even unknown prospects.

“As an AI engineering project, this was an exciting challenge,” said Dr. Mike Musty, a Faraday data scientist, who was instrumental in designing the product. “Delivering self-service recommendations at scale required several novel optimizations, including our hybrid algorithmic approach.”

Existing and new Faraday clients looking to take advantage of the new Recommendations product while in beta should contact their account manager or file a support request.

## About Faraday

Faraday is an infrastructure platform that offers unified AI and data for predicting customer behavior. Based in Burlington, Vermont, the company makes billions of predictions every day for thousands of brands and their partners. Learn more at [faraday.ai](https://faraday.ai).

Faraday has announced the beta release of its new Recommendations product, which predicts the most likely action each customer will take, joining their existing Outcomes and Personas predictions. This new feature aims to enhance personalization in customer interactions and has already demonstrated success in improving conversion rates for early adopters.

Press release: Faraday launches Recommendations product into beta

_This article is a followup to our [bias reporting](/blog/faraday-bias-reporting) article, and aims to provide a high-level overview of how Faraday users can mitigate harmful bias._

## What is bias mitigation in machine learning?

Bias mitigation in machine learning encompasses a set of methodologies and techniques devised to diminish and ideally eradicate unjust and discriminatory predictions inherent in the decision-making processes of artificial intelligence (AI) systems. This is a key component of [Responsible AI](/features/responsible-ai).

Bias, often unintentional, may infiltrate these systems as a result of the historical data they are shown, leading to inequitable treatment of specific demographic groups. For example, there is a history of [discrimination](https://www.investopedia.com/the-history-of-lending-discrimination-5076948) against specific minority groups in financial lending, and this history can perpetuate bias if it is not intentionally mitigated.

It is important to note that not all algorithmic bias is harmful. Predictions about, for example, the likelihood of buying products that are exclusively relevant to, say, certain age groups will and should discriminate in favor of those age groups. For the remainder of this article, when we discuss bias, we’re excluding innocuous bias like this and focusing on harmful bias.

Bias mitigation ensures that AI systems render decisions characterized by impartiality and fairness, irrespective of variables such as race, gender, or other sensitive attributes. This undertaking is pivotal in the development of AI systems that uphold principles of fairness and equality for all individuals.

The subsequent sections of this article describe how Faraday’s bias mitigation feature allows users to create fair propensity predictions. For more on bias management in Faraday, read our [bias management announcement](/blog/faraday-bias-management-announcement).

## How to reduce harmful bias in machine learning

When predicting propensity, Faraday can mitigate bias in two broad ways:

1. The first step is to ensure that the predictive algorithm considers every demographic group as equal. In order to do so, Faraday rebalances your historical data before submitting it to the algorithm for pattern discovery.
2. Second, when later making predictions, Faraday will adjust the nominal propensity ranking to ensure that the highest scorers for each demographic are ranked highly overall. This ensures that the predictions are meeting the fairness objective while maximizing predictive lift.

Some Faraday terminology will be used here, and will link to documentation.

### Unbiasing historical data

In order to mitigate bias in a client’s existing customer data, Faraday assigns different “weights” to each customer in the data to fine-tune their influence on the model's learning process. The goal is to address and rectify biases present in the data, ensuring that the model gives equal consideration to all relevant examples.

To understand this process, consider an example: a home mortgage company. Given historical disparities arising from discriminatory laws and government policies (among other reasons), some demographics are less likely to appear in the company’s customer data.

Formally, Faraday will discover this by noticing a difference between the “attainment” (all customers) and “eligible” (regional adult population) [cohorts](/docs/product/cohorts) in one or more dimensions of concern. For example, we could find fewer women than men in the historical customer data, even though the regional population is an even split: a disparity that has no rational basis. Without correction, the algorithm would believe (incorrectly) that there is some intrinsic quality which makes men more likely mortgage customers than women.

To unbias the historical data, Faraday would give higher weights to customers in underrepresented or disadvantaged groups, increasing their influence on the algorithm while it is looking for salient patterns. Building on the mortgage example above, Faraday could give higher weight to the proportionally fewer women customers, evening out their influence versus men. This results in more equitable predictions for individuals belonging to these groups.

By adjusting the weights assigned to different customers, Faraday aims to create a more balanced and fair learning process, ultimately leading to AI models that provide fairer outcomes across diverse demographic groups.

### Unbiasing predictions

When you include a propensity [outcome](/docs/product/outcomes) in your [pipeline](/docs/product/pipelines), Faraday can take additional measures to rectify potential biases in the resulting predictions. After the initial predictions are made, this approach evaluates and adjusts them to ensure fairness and reduce discriminatory outcomes.

Given that the first step eliminates bias in the historical customer data, why is this second step even necessary? When we unbias historical data, we can't alter the algorithm to completely ignore the overrepresented population in favor of the underrepresented one–we’re simply telling it to pay more attention to the latter until it has a roughly equal representation. This weighting can have a cascading effect that creates biases of its own, though.

To take the mortgage example even further, maybe the women customers tend to be older than the non-customers, whereas the men customers tend to be younger than their non-customer counterparts–after the weight adjustment, this can make the trait _age_ mean that an individual is more likely to attain the outcome if they are older, unfairly penalizing those without that characteristic, which is why this step is crucial.

In this step, Faraday looks for differences between the demographic makeup of the pipeline’s eligible population and the subgroup receiving the highest scores. The predictions are then adjusted to align more closely with the desired fairness objectives.

This method allows for the correction of biases that may have been present in the original predictions, promoting a more fair distribution of outcomes across different demographic groups. By incorporating fairness considerations after the initial predictions are generated, prediction this re-ranking contributes to developing machine learning models that exhibit improved fairness and reduced problematic disparity in their decision-making processes.

## Mitigating harmful bias with Faraday

As of the publication of this article (February 2024), Faraday currently allows for bias detection and mitigation for the **age** and **gender** dimensions (these features are often referred to as protected classes or Dimensions Of Concern).

When you define a propensity outcome in Faraday, you can choose how you want the system to handle bias:

- **No mitigation**: The outcome fairness is only reported as part of the outcome dashboard. No further transformation is applied to ensure that protected classes are treated fairly. This is perfectly appropriate in cases where the product or service being sold is intrinsically linked to certain demographic groups.
- **Equality**: This mitigation strategy ensures that demographic patterns are similar between the eligible population and the target group (those with the highest scores). For example if “young adult, women” represent 25% of your eligible population, we ensure that any population targeted from the pipeline will contain 25% of “young adult, women”.

  The chart below depicts the effects of historical data bias mitigation using the “equality” setting with three populations (A, B, C). Population **A is underrepresented** in the customer data (attainment population)—30% are customers vs. 47% in the overall population while **B** and **C are both overrepresented**. The mitigation process ensures that in the target group, population **A** accounts for 47% while **B** and **C** account for their original 14% and 39% respectively.

  [![Equality bias mitigation](/images/blog/faraday_bias_mitigation/equality-mitigation.png)](/images/blog/faraday_bias_mitigation/equality-mitigation.png)

- **Equity**: This mitigation strategy “inverts” bias so that populations that were historically underrepresented are now instead prioritized.

  In the example below, the customer group (attainment cohort) consists of 30% of population **A**, 20% of population **B**, and 50% of population **C**. The equity mitigation is going to promote the underrepresented populations **A** and **B** at the expense of population **C**. The target population will contain 37% of **A**, 47% of **B**, and 16% of **C**.

  [![Equity bias mitigation](/images/blog/faraday_bias_mitigation/equity-mitigation.png)](/images/blog/faraday_bias_mitigation/equity-mitigation.png)

Note that when mitigating bias, Faraday will consider all dimensions simultaneously, leading the system to consider subpopulations like “young adult, women” and “middle-aged men.”

## Does mitigating bias hurt performance?

Because the bias mitigation process necessarily alters the ranking and promotes sub-populations with otherwise lower predicted scores, it can reduce lift.

That said, the impact is often small. The top two graphs below depict the impact of **equality** mitigation on Gender. The lift curve for the un-mitigated model is only slightly better than the one for the mitigated model. The main difference lies in the 20% to 40% score percentile range with lift in that range decreasing from ~300% to ~220%.

The two bottom graphs show the mitigation at work. On the left, the un-mitigated model will create a target group with 2x too many males.

After mitigation, the graph on the right shows only a slight discrepancy between the fair proportion of each gender, within 3%. This is a significant fairness improvement over the initial target population breakdown with only a minor reduction in performance.

[![Bias mitigation comparison](/images/blog/faraday_bias_mitigation/mitigation-comparison.png)](/images/blog/faraday_bias_mitigation/mitigation-comparison.png)

## Why mitigate bias?

Bias isn’t inherently harmful. Earlier in this article, we touched on innocuous bias that occurs when products or services are explicitly designed for and marketed toward certain groups, like a womens’ swimwear brand marketing mostly to women. It’s the _other_, harmful–and often hidden–bias that should be surfaced and mitigated.

With Faraday, bias mitigation is easy. If you find that a predictive outcome you created has bias that you want to mitigate, simply create a new outcome using the same cohorts of people, toggle mitigation on for the sensitive dimension in question, select your mitigation strategy, and save the outcome. Once it finishes building, compare the bias report from the un-mitigated outcome with the report from the mitigated outcome. Alternatively, use the **bias_mitigation** object with the API call to [create](/developers/reference/createoutcome) an outcome.

[![Bias mitigation comparison](/images/blog/faraday_bias_mitigation/bias-mitigation-in-app.png)](/images/blog/faraday_bias_mitigation/bias-mitigation-in-app.png)

It’s in the best interest of anyone developing or using AI to do so responsibly. Whether you’re using [generative AI](https://www.bloomberg.com/graphics/2023-generative-ai-bias/) for email copy and creative, predicting customer behavior like we walked through in this article, or something else entirely, you should be wary of the existence of bias in the data that the algorithms are using, and how it impacts who you market and sell to. By employing responsible, [explainable AI](/blog/what-is-explainable-ai-why-is-it-important), you can build a foundation of trust between not only yourself and your customers, but between your business and your data.

Responsible AI isn't just good business–it’s the right thing to do.

Mitigating harmful bias with Faraday is simple, whether through our developer API or in-app. Learn more about how Faraday mitigates bias in your customer predictions.

How Faraday helps mitigate harmful bias in machine learning

With increased expectations to provide meaningful customer experiences—and shrinking budgets—brands are looking for smart engagement techniques that use the tools they already know, love, and pay for. Join our team of experts from Faraday, Google, and Hightouch as we delve into the world of predictive customer engagement, offering insights on how you can elevate your strategy with your existing stack.

**In this session you’ll:**

- Uncover the potential of predictive models like propensity, spend forecasting, next-offer recommendations, and personas in enhancing and personalizing your customer engagement strategy.
- Gain valuable insights into the mechanics behind crafting precise predictions.
- Learn practical strategies for seamlessly deploying insights into your customer engagement tools.

With increased expectations to provide meaningful customer experiences—and shrinking budgets—brands are looking for smart engagement techniques that use the tools they already know, love, and pay for.

How to unlock data-driven customer engagement with predictive AI

Quantifying bias in predictive models is the first step towards building models with fair treatment of protected groups.

Faraday bias reporting: How we measure & report on bias

One of Faraday’s founding pillars is the use of [Responsible AI](/features/responsible-ai), which includes both preventing direct harmful bias and reporting on possible indirect bias. We’ve been building these features directly into the platform over the past several months, and today we’re going to walk through two new features in Faraday with **bias reporting** and **bias mitigation**.

## Bias reporting

To see the bias report on an outcome, head to [outcomes](https://app.faraday.ai/outcomes) and scroll down to the **bias** section to see the breakdown of bias discovered in the _data_, _power_, _predictions_, and _fairness_ categories, which we'll dive into below. These can also be retrieved via the API using [get outcome analysis](/developers/reference/getoutcomeanalysis).

[![Bias reporting on an outcome](/images/blog/faraday_bias_reporting/outcomes-bias-reporting.png)](/images/blog/faraday_bias_reporting/outcomes-bias-reporting.png)

Faraday uses four categories to report on bias for an outcome:

- **Data**: Measures selection bias in the underlying cohorts used in the outcome. In a Faraday outcome, two labels exist for the purpose of this blog post: _positive_, or the people from the attainment cohort that were previously also in the eligibility cohort, and _candidate_, or the people from the eligibility cohort.
- **Power**: Measures outcome performance on a subpopulation compared to baseline performance–for example, Faraday will compare how well the outcome performs on the subpopulation “Senior, Male” compared to everyone else.
- **Predictions**: Measures proportions of subpopulations being targeted compared to baseline proportions in order to see if the subpopulation is over or under-represented.
- **Fairness**: Measures overall fairness using a variety of common fairness metrics in the literature. For example, Faraday will look at whether or not the subpopulation "Senior Male" is privileged or underprivileged.

Each bias category uses groups of people called subpopulations, which are combinations of people defined in the [understanding bias reporting](/docs/product/outcomes#understanding-bias-reporting) section's **sensitive dimensions** dropdown table. Examples of these are "Teens with gender unknown" and "Senior Females." We're launching with two of these sensitive dimensions in **age** and **gender**, but will be rolling out more over time.

If you discover that one of your outcomes includes harmful bias, you can create a new outcome using the same cohorts, but enable bias mitigation options while creating it to compare results.

## Bias mitigation

When creating a new outcome, you can choose to mitigate bias by using the toggles in the **bias** section. Check out our [outcomes documentation](/docs/product/outcomes#mitigating-bias) for info on different bias mitigation strategies.

[![Bias mitigation in an outcome](/images/blog/faraday_bias_reporting/outcomes-bias-mitigation.png)](/images/blog/faraday_bias_reporting/outcomes-bias-mitigation.png)

<Callout type="warning" title="Taking action on bias">
Mitigating bias isn't something you should set-and-forget, meaning it's not recommended that you toggle both **age** and **gender** on for every single outcome you create. For example, if you're a womens' swimwear brand, your data will skew heavily toward women, and mitigating gender would negatively impact your lift.
</Callout>

Always keep the populations that you're making predictions on in mind when considering bias. In the above example, it's _totally_ fine to ignore the biased skew towards a certain gender or age group because they're your ideal customers.

We're thrilled to continue to drive Responsible AI forward by providing agency for our users to both discover and mitigate bias in their customer predictions. Stay tuned for a deep dive into the science behind how Faraday determines bias.

Learn how Faraday's bias management suite enables you to analyze and mitigate bias found in your customer predictions.

Introducing Faraday bias management

I've always loved Zapier. It really does feel like a superpower to set up a workflow with a zap: the gift that keeps on giving. This section has been on their homepage longer than anything else and I can see why — Zapier users all feel this way.

[![Testimonials from Zapier users](/images/blog/zapier/testimonials.png)](/images/blog/zapier/testimonials.png)

That's why I'm so excited and proud to share that you can now [incorporate Faraday predictions into your zaps](https://zapier.com/apps/faraday/integrations). With this new integration, you can score leads, assign customers to personas, make product recommendations to shoppers, forecast spend, and more, all within your automated Zapier workflows.

Predictions are the key to improving customer engagement throughout the lifecycle—you can personalize outreach, customize offers, prioritize better opportunities—the list goes on.

## Working through an example Zap with Faraday predictions

Let's say we want to score new leads when they arrive in HubSpot, evaluating them for their likelihood to close and assigning them to a persona. This way, you'll be able to tailor your handling of these leads differently according to their predictions.

### Faraday setup

We'll assume you already have basic customer and lead data, along with their respective cohorts, set up in your Faraday account. If you're new to Faraday, check out the [docs](/docs/intro/welcome) and [Faraday Academy](/docs/academy) to get up to speed.

Next, you'll need to set up your two predictions if you don't have them already:

1. **Lead scoring outcome** — [recipe](https://app.faraday.ai/recipes/lead_scores_in_lookup_api) or [docs](/docs/how-to/predictions/likelihood-to-convert)
2. **Customer personas** — [recipe](https://app.faraday.ai/recipes/persona_assignments_in_lookup_api) or [docs](/docs/how-to/predictions/customer-personas)

Finally, if you don't already have it, create a pipeline to get these predictions deployed:

- **Population** – everybody
- **Payload** – the two predictions above
- **Deployment** – Lookup API

That last bullet is important to get things working right on Zapier.

### Creating your zap with Zapier

I love Zapier's new LLM interface for creating zaps, and it works great here. Just head to your dashboard and type "When a new lead comes into HubSpot, score it using Faraday," like so:

[![Zapier prompt saying "When a new lead comes into HubSpot, score it using Faraday"](/images/blog/zapier/prompt.png)](/images/blog/zapier/prompt.png)

Zapier should automatically calculate your zap template for you—something like this:

[![Zapier template for "Score new leads in HubSpot with Faraday"](/images/blog/zapier/template.png)](/images/blog/zapier/template.png)

Hit that "Try it" button. This should load the template up in the zap editor, which will look something like this:

[![Zapier zap editor for "Score new leads in HubSpot with Faraday"](/images/blog/zapier/editor.png)](/images/blog/zapier/editor.png)

Click the HubSpot box to pull up settings for this trigger step. You'll have to choose your HubSpot account and then pick your lead capture form from the dropdown. It will look like this:

[![Zapier HubSpot settings for "Score new leads in HubSpot with Faraday"](/images/blog/zapier/hubspot.png)](/images/blog/zapier/hubspot.png)

Hit the test button, and it will find a recent lead from your form data. When you hit continue, you'll make your way into the Faraday action step settings.

Choose or add your Faraday account (you'll need your Faraday API key for this), then proceed to the Action section. In the "Select pipeline" dropdown, choose the pipeline you created earlier. If it's not on the list, make sure your Lookup API deployment was properly added to this pipeline in Faraday.

Then, use the form data from HubSpot to fill out as many of the identity fields as you can. In this case, we capture email in our lead form, so I'm using that (actually, the Zapier LLM had that preconfigured for me ✨):

[![Zapier Faraday settings for "Score new leads in HubSpot with Faraday"](/images/blog/zapier/faraday.png)](/images/blog/zapier/faraday.png)

After testing what you've got so far, it's time to store your predictions for this new lead somewhere. You could put them back in HubSpot with an "Update Contact" action, or whatever you'd like. Then publish your zap and you're all set!

**Ready to try Faraday with Zapier? [Create a free account](https://faraday.ai/signup) to get started today.**

Zapier is everybody's favorite workflow automation tool. Now, you can retreive predictions from Faraday in your Zaps.

Launch: Faraday predictions now in Zapier

AI is on everyone’s minds due to its myriad use cases and applications across industries. From chatbots getting people where they want to be faster, to providing [financial planning advice](https://quickbooks.intuit.com/au/blog/accountants-and-bookkeepers/accounting-ai/), to using ChatGPT to help out with a resume or piece of code, to making [customer predictions](/features/predictions) for likelihood to take an action and personalization–AI isn’t going anywhere anytime soon.

[64% of businesses](https://www.forbes.com/advisor/business/ai-statistics) feel that various aspects of their organization will see increased productivity due to AI, and it’s no surprise when you think about the countless tasks that people have been doing for ages–segmentation, data analysis, and so on–and how much time & effort is saved by offloading them to AI. The use of AI for generating content in particular has scaled so fast that discussions abound on the impact that this AI revolution can and will have on human jobs, and brands like [NPR](https://www.npr.org/2023/06/07/1180768459/how-to-identify-ai-generated-deepfake-images) and [MIT](https://www.technologyreview.com/2022/12/19/1065596/how-to-spot-ai-generated-text/) are publishing guides for how to spot AI-generated content.

## The black box problem

The issue at the heart of this AI quandary is known as the [black box problem](https://umdearborn.edu/news/ais-mysterious-black-box-problem-explained). In short, the problem is that while what we’re seeing AI accomplish is incredible–and seemingly accurate in many ways, at first glance–by and large, we don’t understand how it reaches its particular conclusions, and what data it uses to reach them.

Faraday has long pursued making AI that uses [responsible data](/features/responsible-ai) to make responsible predictions accessible for all, and part of this pursuit has been tackling the black box problem. Starting today, every customer propensity prediction made using Faraday infrastructure can include _explainability metadata_.

## Explainable AI now in Faraday

[![Image of explainability checkbox in Faraday](/images/blog/product/explainability/explainability-checkbox.png)](/images/blog/product/explainability/explainability-checkbox.png)

Sounds good, but what does this mean?

Faraday has always provided detailed technical reports that analyze the impact of key data points on predictions at the overall level.

Now, each prediction, down to the individual level, has its score explained by the features in your data that had the highest impact on that individual’s prediction.

[![Image of a Faraday outcome feature importance for number of children and age](/images/blog/product/explainability/explainability-features.png)](/images/blog/product/explainability/explainability-features.png)

In the above example outcome, we see that, overall, the **approximate number of children** and **age** are the most predictive factors. And we can see what ranges lead to what average impact. For example, people with no kids have—all else equal—a slightly higher probability than those with 1–2 kids. This kind of aggregate reporting has always been available in Faraday.

[![Image of a Faraday score explainability output](/images/blog/product/explainability/explainability-output.png)](/images/blog/product/explainability/explainability-output.png)

Above, we see an example of CSV output from a pipeline. John is impacted by this outcome’s major factors, age and number of children, and leaves him with a low probability.

Jane, on the other hand, exhibits a more unusual combination of traits that give her a much higher probability for different reasons. For Jane, both her household income and millennial lifestyle saw her conversion probability higher than John because this business often sees conversions from people with those traits–even if they’re not the dominant traits.

## Explainable AI is Responsible AI

The central idea of Responsible AI is to understand our models—how they work, why they produce the results they do—enough to identify and correct their shortcomings. That’s why we have invested so heavily in, for example, bias detection and mitigation here at Faraday.

But even as we develop a strong understanding of and confidence in our models in the aggregate, we must not lose sight of the impact that prediction can have on individuals. We should demand nothing less from our AI models than that they explain themselves. Explainability is as core to Responsible AI as transparency and fairness.

A simple statement that explains a given prediction may not seem like much at first glance, but we feel otherwise. The _just trust us_ hand-wave that has been the motto engraved on the black box around AI for decades has been preventing our industry from moving forward with respect and responsibility for businesses and people alike–until now. Now you can see exactly what data helped AI reach the conclusion that it did, and feel confident that you’re making responsible predictions with responsible data.

**[Create a free Faraday account](https://faraday.ai/signup) today.**

Learn how every individual you make customer predictions on had their score calculated.

Removing the black box around customer predictions with Faraday score explainability

Traditional approaches to personalization and customer engagement often miss the mark in understanding the constantly evolving preferences and behaviors of individual customers. In this webinar we'll explore how predictive AI can transform [personalization](/blog/predictive-personalization-machine-learning) and engagement to deliver relevant and timely interactions.

Join Directors of Accounts Alex Noyes and Nick Haggerty, as well as Director of Marketing, Perry McDermott as they discuss how predictive AI can transform personalization and engagement to deliver relevant and timely interactions.

**From this workshop, you'll learn**:

- How machine learning can be used to improve and scale personalization efforts.
- Why first-party and third-party data are crucial for personalization at scale.
- How to use product-level predictions to personalize offers and increase LTV.

Join Directors of Accounts Alex Noyes and Nick Haggerty, as well as Director of Marketing, Perry McDermott as they discuss how predictive AI can transform personalization and engagement to deliver relevant and timely interactions.

Delivering personalized experiences at scale with predictive AI

[Personalization](/blog/predictive-personalization-machine-learning) is the holy grail of growth teams everywhere. At this point, everyone knows how important it is to speak to their customers and prospects like they’re real people, and not just orders and clickstreams. It’s often easier said than done, though, and brands with a significant amount of data and a large customer base may find themselves struggling to make heads or tails of how to organize it, or even know what data is worth focusing on.

## Challenges of personalization at scale

Doing personalization at scale right can provide incredible value to your brand, and has a major impact on how your customers see and interact with you. A daunting set of challenges is presented to most brands that try to tackle the issue themselves–let’s dive into a few of the most glaring challenges.

### Data, and a lot of it

To know more about–and how to speak to–your customers, you need to have data on them. Through the data you currently have, you’ll likely have some level of insight into the demographic, financial, and transactional background of your customer base as a whole. Though this data is strong, and gives a unique perspective due to it being exclusive to your brand, your customers exist outside of interacting with your brand. A healthy mix of [first-and-third-party data](/blog/first-party-vs-third-party-data-marketing) can provide additional layers of complexity and nuance to how you view your customers–and how you engage with them.

The problem is that you need a lot of strong data to personalize meaningfully, and the more data you collect, the harder it becomes to personalize at all.

### Finding the relevant data

Collecting the data is the first step to creating an impactful strategy for personalization at scale–but what do you do with it? Historically, data-savvy marketers have sifted through the terabytes of data they’ve collected to try and piece together what makes up their ideal customer, and what their key levers are. If your brand is thorough, this is a _ton_ of data to sift through, and the process of finding what’s relevant–on the whole as well as on the individual level–is colossal.

### Personalization at scale

By using machine learning for predictive personalization, the effort involved in scaling personalization across your entire customer base–past, present, and future–isn’t as herculean as it sounds. Machine learning algorithms can segment your customers into bespoke, cohesive personas that you can deploy [anywhere in your stack](/integrations), allowing you to create individualized, personalized campaigns for any given subgroup of people you’re interested in.

### Responsibly-sourced data

Privacy is front-and-center of any data-focused discussion in 2023–as it should be. It began with the (albeit slow-rolling) [death of third-party cookies](/blog/death-of-third-party-cookies-strategies), and will continue to be a primary focus of consumers moving into the future. Privacy matters, and largely, people are uncomfortable with things like cookies–previously critical to a third-party data strategy–tracking them creepily across the web to find out what they’re searching and where they’re visiting, even if the end-result is personalized content.

The data you collect and use should be [responsibly-sourced, opt-in, and permissioned](/features/data). It’s possible to provide a perfectly personalized experience while simultaneously being a privacy-forward brand, but it takes effort.

## Benefits of personalization at scale

With the barriers to true personalization at scale being so high, is the end-result worth it? With companies that personalize collecting [40% more revenue](https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/the-value-of-getting-personalization-right-or-wrong-is-multiplying#0) than those that don’t, the answer is a resounding _yes_.

### The right message, at the right time, every time

It’s an adage as old as marketing itself, but it’s always going to be true when a single poorly-personalized, or otherwise irrelevant message could mean the loss of a customer–or potential customer–forever. With personalized content delivered to every individual you’re engaging with, at any scale of your campaign, you’ll always be delivering the right message, at the right time, every time.

### Consumers demand personalized content

With [61% of consumers](https://www.salesforce.com/resources/research-reports/state-of-the-connected-customer/) reporting that they feel that most companies treat them as a number rather than a person, it’s clear that personalized content is near and dear to the modern consumer’s heart. As a dog lover, you wouldn’t want to receive a postcard from a pet food brand focused exclusively on cats, would you? Your customers don’t either.

### Building brand loyalty

[78% of consumers](https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/the-value-of-getting-personalization-right-or-wrong-is-multiplying#0) say that they’re more likely to buy again from brands that personalize their content. Loyal customers are also more likely to recommend brands they love to their friends & family, ensuring that, if you focus on keeping your customers happy, they’ll return the favor by shouting your brand’s name from the rooftops. Keeping a strong base of loyal customers is key to a good [churn prevention strategy](/blog/churn-prevention-machine-learning).

## How to achieve personalization at scale

It’s clear that the benefits of personalization outweigh the challenges, but how exactly can it be achieved? There are three key pillars to successful personalization at scale, and they all focus on your customer data.

### Collect good first-party data

Data is the most important feature of propensity to buy and customer predictions. Without a strong enough foundation, the predictive models that your data science team builds won’t have enough data to predict strongly off of, leaving you with questionable-at-best insights. You’ll want a strong mix of first-and-third-party data collected through tactics like web forms, surveys, and loyalty programs, where you can _really_ draw out what makes your customers unique.

### Optimize your data with first-party data enrichment

As good as your own customer data is, your customers have more going on in their lives than you can see through their interactions with your brand. You can buy data from companies like Acxiom or Epsilon, who have been collecting data for years and years, or you can leverage assets like publicly available census data to find more about who your customers really are.

### Use AI for customer segmentation

With your data collected and enriched, you need to segment your customers into unique subgroups, or personas. It can be manually done, but the process is incredibly labor-and-time-intensive, and is prone to human error. Using [AI for customer segmentation](/predict/category/persona_assignments) takes the load off of your team, leaving you ready to personalize campaigns at any scale, create internal, [persona-based talk track alignment](/blog/sealed-lead-optimization), and more.

## How Faraday supports personalization at scale

Faraday lets you easily identify distinct personas and predict customer behavior with machine learning, unlocking powerful insights to shape better personalization strategies at scale.

### Unified, responsible data

Faraday unifies your first party customer data with detailed, opt-in, permissioned, cookieless consumer data, giving you the insights your brand needs while simultaneously being privacy-centric. Hundreds of datapoints through demographics, psychographics, financials, property information, and more to be leveraged for your customer predictions.

Consumers have adapting needs, and [65% of consumers](https://www.salesforce.com/resources/research-reports/state-of-the-connected-customer/) expect brands to adapt to them. This is a laborious task manually, but with Faraday’s built-in data always staying up-to-date, your brand can adapt alongside your customers.

### Delivering personalized messaging and creative across multiple channels and touchpoints

Faraday takes your data–combined with Faraday’s–and uses AI to segment your customers into unique subgroups, or personas, based on hundreds of datapoints.

[![Faraday customer personas](/images/blog/personalization_at_scale_guide/personalization-at-scale-guide-personas.png)](/images/blog/personalization_at_scale_guide/personalization-at-scale-guide-personas.png)

With your customer personas predicted, easily kick off campaigns by deploying your entire persona set, or a single persona, to anywhere in your stack. Every individual your brand is interested in, revealed with clear-cut traits like _dog owner_, grouped with people that share similar interests, shopping behaviors, and more.

[![Faraday prediction deployment](/images/blog/personalization_at_scale_guide/personalization-at-scale-guide-deployment.png)](/images/blog/personalization_at_scale_guide/personalization-at-scale-guide-deployment.png)

Personalization at scale has never been easier, or more accessible.

### Predicting the next best product for every individual

Alongside your customer personas, you can optionally predict the propensity, or likelihood, that any action is taken by an individual. For example, take your Emerging Adult Eli customer persona and predict the likelihood that anyone in that persona will [repeat purchase](/predict/category/repeat_purchasers), and, further, predict the next best product for them based on similar shoppers.

### Providing personalized discounts to foster customer loyalty

Personalization extends beyond the copy and creative that’s in your engagement with a given customer–it can include providing strategic discounts to customers that have, for example, made a second or third purchase. This way, your customer predictions guide you toward the people that are most likely to repeat purchase so that you can then turn around and give them _just_ the right discount they need to keep coming back.

**Ready to achieve true personalization at scale? [Create a free account](https://faraday.ai/signup).**

Consumers are demanding personalized content at an increasing rate, but personalization at scale has traditionally been out of reach for many brands. Using a customer insights platform makes true personalization at scale more accessible than ever.

Product