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

## Dependency visualization

In Faraday, any given resource can be connected to one or more other resources, which we call dependencies. For example, a cohort you’ve built might be used to define an outcome, which is then used in a pipeline. This type of interconnectedness can make it difficult to troubleshoot issues, to see what impact editing a resource might have on other resources, and so on.

We’ve found that it can be helpful to zoom out and look at the bigger picture of where a resource fits in the larger picture of your Faraday account. To achieve this, we’ve added a dependency visualization to every resource’s detail page in the Dashboard.

[![Screenshot of the resource dependency graph](/images/blog/product/dependency-visualization.png)](/images/blog/product/dependency-visualization.png)

The dependency visualization reveals the overall layout of resources in your Faraday account, and highlights which individual resources are connected to the one you’re currently viewing. In the above example, a _signups_ event stream was used to define the _leads_ cohort, which was used to define the _lead conversion_ outcome.

This new visualization should help you understand which resources are necessary for your predictions, and which you can safely [archive](/blog/archive-resources-and-updated-deployments) or remove.

## Sortable date columns

In the past, resources you created in Faraday were sorted alphabetically by name. While alphabetical sort is simple to understand and use, we found it a bit awkward to create something only to have it be dropped into the middle of a list.
To improve this workflow, we’ve made the majority of columns sortable, and added new columns for relevant date information.

[![Screenshot of sortable columns](/images/blog/product/sortable-columns.png)](/images/blog/product/sortable-columns.png)

The created date is–you guessed it–the date that the resource was created. Modified refers to the date that the resource’s configuration was last modified (e.g. a trait was added to a cohort’s definition). Lastly, processed is the last time the output was processed by Faraday (i.e. the output changed).

---

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

Explore the latest Faraday updates with a visualization of how resources interact with each other, and new sortable list columns.

What's new: dependency visualization and sortable list columns

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

We're happy to announce a new partnership between Faraday and Simpli.fi, the leading Advertising Success Platform, which provides programmatic advertising solutions and workflow software to over 2,000 media teams, agencies, and brands.

Under this partnership, Simpli.fi will incorporate Faraday's [identity graph](/features/faraday-identity-graph) of more than 1,500 consumer attributes and individual-level [customer behavior predictions](/features/customer-behavior-predictions)into its sophisticated media buying program. Simpli.fi users can now create highly effective lookalike audiences, powered by Faraday, that enable marketers to target their best-fit customers with relevant engagement at scale.

"It's about combining the science and the art," Faraday CEO Andy Rossmeissl says while chatting with Simpli.fi's David McBee about lookalike audiences on [Simpli.fi TV](https://simpli.fi/simplifi-tv/andy-rossmeissl-simpli-fi-tv). Faraday's data science–combined with the artistic approach in which Simpli.fi makes these insights actionable–results in campaigns that show real ROI.

Clients interested in leveraging customer behavior predictions to create lookalike audiences of likely buyers and deploy them through Simpli.fi can contact their account manager to get started.

Faraday's identity graph and customer behavior predictions come together with Simpli.fi's media buying program to create lookalike audiences that enable marketers to target likely buyers with relevant engagement at scale.

Introducing a new partnership between Faraday and Simpli.fi

Is the artificial intelligence hype cycle approaching the inevitable trough of disillusionment, as hype and reality collide head-on? Is the AI bubble, much like the internet bubble of years past, about to burst?

AI is moving at a frantic pace, and with its advancement comes a massive amount of investment, which has led to inevitable market saturation and a lot of inflated expectations. As of the summer of 2023,[ ](https://www.gartner.com/en/newsroom/press-releases/2023-11-28-gartner-hype-cycle-shows-ai-practices-and-platform-engineering-will-reach-mainstream-adoption-in-software-engineering-in-two-to-five-years)**[Gartner](https://www.gartner.com/en/newsroom/press-releases/2023-11-28-gartner-hype-cycle-shows-ai-practices-and-platform-engineering-will-reach-mainstream-adoption-in-software-engineering-in-two-to-five-years)** had already identified that generative AI was reaching the 'peak of inflated expectations' and was about to enter the dreaded 'trough of disillusionment'.

Even the ever-promoting, all-is-well tech titans are starting to sound the alarm on overpromising about the power of AI. Recently both Amazon and Google have reportedly been[ ](https://www.theinformation.com/articles/generative-ai-providers-quietly-tamp-down-expectations?rc=kz8jh3&ref=wheresyoured.at)**[telling their sales teams](https://www.theinformation.com/articles/generative-ai-providers-quietly-tamp-down-expectations?rc=kz8jh3&ref=wheresyoured.at)** to tone down on hyping AI too much to customers. Today's AI leaders are concerned that they are overpromising, and underdelivering.

But all is not hype when it comes to AI, there are many real-world business applications where AI is making a big difference and delivering true value to both the business and its customers. Accordingly,[ ](https://www.gartner.com/en/webinar/567177/1278324)**[Gartner](https://www.gartner.com/en/webinar/567177/1278324)** expects 2024 to be the year where the companies that successfully deploy AI do so through a more targeted and disciplined approach.

So, what is AI hype and what is AI reality? And how are people using AI successfully in the real world to make a measurable difference to a business's bottom line?

## AI hype

The amount of AI hype is extraordinary, even in the world of technology hype. Many of us who have been around the block a few times have heard this story before. An amazing new technology arrives that will change the world, and then companies hype the technology endlessly, leading to market saturation, end-user fatigue, and missed expectations. Eventually, the technology makes its way into mainstream adoption, but typically through more distinct use cases and targeted real-world applications. This process is what **[Gartner's Hype Cycle](https://www.gartner.com/en/articles/what-s-new-in-artificial-intelligence-from-the-2023-gartner-hype-cycle)** attempts to represent graphically.

What is AI hype? When you go to a technology conference and an AI futurist talks in vague generalities about how AI will solve every future problem while giving no specific use cases. Or when omnipresent tech CEOs recycle their chatbot marketing from 30 years ago, put an AI in front of it, and tell you your AI assistant will help you in every conceivable way, like a virtual butler. This is AI hype.

For the past year or so, the AI hype discussion has been dominated by generative AI. And while some demos of generative AI were promising, their real-world application at this time has been oversold somewhat to the general public and business world.

## AI reality

While up to now generative AI has gotten much of the attention, predictive AI has become the hot topic of conversation in business circles for delivering immediate business value in challenging economic times. Predictive AI can work in tandem with generative AI leveraging the best of both technologies, but you need the right software to combine the algorithms for true value.

With predictive AI, more specifically predictive modeling and predictive analytics, companies today generate real business value in the short to mid-term by predicting customer behaviors by enabling more effective and efficient customer engagement within the customer experience.

By using software like the[ ](https://faraday.ai/)**[Faraday](https://faraday.ai/)** predictive analytics and predictive modeling platform, you can have the AI infrastructure needed to predict customer behaviors like likelihood to convert, likelihood to churn, forecast spend, and product recommendations. With these predictions, you can drive data-driven customer experience use cases like AI lead scoring, churn prevention, adaptive discounting, and next-best offer.

Platforms like Faraday are real-world targeted AI, not some general-purpose AI platform for generating news headlines. Faraday focuses on predicting the behavior of individuals, with built-in Responsible AI, including features like bias mitigation and prediction explainability.

## More real-world AI use cases

While improving the customer experience with predictive modeling and predictive analytics is a top use case for real-world AI implementation, there are several other primary use cases where AI is making significant inroads.

One of the areas already making a huge impact is predictive AI and generative AI being used to empower software developers, aka employee empowerment, allowing them to code quicker, improve time-to-market, and enhance their data with predictive algorithms.

Let's face it, existing developers are maxed out with work, and they can't build things as fast as companies want, plus they don't enjoy their coding jobs when they work on an assembly line. Today's AI coding assistants are intended to handle more routine tasks, allowing developers to focus on higher-value work. But to work well, they need to be combined with predictive AI engines like Faraday powered with Responsible AI.

"AI-augmented and machine learning (ML)-powered software engineering is changing the way software is being created, tested and operated, and the need for responsible AI is growing," said Dave Micko, Senior Director Analyst at Gartner. "Practices such as platform engineering will begin injecting insights from deployed systems into the systems being developed."

## Conclusion

The AI hype machine has done its job. People are inundated with AI marketing to the point no one knows what is hype and what is reality. Yes, generative AI is exciting and has tons of future promise. But right now it's more in the demo stage of development.

While many generative AI apps do not need predictive algorithms, for many traditional business applications, using predictive AI plus generative AI can be a good optimization to move the needle on your business.

Predictive modeling and predictive analytics are where businesses need to put their technology investment dollars to start to see real returns from AI, in the short-to-mid term.

But how a company moves forward in this process is critical. AI must be done safely and responsibly to ensure that the consumer's, business's, and community-at-large best interests are maintained.

This post explores the current state of AI, contrasting the widespread hype with real-world applications that deliver tangible business value. It highlights the overpromising nature of generative AI and emphasizes the immediate benefits of predictive AI, urging companies to adopt a targeted and responsible approach to AI implementation.

AI hype vs reality: AI in the real world

**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

The first step to grow your sales pipeline is to improve your lead generation management. By optimizing your lead generation with predictive lead scoring, you can generate more qualified leads, improve profitability, and ROI, and lower costs.

Some of the most forward-looking businesses in the world are leveraging AI and predictive lead scoring to help them identify and prioritize the most promising and valuable leads based on the criteria that matter most to their particular organization and industry. AI and ML algorithms can ingest and analyze all sorts of data, from behavioral to transactional to demographics, and from multiple sources, including CRM platforms, websites, email, social media, and analytics platforms, to score those leads based on the likeliness to convert.

AI, predictive lead scoring, and predictive modeling allow your company to focus on the leads that matter to the health and profitability of your business, aka your most valuable customers, and avoid wasting your time and resources on low-quality leads.

By using AI, companies can optimize their sales and marketing pipeline by automating and personalizing some or all of it. Depending on the software and technology you apply to your pipeline, AI-driven tools can assist in the optimization of your customer communications in their buyer journey, triggering them at the right times, frequency, and channels to improve response rates, customer engagement, and conversions.

## How to use AI to forecast spend and optimize sales

Leveraging AI and predictive insights, businesses can now get a better handle on forecasting and optimizing sales. By feeding AI/ML algorithms with not just historical and real-time observable data, but also loading them with predictive data informed by machine learning, you will be better able to predict accurate outcomes, estimate revenue, and forecast spend.

With AI and predictive modeling, business can now optimize their sales and marketing pipelines. Predictive AI software allows companies to recommend next best actions, like being offered a discount or following up, to trigger them and move them along the customer journey toward a sale. The right AI-driven software will embrace predictive AI algorithms to detect churn and determine the likelihood of a prospect or customer buying, converting, or churning.

## The power of modeling lead prioritization and lead assignments

Lead routing, or lead assignment, is where a lead goes and who it gets assigned to when it comes into your company. Lead routing can be done manually like a lot of other processes, but as your company grows, and your customers become more diverse, automating lead routing, lead prioritization, and lead assignments with AI and predictive modeling to better route those leads. With predictive AI, you can do much more advanced routing and prioritization, based on criteria such as geography, deal size, industry, and other factors germane to your sales makeup. With the right AI-driven software, organizations can use rules-based routing to assign leads to the right sales associates, ranking and prioritizing those leads based on the likelihood to convert.

## Get AI lead prioritization with Faraday  

By using **[Faraday](https://faraday.ai/)** predictive modeling and AI platform, businesses can convert more leads by using AI-driven predictions to optimize how you prioritize, route, and engage leads. With their software, companies can anticipate the behavior of each lead in their funnel, essentially turbocharging your lead handling workflows with AI.

Faraday provides the data and the predictive AI modeling infrastructure (IaaS) that enables organizations to iterate on their business challenges, instead of their technology infrastructure. With our platform, brands, and the software companies who support them, can now develop end-to-end customer experiences, at scale, driven by predictive insights.

In the end, a data-driven CX is a better CX. Faraday is the only predictive AI platform, both for **[software teams](https://faraday.ai/solutions/powered-by-faraday)** and **[brands](https://faraday.ai/predictions-for/brands)**, dedicated to predicting customer behavior. Everything you need to make customer experiences predictive is built into our software.

## The new normal

Predictive customer behaviors will increasingly drive the unified customer experience until all customer interactions are predictively AI-driven. Brands and organizations looking to leverage AI and predictive insights for business and customer value should start by using AI and predictive modeling to improve the quality and routing of their marketing and sales pipeline.

By optimizing your lead generation with AI-powered routing, lead scoring, and lead prioritization, organizations can run programs more effectively and efficiently, improving business results and consumer value.

Businesses can enhance their sales pipelines by using AI and predictive lead scoring to improve lead generation, prioritize high-value leads, and optimize customer interactions. AI-driven tools can help forecast sales, route leads, and create predictive customer experiences.

How to use AI to deliver a better quality pipeline

When you are marketing and selling consumer products, every lead counts. However, not all leads are created equal. As your business matures, it's crucial to move beyond simply filling your pipeline to effectively prioritizing leads that are most likely to convert. This is where the power of predictive lead scoring comes into play.

## What is lead scoring?

Lead scoring is the method marketing and sales departments use to rank the quality of a lead, typically in the form of a numerical point system. Leads get scored based on attributes critical to the business, including demographics, transactional behavior, and customer engagement, giving sales professionals the most promising potential customers to contact. Lead scoring systems are fed with past leads and historical data to inform the current scoring model and are typically handled within an enterprise's CRM system.

But traditional, manual lead scoring has its drawbacks. It's time-consuming for employees who could be spending their time in more productive ways, and it can result in best guesses and other types of inconsistencies in the application of the methodology.

## What is predictive lead scoring?

Predictive lead scoring represents a huge step forward for marketers and sales departments for qualifying leads at scale. By using machine learning algorithms, organizations can analyze enormous amounts of data and identify patterns to sort and prioritize leads, improving the quality of lead scoring mechanics. Predictive AI engines can ingest anything from CRM data, online data, IoT data, and behavioral data to help predict what leads will be the most likely to convert into customers.

During challenging economic times, budgets are tight, and sales and marketing teams are looking for ways to reduce costs and improve the efficiency and efficacy of their sales funnels. By leveraging AI-driven predictions, businesses can convert more leads by using those predictions to optimize, prioritize, route, and engage leads. Marketers can anticipate each lead's behavior in their funnel by infusing AI into their lead-handling workflows.

Predictive lead management works better than traditional lead management. The challenge is good predictions require robust and accurate data, which means resolving every detail down to the identity level. Only the right predictive software infrastructure will be able to address the data-oriented challenges to deliver effective predictive AI lead scoring.

## The benefits of predictive lead-scoring

With predictive lead scoring, organizations can expect to see a variety of benefits. They will be more data-driven overall, leveraging not just observed data, but predictive data to improve algorithm performance. You will get better efficiency in your prospecting with better conversation rates and shorter sales cycles. Plus, you can look forward to an improved alignment and democratized data connection between marketing and sales establishing a single source of data-driven truth for both departments to rally around.

## Three kinds of predictive lead scoring to consider

There are different types of predictive lead scoring, so it's important whatever software you choose can accomplish these three primary activities.

1. **Lead rejection** — Does your organization purchase leads from lead generators and/or aggregators? Many of these leads are not worth buying. If your lead platform supports it, you can use predictive lead scoring to reject leads that don't meet a minimum conversion probability threshold, saving you significant time and money.
2. **Lead suppression —** Does your organization spend money engaging every lead—for example, with direct mail—whether self-generated or purchased? Not every lead is worth the expense. You can use predictive lead scoring to suppress leads below a minimum conversion probability threshold—another way to save significant time and money.
3. **Lead prioritization —** Does your organization engage your incoming lead funnel with a resource-limited sales team? In many cases, this means not every lead can get the same attention, and few can get the kind of attention and engagement you know will maximize conversion. You can use predictive lead scoring to ensure your best leads get handled first and with the most robust level of engagement. This is another way to boost revenue with predictive lead scoring.

## The challenges of predictive lead scoring

One of the top challenges of deploying any type of predictive engine is making sure it is fed with quality data. The accuracy of the predictions will be driven directly by the quality of the data input into it. As the old saying goes, garbage in, garbage out (GIGO). You must have the right software platform necessary to make sure that data is complete, consistent, and accurate.

The other primary issue with predictive AI lead scoring is there is an ongoing process that involves the consistent refinement and retraining of data sets so that the predictive models remain accurate as new data comes in. These patterns can change over time.

## Faraday delivers predictive lead scoring

Patterns that predict behavior in the early stages of a customer journey eventually stop working and new ones take over. Only[ ](https://faraday.ai/)**[Faraday](https://faraday.ai/)** uses time-linked model ensembles to tell the true customer story. Faraday detects and selects the best tools for the job, from time-tested algorithms to cutting-edge generative AI Everything you need to predict customer behavior is built-in to the platform. With Faraday's rich, built-in data, you can skip expensive licensing deals and start predicting behavior immediately.

**[Faraday](https://faraday.ai/)** provides the data and the predictive AI modeling infrastructure (IaaS) that enables organizations to iterate on their business challenges, instead of their technology infrastructure. With our platform, brands, and the software companies who support them, can now develop end-to-end customer experiences, at scale, driven by predictive insights.

In the end, a data-driven CX is a better CX. Faraday is the only predictive AI platform, both for **[software teams](https://faraday.ai/solutions/powered-by-faraday)** and **[brands](https://faraday.ai/solutions/lead-management)**, dedicated to predicting customer behavior. Everything you need to make customer experiences predictive is built into our software.

Predictive lead scoring uses machine learning to analyze large datasets and prioritize leads most likely to convert, enhancing efficiency and effectiveness for marketing and sales teams. By leveraging AI, businesses can improve conversion rates, shorten sales cycles, and ensure alignment between marketing and sales, provided they maintain high-quality data and regularly refine their predictive models.

Why you need predictive lead scoring today

_This customer story has been anonymized to respect the company's competitive strategy._

## The problem: optimizing their lead scoring methodology

A national home services brand was trying to optimize their lead-to-customer conversion rate. Their existing lead handling methodology was to engage all leads upon entry into their funnel via a call center, then an SMS and email engagement drip for those leads that did not convert. Their active lead engagement window was 19 days from funnel entry to lead conversion. Leads were scored using a rules-based methodology of first-in, first-out with identity verification completed upon form fill completion. Beyond this, all leads were treated the same.

This longer sales cycle resulted in sales representatives having more leads than they could consistently follow up with. The company believed this dynamic was negatively impacting conversion rates with sales reps spending too much time on prospects with a low chance of converting into customers, and too little time on those who were a great fit for their services.

## The obstacle: limitations of rules based scoring models

Conventional B2C lead scoring models commonly utilize a rules-based approach, where human judgment determines the rules for scoring leads. Typically, this involves calculating a lead score by establishing rules from a blend of basic demographic, property, or identity details (e.g. owner-occupied household, identity verified, etc.) and observed behaviors such as website visits, email interactions, form completions, and/or call length.

Although these rule-based models aid marketing and sales teams in prioritizing their efforts, they carry notable limitations.

1. The criteria for lead scoring in a rules based approach are often subjectively chosen, leading to inaccuracies in assessing a lead's actual potential.
2. There is a cold start problem for a new lead that has exhibited little to no behavior. A new lead that has just entered the sales funnel often can't be scored, or the score is inaccurate due to a lack of observed behavior.
3. Rules reliant on demographic, property, and/or identity criteria often involve either having to ask the lead for this information, which impacts conversion rates, or licensing 3rd party data, which can be costly and increase a company's privacy compliance and liability burdens.

As a result of these limitations, this national home services team started exploring alternatives to traditional lead scoring.

## Why Faraday?

### Quantitative rather than subjective lead scoring

Unlike a rule-based lead scoring approach, [predictive lead scoring](/blog/predictive-lead-scoring-b2c) uses machine learning to analyze historical lead data and create statistical models. This approach predicts the likelihood of a lead becoming a customer through quantitative analysis, rather than qualitative assessment.

Faraday's predictive modeling eliminates subjectivity from lead scoring by incorporating both intuitive and non-intuitive data points that traditional lead scoring might miss. The model updates continuously with new customer data, ensuring that changes in the market or product offerings are reflected in the score. Unlike rule-based scoring, which can become outdated, predictive scoring always evaluates leads based on the most current and relevant information.

Additionally, Faraday's [out-of-the-box identity graph](/features/faraday-identity-graph) eliminates the need to collect identity criteria or rely on expensive data vendors. With just an email or physical address, Faraday can instantly access over 1,500 data points on nearly every adult consumer in the country, both providing a deep well of data to find signals and eliminating the cold start problem of how to score leads with very little data.

## The results

### 4x higher lead conversion

The marketing and analytics team at this nationwide home improvement brand integrated their lead data into the Faraday platform. The data used for training these models included multiple years of lead, appointment, and customer records. The Faraday platform automatically associated these records with thousands of demographic, property, financial, and lifestyle attributes found in the Faraday Identity Graph.

From there, the brand quickly built predictive lead scoring models to assess and rank leads based on their likelihood to convert into paying customers. The models found objective patterns across the client and Faraday's data that clearly distinguished a good lead from a bad lead.

To validate the accuracy and effectiveness of Faraday's lead scores, the home improvement brand engaged the agency Bold Orange to conduct a thorough analysis of the efficacy of the predictive lead scores, comparing Faraday model predictions against actual lead conversions and outcomes.

[![Image of a third party holdout analysis](/images/stories/screenshots/bold-orange-analysis.png)](/images/stories/screenshots/bold-orange-analysis.png)

The analysis revealed that the conversion rate of leads in the top 10% of lead scores was **4x higher than the brand's baseline conversion rate**, confirming the model's ability to accurately identify high-potential leads.

With this third-party verification in place, the brand deployed the models into production. The models scored the brand's leads in real-time the moment they entered their sales funnel and used the scores to prioritize leads within their call center. Now, the sales team has the confidence to know that they are talking to the best opportunity at any given moment.

A third-party analysis revealed that the conversion rate of leads in the top 10% of lead scores was 4x higher than the brand's baseline conversion rate, confirming Faraday's ability to accurately identify high-potential leads.

How Faraday's predictive lead scoring helps a national brand identify leads with a 4X conversion rate

[![DataGrid header image](/images/stories/screenshots/salesrabbit-story-header-image.png)](/images/stories/screenshots/salesrabbit-story-header-image.png)

## Making a powerful tool even smarter

SalesRabbit is the leader in software for field sales teams. Their [DataGrid](https://salesrabbit.com/datagrid/) solution supplies canvassers with household data on specific properties and homeowners in their territories, providing essential context as they knock on doors. However, the SalesRabbit product team recognized that static data points on each property provided limited context for sales reps.

They realized that, if they could identify which attributes of a homeowner or property could predict the likelihood that a prospect would become a customer, then they could use that information to inform how their sales reps approached their day-to-day responsibilities. For example, solar clients would not only know which houses had the right amount of sun exposure for solar panels, but also which of them most closely fit the model of historical solar adopters. Pest control clients would not only know what material houses were made out of, they could also know if the homeowner would be more or less likely to say yes to regularly scheduled inspections or not, and so on.

Furthermore, if they had a way to predict which neighborhoods had the highest concentration of likely customers, then DataGrid could direct sales reps' efforts more efficiently.

## The challenges of building predictive AI models from scratch

SalesRabbit identified three main obstacles to them building effective predictive AI models in-house:

- A lack of data about consumers.
- The long ramp time to build effective predictive models.
- Resources to test models for accuracy and continue to provide transparency about the models efficacy.

### Licensing more customer data

Predictive models need a deep well of data to find an accurate signal. Despite the wealth of data SalesRabbit had on homeowners, adding an ML layer to DataGrid would require thousands more data points to ensure accurate models. Psychographic data, shopping behaviors, demographic data, financial information, and more would all need to be vetted, licensed, and reconciled with their existing data before predictions could be made.

### Extensive time and resources dedicated to prediction

To build a customer prediction pipeline in-house, the SalesRabbit team would also need to grow their existing data science team. They couldn't just reorganize other team members for a temporary sprint. Not only is the typical timeline for an ML program multiple years, they’d need to invest to continuously maintain those models over time.

### Limited transparency and scalability of cost

For the C-level team at SalesRabbit, it was critical for the solution to their challenge to be scalable. The original timeline for building this feature entirely in-house was multiple years and required a major investment from their data science team to get started.

Additionally, SalesRabbit's customer base includes both enterprise home service companies and small, regional installers. Creating a customer prediction pipeline comes with a tremendous computing cost, and running hundreds of millions of scenarios to build a model that works quickly can grow that cost exponentially–and unpredictably. How do you forecast the budget for the computing cost when the number of predictions made monthly for large and small organizations varies dramatically from month to month?

## The partnership

[![DataGrid mock](/images/stories/screenshots/datagrid-mock.png)](/images/stories/screenshots/datagrid-mock.png)

Realizing that an in-house build carried these risks, the SalesRabbit team shopped for existing best-in-class infrastructure that they could build on.

Faraday was the perfect fit for the team to implement customer predictions quickly, and at scale.

### Built-in consumer data for predicting customer behavior

Faraday has over [1,500+ attributes](/features/faraday-identity-graph) built into its platform, which mean all of this data was available on day one. Even better, it could be seamlessly combined with their existing data sets–no identity resolution required.

Employing this font of data on nearly every U.S. adult, SalesRabbit had everything they needed to understand the patterns and trends of their customers' ideal homeowners, regardless of which vertical their customers work in.

### Preserving existing data science resources and cutting product build time with Faraday recipes

Faraday’s plug-and-play [customer behavior predictions](/features/customer-behavior-predictions) meant there was no need to pull their data scientists off of other projects to build a customer prediction pipeline from scratch. SalesRabbit used pre-built prediction recipes as an accelerator to test and refine models before incorporating them into their product. This allowed them to shrink a multi-year roadmap to a matter of days, and free up their data science team to focus on other efforts.

### Peace of mind with transparency and scalability of cost

To the product team at SalesRabbit, building a customer prediction pipeline from scratch came with no small level of forecasting uncertainty. How much should they invest to launch? How much will it cost to scale up as their customer base grows?

With Faraday, they found pricing certainty, which allowed SalesRabbit to know exactly what to expect as they scale up.

## The results

[![DataGrid UI screenshot](/images/stories/screenshots/datagrid.jpg)](/images/stories/screenshots/datagrid.jpg)

To SalesRabbit, choosing to embed customer predictions into their software rather than build it in-house was a game-changer. Once SalesRabbit connected their data and created their initial predictive models, Faraday’s developer API made it easy to start embedding Faraday’s predictive features into DataGrid, and in a fraction of their original timeline.

Launched as “DataGrid AI,” SalesRabbit’s predictive lead scoring tool guides canvassing teams to best-fit territories, helps with route planning, and leaves reps with clear guidance on which households to prioritize when in the field. And because all of Faraday’s predictions are cached in advance, overall app performance improved once they were integrated.

Launched as “DataGrid AI,” SalesRabbit’s predictive lead scoring tool guides canvassing teams to best-fit territories, helps with route planning, and leaves reps with clear guidance on which households to prioritize when in the field.

How SalesRabbit uses Faraday to enhance their field sales app with data & AI

Customer experience is THE differentiator for a modern, successful business. Being able to provide a compelling, valuable CX can lead to a variety of benefits, including improved customer satisfaction, better retention, and improved business performance.

Today's businesses are well aware of the critical nature of the customer experience in their future. **[80% of companies](https://www.zippia.com/advice/customer-experience-statistics)** that focus on CX report seeing an increase in revenue, while brands that are more customer-centric and focused on the customer experience report **[60% higher profits](https://cxindex.com/company/blog/20-statistics-to-understand-the-rise-and-value-of-cx/)**. Companies that update their CX to be more data-driven find that by improving the customer experience they can **[reduce customer support costs by up to 33%](https://cxindex.com/company/blog/20-statistics-to-understand-the-rise-and-value-of-cx/)**. And companies are putting their money where it counts when it comes to the customer experience. Global CX spending is expected to reach **[$641 billion in 2024](https://cxindex.com/company/blog/20-statistics-to-understand-the-rise-and-value-of-cx/)**.

It's not just businesses that want a quality, value-driven CX. Consumers are willing to pay good money when interacting with a high-level customer experience, so long as they get something in return for it. For instance, consumers are willing to pay a **[16% price premium](https://cxindex.com/company/blog/20-statistics-to-understand-the-rise-and-value-of-cx/)** in return for a superior customer experience. Data shows that customer spending increases by up to 140% following a positive customer experience.

And don't kid yourselves, today's consumers are more empowered than ever before and will bolt for another brand if they don't have a good customer experience. **[49% of customers](<https://chainstoreage.com/bad-customer-service-puts-brands-risk#:~:text=Roughly%20half%20(49%25)%20of,24%2F7%20customer%20service%20support>)** who left a brand that they are loyal to say it's due to poor CX.

But for brands to be successful in executing a superior CX, they must come to terms with the fact that the foundation of a value-driven customer experience is quality, accurate customer data. The reality is a data-driven customer experience is a better CX.

And while up to now the CX was predominantly fed with more traditional "observable" data, including first-party data like pages visited, clicks, adds to cart, and time on site, for the customer experience to proactively understand customer behaviors it must deploy predictive AI algorithms to deliver to consumers what they desire before they have had a chance to act on it.

## What is predictive AI and generative AI?

The two primary AI models being deployed today in the enterprise business world are predictive AI and generative AI.

Predictive AI uses historical data to find patterns and anticipate future potential outcomes, while generative AI is more focused on the creation of new content assets, like images, text, and video from that data.

With predictive AI, businesses can better understand what their customers need, and what they care about. In this, businesses can finally move beyond making best guesses when detecting churn, scoring leads, recommending next best actions, forecasting spend, and making smarter decisions based on solid predictive data.

These two approaches to AI can be complementary if their algorithms are combined with the right software infrastructure that addresses data-oriented challenges like data quality, unified data integration, data integrity, identity resolution, and global data privacy compliance.

## To be proactive, not reactive

Making decisions from aging data is always going to put a business on the back foot. They will be forever reacting to something that happened in the past. But if an organization can predict future outcomes, it can be prepared ahead of time for the actions of customers, market fluctuations, and even challenging economic headwinds. Predictive AI delivers the knowledge of the possibilities of what might occur and allows a brand to prepare a response before it happens.

To do this companies must invest in the right AI technology infrastructure that will give them robust, dynamic predictive data to feed their CX systems. **[19 out of every 20 CX leaders](https://www.cxtoday.com/data-analytics/96-cx-leaders-are-expanding-their-customer-data-strategies/)** have invested in or plan to invest in data integration, data integrity, or data enrichment technologies.

**[Faraday](https://faraday.ai/)** provides the data and the predictive AI modeling infrastructure (IaaS) that enables organizations to iterate on their business challenges, instead of their technology infrastructure. With our platform, brands, and the software companies who support them, can now develop end-to-end customer experiences, at scale, driven by predictive insights.

In the end, a data-driven CX is a better CX. Faraday is the only predictive AI platform, both for **[software teams](https://faraday.ai/solutions/powered-by-faraday)** and **[brands](https://faraday.ai/features/customer-behavior-predictions)**, dedicated to predicting customer behavior. Everything you need to make customer experiences predictive is built into our software.

## Conclusion

How important is the customer experience to the modern business? According to **[Gartner](https://www.gartner.com/en/doc/3874972-realizing-the-benefits-of-superior-customer-experience-a-gartner-trend-insight-report)**, 80% of organizations expect to compete mainly based on CX. **[90% of businesses](https://cxindex.com/company/blog/20-statistics-to-understand-the-rise-and-value-of-cx/)**, regardless of their vertical market, have said they have made CX their primary focus.

For companies to realize this potential, they must have quality, accurate data to feed the CX system. By only using observed data companies remain inherently reactive, not proactive. To be proactive in serving their customer's needs, brands must deploy predictive AI software technology to deliver the dynamic, enhanced data needed to power the CX.

Through predictive analytics, brands can deliver a superior omnichannel customer experience, anticipate future trends, identify customers' behaviors and future needs, and proactively prevent customer churn.

To deliver the best customer experience, businesses must utilize predictive AI to anticipate and meet customer needs proactively, leveraging quality, accurate data to enhance customer satisfaction and drive business performance.

All posts