What Goes Into A Predictive Lead Scoring Model?

What’s the most important attribute to consider when building a lead scoring model? Well, that depends on who you ask. Every demand gen leader has a different model for scoring their leads, with values and attributes often based on nothing more than gut instinct.

With ineffective, guesswork-led models running rampant across the industry, it’s no surprise that most people consider lead scoring to be a joke.

Predictive Lead Scoring Is Changing The Game

Predictive lead scoring is lead scoring that actually works. No guesswork, gut instinct or dumb luck required. Instead, advanced machine learning algorithms do the work for you. No, really.

Building A Predictive Lead Scoring Model For Your Company

If you want predictive scores that are statistically proven to be accurate, it’s time to dump the useless, outdated models and upgrade to an AI-led predictive lead scoring model. Here’s how Infer builds industry-leading models:

Start With Your Existing Data

Your historical data is littered with valuable insights. So, that is where predictive lead scoring starts. Infer builds each custom model by first connecting with the existing data from your CRM or marketing automation system.

Add Thousands Of Externals Signals

Then, new data is added. Infer matches each of your records against our proprietary dataset to expand the information base. We identify thousands of new signals for each record including social media presence, website technology and more.

Determine Which Signals Are Predictive

Not every signal we unearth will be truly predictive. We use machine learning algorithms to search the data for deep patterns, highlighting the predictive signals and discarding signals that could lead our scoring off track.

Create The Optimal Formula

Once predictive patterns have been identified, we create a custom formula for your organization. This formula is then used to automatically score any leads that enter your system and assign a predictive value to each one.

Test The Accuracy Of The Model

Infer promises statistically-accurate results. And we mean it. We test and retest the accuracy of your predictive model to ensure it delivers results your sales and marketing teams can rely on.

Push It Live Into Production

We will have you up-and-running with Infer in days, not months. And since Infer seamlessly integrates with your sales and marketing stack, you can start focusing on your best leads immediately.

Measure The Results

You will see real results, right away. Don’t believe us? Just ask Suresh.

“We have one set of leads that converts at 4x the baseline, and one set that’ll never convert. Infer helps us tell the difference between the two.” – Suresh Khanna, Chief Revenue Officer AdRoll

Or anyone else from our happy community.

Want to learn more about predictive lead models? We made this eBook just for you:

What Is Predictive Lead Scoring And What Can It Do For Me?

We hear it every day:

“I need more pipeline.”
“My sales reps don’t trust our manual lead scoring system.”
“My team is overwhelmed and doesn’t know which leads to focus on.”
“These leads are garbage.”

The constant pipeline battles wear down even the best demand generation leaders. Yes, even those that consider themselves well-versed in traditional lead scoring methods still struggle to identify and prioritize truly good leads. It’s enough to drive any marketer mad.

Fortunately, we have the magical (okay, it’s AI-driven) solution.

Intelligent Predictive Lead Scoring Is Right Here To Save You

When it comes to making critical sales and marketing decisions, traditional lead scoring is like stopping at the gas station to ask for directions, while predictive lead scoring is Google Maps – the instant clarity of knowing where you are and what to do next.

No more long, frustrating meetings debating what factors should be considered in scoring a lead or how much weight each factor should carry. No more shifting the blame between sales and marketing when you aren’t seeing the results you want.

Predictive lead scoring takes the guesswork out of demand generation and delivers lead scores every member of your sales and marketing teams can rely on.

How Does Predictive Lead Scoring Work?

We’re glad you asked. At first glance, a tool like Infer seems too good to be true. But go ahead and pinch yourself because Infer really is all that and a bag of chips. (Disclaimer: chips not included.)

Infer connects with existing data from your CRM or marketing automation systems, whether that be HubSpot, Marketo, Salesforce, etc. After connecting, our tool then matches each of your records against our proprietary dataset to identify thousands of signals, including social media presence, website technology and more.

Next, Infer uses advanced machine learning algorithms to discover the deep patterns within your marketing data that have been escaping your notice. Infer then analyzes this data and identifies which prospects are most likely to become customers, assigning a predictive value to each one.

P.S. Our predictive scores are statistically proven to be accurate. Let’s see a manual lead scoring system do that!

Sounds Great. But What Difference Will This Make For Me Really?

The better question: What aspects of my demand generation won’t see improvement? The answer: None.

Everyone on your demand generation team will reap the benefits of Infer’s predictive lead scoring system. And we guarantee you’ll get results quickly.

Your marketing team will enjoy increased effectiveness thanks to their ability to better understand and target leads. Additional benefits include a lower cost per good lead, faster pace of innovation and increased lifetime value of customers.

Your sales team will also enjoy increased effectiveness thanks to Infer’s ability to score, filter and prioritize leads instantly. You can expect to see an increase in average deal size, efficient and personal nurture follow-ups as well as a higher conversion rate.

Big growth plans? Predictive lead scoring will ease those growing pains. Infer’s intelligent model scales from thousands to millions of leads without breaking a sweat.

Plus, did we mention more revenue?

Getting Past The Third-Party Data Hype: Why You Need First-Party Data

Countless marketers, frustrated by what seems like an onslaught of never-ending demand generation problems, are turning to data for a solution. This, in and of itself, is a great idea. Incorporating intent data into your sales and marketing decisions enables you to build informed and strategic campaigns.

Intent data is the online behavior of consumers that links them to a certain topic and it is often a powerful predictor of their future actions. By highlighting the leads and accounts that are actively researching a particular industry, it effectively shows companies which accounts are ready to buy. However, not all intent data is created equal.

There are two main types – first-party data and third-party data. First-party intent data, or internal intent data, is the activity that you’re capturing on your website, inside your marketing automation system or through application logs. It is data that you already own.

Third-party intent data, or external intent data, is the data collected by networks of B2B publishers. This data is available for purchase through third-party providers, and is often a favorite of sales and marketing teams that think it will be a magical solution for their demand generation woes.

What many marketers are failing to realize is that, while using third-party data can be helpful, relying solely on it will set your campaigns up for failure. If you are not also pulling insights from first-party data, you are missing out on the best, most highly relevant information available to you. Not only is third-party data less relevant to your customer base, it can actually mislead your marketing efforts if it’s all you use to make marketing decisions.

Don’t let the allure of “big data” limit your campaigns. We explore the many reasons why relying solely on third-party data is not all it’s cracked up to be:

Data Decay

Today’s consumers move quickly. Their interests and objectives change based on what their most pressing issue is at any given time. The problem is that the third-party “intent data” that vendors are selling doesn’t actually keep up with consumers’ changing intents.

The shelf life for third-party data is alarmingly small. By the time your company buys the intent data, analyzes the trends and builds the campaigns – those potential customers have already moved on. So if this is all you’re using to plan campaigns, you could end up taking actions based on data that is no longer relevant.

To be able to quickly and relevantly address consumers’ needs and interests, you need to look at first-party intent data.

Broad Identification

When you purchase third-party intent data, you are not given the person or persons who have shown interest in your industry. Instead, you are given a broad account or company name. This leaves you scrambling to determine who you should reach out to, and what their specific interest in your product or service is.

To be effective in your outreach efforts, you need to identify a specific lead name. This can easily be achieved with first-party intent data tools such as Infer.

No Fit Scoring

Just because someone is looking into your industry it doesn’t mean they are necessarily a good fit for your specific product or service, nor does it mean they have any purchase intent. Third-party data offers no insight into whether or not any given lead is actually a good fit for your company, leaving you to waste valuable resources chasing down low-quality leads.

To truly know your customer, you need to take your data-driven marketing to the next level with first-party data and fit scoring. For example, Infer uses machine learning to create a predictive model of your ideal customer profile and instantly score each new lead to immediately determine how likely they are to buy.

Black Box Methods

Third-party data vendors often strive for quantity, often to the detriment of quality. And, in a quest for scale, it is easy for marketers to overlook the inaccuracies. However, if the third-party intent data you purchase is stuffed with filler accounts, you will have wasted your investment and your campaigns will not succeed.

It is difficult to know if you can trust the data given to you by large third-party data vendors, especially as they are rarely transparent about their collection methods and the information is mostly anonymous. This means that unless you are willing to purchase another tool to score the third-party data leads you receive and identify the ones that are worth pursuing, your sales reps will struggle with outbound and your progress will be slow.

It is better for both your team’s sanity and your bottom line to use one tool that can analyze both third-party and the always accurate first-party data you already own, surface leads, and then instantly score them with transparent rankings. That way, you will never have to question if your data is dependable.

How To Make The Most Of Your Data

Don’t get us wrong, third-party data can be a valuable asset but only when it is used in conjunction with accurate and personalized first-party data. Knowing that someone showed interest in a certain article can be useful, but it is not conclusive enough to actually know if or when that account will convert. You need to dig deeper.

By using first-party data to examine the actions that buyer takes on your site and narrow the lead down from an account to an individual, you can accurately assess the value of each and every lead. This information will help you supercharge your pipeline and have more informed sales conversations with potential customers.

Want to learn more about how first-party data can transform your sales and marketing?

Oh How the Tables Have Turned: What B2C Marketers Can Learn From B2B

This co-authored byline was originally published on CMSWire by Sean Zinsmeister, Vice President of Product Marketing at Infer, and Adrian Chang, Director of Customer Marketing at Oracle Marketing Cloud.

Business-to-business (B2B) sales and marketing are entirely different from business-to-consumer (B2C) tactics — or that’s the general assumption.

One is relatively low volume and high budget, with lengthy, consultative selling processes and lots of personal relationships at play, while the other usually means huge volumes and low price points, with fast, direct sales processes.

B2C relies on consumers, transaction events, impulse buys and coupons. B2B is all about prospects, “journeys” and reaching people through content in context.

But more and more, we’re seeing these two universes converge.

Although we often perceive B2B companies as one step behind their B2C counterparts when it comes to adopting the latest sales, marketing and advertising techniques, that is quickly changing. B2C marketers have naturally excelled at bringing as many people as possible into the top of the funnel, but B2B companies have perfected the use of intelligence and personalization to move multi-stakeholder buying committees through non-linear customer journeys.

Today, many B2C businesses are finding that at a certain threshold, the price points and buying cycles of a considered consumer purchase are beginning to look almost like a B2B deal.

In this environment, B2C companies are turning the tables and adopting the latest B2B marketing approaches and technologies, to great benefit.

4 Ways B2C Can Benefit From a B2B Martech Stack

1. Adopt Hyper-Segmentation for a Better Understanding of the Ideal Customer

Most B2C technologies focus on point-in-time transactions across a massive prospect universe. But they aren’t great at capturing detailed profile and activity data on each of those customers.

While they ingest plenty of basic demographic signals from search patterns and lifestyle purchase history, much of that information is anonymous. Anonymization makes it challenging for marketers to understand and analyze the data, since they can’t link consumer patterns with internal data from custom forms or other identifiable insights.

B2B systems, on the other hand, track a complex web of detailed attributes and match activities to purchases across each customer at every stage in the buying journey. They are well-equipped to help marketers segment their total addressable market deeply and precisely across all types of customer signals and behaviors.

A consumer tech or financial services company might want to segment its customers based on their travel profiles (many of which overlap), or a home rental agency might consider not only what types of properties a consumer is interested in, but also what city they live in or which industry they work for.

B2B predictive platforms can deliver valuable insight into these situations by producing easy-to-understand customer predictions that inform the segmentation process so marketers can determine which profiles should receive high-touch, personalized outreach verses low-cost, low-touch campaigns.

2. Automate Customer Journeys to Reach Buyers at Every Stage

As opposed to the point-and-click purchases of the consumer world, in the B2B model, buying processes tend to be more intricate and involve multiple people. B2B marketing automation platforms support a more complicated funnel that helps marketers plan, prioritize and execute campaigns knowing that certain interactions are bigger influencers to winning a customer than others.

Think about the evolution fast-food chains like McDonald’s have undergone over the past few decades. Rather than organizing their prep staff to assemble Big Macs and other menu items based on a rough estimate of what most people usually order and when lunch and dinner rushes occur, most of these businesses have switched to made-for-you systems in which food is assembled as it is ordered for better quality.

B2C companies are poised for a similar transition away from one-size-fits-all marketing as they adopt B2B martech systems to better orchestrate end-to-end, dynamic automation across their many channels and segments. Predictive technologies inform these tailored workflows by showing which assets, campaigns or channels reach a company’s best-fit customers or profiles.

3. Increase Personalization to Boost Advertising ROI

While B2C marketing can rely heavily on brand advertising to push decision makers over the line, in B2B, brand awareness might help get a vendor into consideration, but won’t close the deal alone. Because B2B systems collect all kinds of customer intelligence, they do an excellent job of helping marketers go beyond general brand-building through profiling, predictive modeling and personalization at scale.

B2B martech stacks bring better data and richer profiles to the task of executing hyper-segmentation strategies, workflow automation and dynamic content — all of which combine to deliver both higher short-term conversions and greater customer loyalty over the long-term.

For example, Dell leverages the same marketing automation platform to power both the B2C and B2B sides of its business. As a result, its marketers have a better understanding of their ideal customers and can draw these prospects into more intimate, personal conversations with the brand.

4. Use Full-Funnel Insights to Drive Engagement

With all of the apps launched during the recent martech explosion, integration causes major headaches, leaving dangerous blind spots between some go-to-market workflows. It doesn’t help that many B2C interactions happen offline, compounding the challenge of collecting and interpreting data across the entire customer funnel.

The good news is B2B systems generally “play together” better than B2C tools because they offer more open APIs, and were made from the get-go to deliver end-to-end insights to a sales audience.

As opposed to fueling the sales and marketing divide by keeping people siloed in different B2C systems, seamlessly integrated B2B platforms can bring all of the functions together. For instance, predictive platforms can learn from historical sales data to model what a good customer looks like, and apply that intelligence to the top of the funnel, which then cascades through marketing automation programs and all the way down the funnel to help customer success teams load-balance customers.

B2C companies using this type of intelligence throughout their business will grow consumers’ loyalty, increase retention and find new ways to delight people.

It’s Time for B2C to Flip the Funnel

Over five years ago, Joseph Jaffe started a B2C conversation about “flipping the funnel” (a term which has since been hijacked and distorted by the ABM community), yet few B2C companies have fully adopted the mindset. In comparison, B2B marketers inherently follow Jaffe’s approach, because it is built into the very way they sell.

As more noise infiltrates a consumer’s purchase decision, it’s becoming harder for B2C marketers to drive conversions with mass-messaging. It is time for B2C to take a cue from B2B and finally double down on one-to-one relationships.

AI 101, Part II: How to Deal with Data Preparation

This article was originally published on MarTech Series by Sean Zinsmeister, Vice President of Product Marketing at Infer.

My first post in this series covered what marketers and sales leaders need to know about the four main phases of building predictive models. The second of these steps – data preparation – tends to be the least understood part of AI and predictive analytics in marketing. In this next post, I’ll dig deeper into key considerations surrounding this process, namely related to data volume and data quality. When my company introduces our predictive platform to companies, two of the biggest concerns we hear are: (1) Do I have enough data? and (2) Is my data “clean” enough?

Read This: AI 101, Part I: What You Need to Know about Predictive Models


There’s a rule of thumb for how much data you need in order to be successful with a predictive model, and the most important number is the amount of positive signals or “good” examples there are in your data set. In the case of historical customer data for lead or account scoring, this would be how many total opportunities or closed/won deals you have in your CRM database.

Of course, these positive signals exist among other negatives. Make sure your positive is defined as a relatively significant achievement in the pipeline. For example, the creation of an opportunity is a meaningless milestone if it happens for every single free trial that comes in. Instead, consider going further down the funnel to find a tougher hurdle that really points to lead quality. infographic clean

Predictions will be most accurate when you have around 400 to 500 of these positive results. In that range, they can be randomized and split into two proportions (60% and 40%) for model comparison. If you have fewer than a hundred examples to test your model over, your results won’t be quite as precise as you might want (until you add more data over time and refresh the model).infographic via Sean Zinsmeister


The truth is that no business has perfect quality, complete data, but that’s okay. Modern data preparation techniques are built to work around that very problem, so there’s no need to delay AI initiatives while you wade through cumbersome data clean-up projects. If you do, you’ll just leave revenue opportunities on the table. By matching whatever limited lead data you have with hundreds of external signals from the web, predictive platforms like Infer can build a complete picture of each prospect or customer. In fact, our algorithms can produce lead scores with nothing more than a company name or an email address. That’s thanks to advanced data science approaches like Natural Language Processing (NLP), which can bridge gaps in your data by looking for patterns in the web crawls, performing title normalization and doing spam analysis on form input.


Anyone who has sold into IT or the sales and marketing industry knows that job titles are all over the place (or sometimes not included in the data at all). Title normalization techniques tend to be especially important for lead fit models because you need to know that “Marketing Director” might be equivalent to “demand gen lead,” or that “IBM” and “International Business Machines” are the same company. NLP essentially splits out each word that exists across all of your records and uses an algorithm to assess related patterns and find the words that show up most often in positive outcomes for a particular data set.


Another sophisticated feature to look for is spam analysis – something that’s often used in consumer search algorithms like Google. By analyzing the number of capitalized characters and key input for a name, company, title or email, you can assess the likelihood that each data point is a legitimate input. For example, the way a person’s fingers traveled across the keyboard (i.e. the number of row switches, etc.) often indicates whether their entry is legitimate. An email like asdf@ggg.com doesn’t travel very far and is probably not a real address. Machine-learning can perform these checks on every single record, regardless of whether or not it matches a known website domain.

As you can imagine, NLP alone can help you immediately improve your data hygiene. That’s why, instead of doing months of data cleansing first in hopes of being able to get better intelligence, later on, it’s smarter to get your predictive and AI initiatives started now, with the data you have. There’s no sense in spending time and money augmenting fields and cleaning up data that isn’t helpful for your models anyway. Rather, use machine-learning to figure out what your most important data points actually are, and then focus your data cleanup efforts there as needed.

It’s so important to understand common data science methodologies like these as you move forward, even if you never intend to work with the algorithms yourself. This knowledge will help you spot any flaws, unrealistic expectations, assumptions, and missing pieces in predictive and AI solutions so that you can thoughtfully evaluate them. In my next post, I’ll expand further on basic model types and more problems sales and marketing teams can solve with data.

Getting Started with Machine Learning: 3 Things Marketers Need to Know

This article was originally published on CMSWire by Sean Zinsmeister, Vice President of Product Marketing at Infer.

The buzz surrounding machine learning and artificial intelligence (AI) in the consumer world has rapidly bled over into the enterprise.

Much of this hype stems from the new consumer trends that hint at the possibilities of AI, such as self-driving cars and intelligent voice-first products like Amazon’s Alexa and Apple’s Siri.

At the same time, mainstream cloud adoption and ever-increasing computing power in the form of new solutions like Google Spanner are accelerating the development, accuracy and speed of AI’s underlying foundations, from data availability and spam detection, to machine learning, predictive analytics and natural language processing.

So it should come as no surprise that sales and marketing leaders are questioning what all this means for their departments and companies.

At a basic level, AI is about replacing human function with computers. Without using machine learning, people simply couldn’t sort through the plethora of data that’s out there without making costly errors.

Today’s machines have proven they can successfully process, understand, translate and interpret that data — parsing its meaning into visual and actionable outputs. That said, in AI’s current state, what we’re talking about is computers taking over very simple, mundane tasks as opposed to complicated, nuanced actions.

Machine Learning Already Exists in the Enterprise

Several machine learning and AI related breakthroughs are already having an impact on business outcomes, most notably predictive analytics, chatbots and natural language processing (NLP).

For example, predictive analytics is helping enterprises process their CRM and marketing automation system data, combine it with external signals from across the web, and determine where their highest revenue potential lies so they can double-down on the best bets.

Chatbots are more efficiently routing people to services and quickly answering customer questions 24/7, as well as creating new ways for business people to interact with their company’s data.

B2B enterprises and consumer companies alike are experimenting with these conversational user interfaces, which use NLP to recognize speech patterns, essentially mirror a human, and (hopefully) create better customer experiences. But using these technologies carries risks, and businesses in the US can learn plenty of lessons from mobile-first companies like WeChat out of China and India.

But in spite of the success stories, most companies are still wondering if they’re really ready for machine learning and AI. The truth is nearly any enterprise can benefit from infusing more data into its operations, especially if they watch out for common pitfalls.

3 Requirements for Successful AI

1. Start with a business problem

To start with, ask yourself questions like: How well is my business utilizing data to make decisions? Do we have a problem that better data and accurate insights can solve?

Rather than just inventing a problem to have, look towards the business practitioners who are succeeding with machine learning and AI today. Find out what solutions they’re using, and you’ll uncover communities of innovators that have similar market problems.

The real differentiation among today’s emerging vendors is in how well they address customers’ most important business needs, so choose a platform that’s built for your specific use case and will be able to scale and adapt as your business grows.

2. Optimize your technology stack

Few companies embarking on the AI journey recognize the importance of setting up their systems for success — both from a cost and performance perspective. Rather than thinking strategically about which tools they need in their technology stack, teams are too often distracted by shiny new objects and go on a spending spree only to find themselves with a bunch of fragmented tools (many of which end up as shelfware).

The truth is less can sometimes be more, especially when it comes to a scalable stack that’s architected to leverage AI for greater efficiency and effectiveness.

Another consideration is data storage. It can get very expensive, especially when you’re trying to pull in data to, say, a CRM system from a variety of different sources across and beyond your business. Some vendor’s data fees are in line with hardware storage costs from the ’90s, and can even cost more than your annual license.

Of course, you can’t succeed at AI without data, and even small companies have customer data spread across applications. Unfortunately, most of today’s automation systems weren’t architected to scale to the volumes of data that are now typical for midmarket, high growth companies. Get creative from the get-go and find strategic ways to more affordably connect all of the data you’ll need for your AI initiatives.

3. Make systems smarter through integrations

One of the best ways to address the stack efficiency challenge is to find an AI platform with an open architecture that can serve as your company’s system of intelligence.

When it comes to marketing technology in particular, recent studies suggest only 21 percent of companies use a single-vendor suite, while nearly half go with a best-of-breed approach. The integration of these multi-vendor stacks is key to their ability to deliver value.

Your system of intelligence should seamlessly connect with things like your system of record (CRM) and system of engagement (marketing automation), as well as any other business applications the company uses — and then leverage that data to make accurate and actionable predictions.

The result is you’ll be able to quickly and easily infuse predictive analytics into your business’ decision making without disrupting current workflows or adding more complexity to daily activities. Once you’ve gathered all of your data in one place, you can increase the intelligence of all your systems, thanks to the smart outputs you’ll get from your machine learning and predictive initiatives.

The Road Towards Self-Driving Business

Once enterprises address these considerations, they are well on their way to reaping the rewards of AI.

Yet these technologies will never be a silver bullet for every business woe. The industry is still at the early stages of its journey towards disruptive machine and data-guided business models.

We still has a long way to go when it comes to shifting routine enterprise work away from humans and towards machines. And we will always need the expertise and intuition of people to make AI successful, but I anticipate a bright future ahead.

In the next five to 10 years, we’ll see the business world’s version of self-driving cars (‘self-driving’ enterprise software) make headway in functions like sales, marketing, finance and human resources.

As computers successfully transform unstructured data from disparate systems into actionable intelligence, AI will upend the manual, rules-based workflows of traditional systems of record — causing vendors and early adopters alike to re-imagine what true automation should look like.


AI 101, Part I: What You Need to Know about Predictive Models

This article was originally published on MarTech Series by Sean Zinsmeister, Vice President of Product Marketing at Infer.


While predictive analytics and AI are big topics in the sales and marketing profession these days, it can feel daunting when you’re trying to figure out how to get started with these data-dependent solutions. Although most marketers probably won’t actually be building any data models themselves, it’s vital that the next wave of go-to-market professionals develop a solid understanding of how to solve business problems using data. In this series of articles, I’ll break down key concepts surrounding these technologies piece-by-piece, and provide a helpful look under the hood of predictive modeling.

Business folks who are ready to get their feet wet with AI should first zoom out and learn the basics of predictive modeling – one of the underlying technologies that’s required in order to effectively replace human functions with machines. AI solutions use this advanced data science to process, understand, translate and interpret all the data that’s out there, and parse its meaning into visual and actionable outputs.

Let’s explore each of the four main phases of building predictive models:


When it comes to building a predictive model, the first step is to gather all of your inputs or data sources. There’s no question that sales and marketing teams are acquiring a plethora of data. In many cases, however, marketers are collecting data that only they care about, and it might not be valuable, insightful or actionable for sales (and vice versa). Regardless, given the mass adoption of passive marketing channels, low-barrier free trials, etc., most businesses are gathering a lot of data about their prospects at scale – much of which can be used to inform smarter predictive models.


Before digging into all this data, it’s important to first step back and figure out what business problem you are trying to solve with AI, which will help you prioritize your data preparation tasks. The reality is that it’s very common for data to be incomplete and dirty (there’s no getting around the human error that comes with data entry), so data preparation is crucial to the future of AI. Your data should be properly cleaned up, if you will, so that you can normalize typical errors during the data acquisition phase, and ultimately produce a sound model. Only then, will predictive analytics answer your specific questions and drive the actions you want. Some common ways to prepare your data include enrichment (bringing in external signals to complement current records), spam analysis and title normalization – stay tuned for more on these techniques in my next post.


Once you understand the machine-learning problem you want to solve for, the next step to building a model is to employ data science methodologies like classification or regression. Classification is also known as probability estimation, and it is used to predict which of a small set of classes each individual belongs to. For instance, you might ask “Among customers of Company X, who is most likely to respond to this offer?” There then would be two classes: “Will Not Respond” or Will Respond.”

On the other hand, regression (or value estimation) is used predict the numerical value of some variable for each individual. Looking at historical data, you might produce a model that estimates a particular variable specific to each individual, such as “How much will this customer use this service?” Both of these techniques, and many others, can deliver model outputs that drive powerful AI and predictive analytics use cases in sales and marketing.


For example, sales teams can achieve major performance management improvements by using predictive models to improve the way they filter and prioritize both inbound leads and account-based outreach tactics. With score outputs that indicate which leads look most like the company’s ideal customer, sales can confidently focus their time on just those prospects that are likely to buy. In addition, teams can use AI to more thoughtfully route their leads – either to SDRs for outreach and development over time, to account executives for more aggressive follow-up, or to automated nurture programs – based on each lead’s potential value. Predictive behavior models can also alert sales when an old lead starts acting like a customer. By looking at engagement patterns in marketing automation and web analytics systems, you can determine when neglected leads are likely getting close to a conversion threshold. This helps reps find good qualified accounts and contacts that are “reawakening,” and then trigger data-driven workflows for more aggressive follow-up with just the right message at just the right time.

Another valuable use case for AI is to drive marketing efficiency. With the right customer intelligence, marketing teams can optimize conversions for the greatest possible funnel efficiency. And since predictive analytics outputs deliver immediate feedback on the quality of marketing campaigns, they can easily calculate key performance metrics in real-time rather than waiting for sales cycles to play out. Accurate predictions also add value when it comes to quantifying key marketing performance indicators like cost per good lead, average lead quality, pipeline-to-spend ratio, etc. By using these KPIs to look past traditional vanity metrics and identify top performing campaigns and content, marketers gain deeper insights into which programs attract the highest quality leads, drive larger deals, and accelerate deal velocity.

Each of these four steps to the predictive model build process is important to understand if you want your models to produce statistically accurate predictions, and these phases become increasingly mission-critical as AI takes over more and more everyday tasks from humans. No one wants to miss out on real revenue, just because AI made unnecessary mistakes when determining outbound prospecting lists or writing the content of sales outreach emails.

Hacking Content Marketing With Predictive Analytics

Content Summit is a 5-day virtual conference dedicated to sharing the most effective B2B content marketing strategies & tactics, which are delivered by a B2B marketing executive or thought leader. Our very own Sean Zinsmeister hosted a session on how content marketers can get the most out of their programs using predictive intelligence.

Watch the replay below:

Getting past the AI hype: How predictive analytics fuels conversion optimisation

This article was originally published on MarketingTech by Sean Zinsmeister, Vice President of Product Marketing at Infer.
These days, marketers can’t read about their profession without getting bombarded with wild claims about how AI is going to disrupt everything they do. And with the sales and marketing functions evolving so rapidly in recent years, marketers in particular must embrace an entrepreneurial spirit and constantly explore new technologies in order to give their team a competitive edge. That mindset shift, along with new consumer trends—such as self-driving cars and intelligent voice-first products like Amazon’s Alexa and Apple’s Siri—are bringing the possibilities of AI to the forefront of business-to-business marketing technology discussions.

But all of this begs the question, “Which AI claims are hype and which are reality?”

In order to know what a new technology like AI can bring to the table, it’s important to fully understand the problems you’re trying to solve. When it comes to the current state of AI solutions for marketing and sales, today’s reality is less futuristic robots or automating every single marketing workflow, and more about how data can answer one important baseline go-to-market question: who to sell and/or market to. There’s a wealth of intelligence that predictive analytics and machine learning bring to the task of answering these questions – and that’s the crux of where AI is delivering value today.

Sales performance management

Forward-thinking enterprise sales teams are saving tons of time by simply using predictive solutions to improve the way they filter and prioritise inbound leads. Companies with the “champagne” problem of an overwhelming volume of incoming prospects are using predictive analytics and AI to automatically research and qualify leads who looks like their company’s ideal customer. For example, Shoretel’s market development team found that predictive scores could tell them not just which prospects are the best fit, but also which ones are showing current buying behaviour. As a result, the telco leader’s sales reps can instantly understand who their best prospects are and determine where their time should be spent — insight which has resulted in 8X greater conversions. Now it takes just 12 calls to uncover one marketing-qualified lead (MQL) vs. the 100 calls it took before the company adopted predictive analytics.

Word on the Street: Predictive Advice from Infer Customers

Earlier this month, G2 Crowd sat down with Infer customers InsightSquared and Yesware to ask about how they use Infer’s predictive analytics and AI platform, what benefits they’ve seen, and their recommendations for other early adopters.

First up was Matthew Bellows, CEO of Yesware, a long-time customer of Infer. During the interview, he shared some helpful tidbits about how his sales team uses predictive to automatically qualify the best prospects from their high-volume lead flow, focus on the right accounts, and increase overall sales velocity and performance:

For more helpful advice from Yesware, check out this recent webinar with their director of demand gen, and learn how she was able to build a revenue-centric funnel with Infer. As a result, the company eliminated wasted sales efforts and won more deals.

G2 Crowd also chatted with Joe Chernov, VP of Marketing at InsightSquared, about how the company’s go-to-market teams use Infer to build alignment around the best accounts, and drive engagement as part of their account-based sales and marketing strategies.

Read our full snapshot to learn more about how InsightSquared is using Infer Predictive Scoring to make their marketing and sales machine much more efficient by identifying high-value leads and accounts, increasing conversions from top leads, and reducing overall cost-per-lead.

And for even more “word on the street” comments from other customers, browse Infer’s many reviews at G2 Crowd.