Unboxing the Predictive Intelligence Blackbox: 3 Types of Data You Need

What if you could predict a customer’s next move before they even decide to make it?

In the past, this question was purely hypothetical. But, thanks to technological advancements and mass data collection, predicting customer behavior has become a reality—and it’s changed the face of sales and marketing forever.

One of the benefits of a large volume of information? Predictive intelligence.

What is Predictive Intelligence?

Predictive intelligence falls under the artificial intelligence umbrella. It is composed of statistics, data mining, algorithms, and machine learning to identify trends and behavior patterns.

When applied to sales and marketing, predictive analytics forecasts companies most likely to buy or take future action relevant to your business.

How exactly does that work? Well, basically, a purchase tends to happen at the confluence of three different types of predictive data: Fit, Intent, and Opportunity data. More on that later.

In the case of sales, marketing, and recruiting, this means using historical data to identify potential customers at the moment they are in need of your product or service.

Why is Predictive Intelligence Important?

It’s no secret, personalization is essential to modern marketing success. In fact, 77% of consumers have chosen, recommended or paid more for a brand that offers a personalized experience.

Simply put, predictive intelligence allows marketers to offer personalized marketing. By using past behavior to predict future behavior, marketers can personalize their campaigns not only to certain customer segments—but to each individual customer.

The best part? Predictive intelligence allows this type of analysis and to happen instantly, in a way that human analysis can’t compete with. Thus, predictive intelligence enables marketers to make better decisions, faster.

Predictive Intelligence Stats

  • Predictive intelligence shows a 40.38% increase in revenue after 36 months of implementation.
  • 34% of purchases are influenced by predictive intelligence recommendations.
  • Website sessions that are influenced by predictive intelligence achieve a 22.66% increase in conversion rates.

Source: Salesforce

Convincing, right? But, because it’s still relatively new, predictive intelligence can be intimidating. First, we’re talking about a LOT of data. How can we figure out which data actually matters? Which data points⁠—individually, or together in a “secret sauce” ⁠—predict buying behavior in B2B customers?

That’s why we’ve decided to take a deep dive into the concept, simplifying the various ways salespeople and marketers can use predictive intelligence to improve their campaigns. 

Furthermore, we surveyed 200+ sales and marketing professionals about 78 predictive data points (and “secret sauce” combinations of data points). We asked, “Which data points predict higher conversion rates and more sales?” 

Read on to learn more about predictive intelligence and the results of our survey. We’ll conclude with some ways that you can apply your newfound predictive knowledge to increase leads and sales!

Predictive Intelligence Needs 3 Types of Data

The likelihood of purchasing lies in the middle of a Venn diagram consisting of three buckets: Fit, Opportunity, and Intent.

Behavioral information is only predictive when combined with well-defined firmographic data and demographic criteria that fit the ideal customer profile.

Chart describing difference between fit, intent, and opportunity data.

Type #1: Fit Data

Fit data is the basics: the right contact at the right company. An identified company profile is the primary basic requirement of any kind of scoring or predictive analysis⁠—basic physiological data upon which Maslow’s Hierarchy of Needs for sales and marketing, as it were. 

If the company itself is not a great fit, all other information, no matter how effective at prediction, has no value.

Fit data includes basic demographic, firmographic, and technographic information at the account and contact level. These include data points such as:

  • Industry
  • Job function
  • Department budget
  • Technology stack
  • Gender
  • Location
  • Use of agencies or contract services

What’s the most predictive Fit data point? Job title.

Over 85% of respondents of our survey said job title is effective or very effective at predicting a prospect’s likelihood of making a purchase.

This is because a job title is a basic, fundamental part of the ideal customer profile: Even if every other piece of the puzzle is perfect—the right industry, the right time, a perfect pitch. If the prospect is in the wrong department, or doesn’t have purchasing power…nothing else matters.

Without a customer profile that touches on these points, along with fit criteria at the company level, a sales team is likely to spin a lot of cycles on deals that don’t end up closing.

Type #2: Opportunity Data

Opportunity insights are defined by favorable conditions. Sometimes a prospect stumbles upon a solution at exactly the moment they need it…but luck has never been a great sales strategy.

That’s why Opportunity or “trigger” information becomes a predictive piece of the purchasing puzzle—when it’s layered on top of Fit and Intent data. These are the data points that indicate that conditions are favorable for a change.

Action-based Opportunity signals show favorable conditions for purchase. These types of data points include:

  • Leadership change
  • Funding
  • Pain point
  • Hiring plans, promotions, layoffs
  • Company events
  • Merger
  • FCC fine

So what’s the most effective Opportunity data point?
84% of our survey respondents said Requests for Proposal (RFPs) and Projects/Purchase initiatives are effective or very effective at predicting a prospect’s likelihood of predicting a purchase.

If a sales team can get a seat at the table when a company is requesting proposals, or during the project planning phase, it stands to reason that they will select those pitches over those who don’t get in the door in time.

The least predictive Opportunity data point was Company Awards. (While this might not predict purchase behavior in itself, one of ZoomInfo’s top-performing SDRs would be quick to point out that a company event like an award is a great excuse for outreach.)

Overall, just 29% of respondents use Fit AND Opportunity data.

Type #3: Intent Data

The third layer of data that makes up predictive intelligence is Intent data: information on implicit behavior.

With a foundation of basic demographic and firmographic details in place, and favorable conditions present, Intent data is the lynchpin for predicting success.

Intent data is the behavioral activity that links target buyers and accounts to a solution, solution category, or related topics. This includes:

  • Time on website
  • Form-fills / downloaded your content
  • Comparing your product with a competitor’s
  • Lead source
  • Social media follows
  • Commented or ‘liked’ your content
  • Spikes in content on a given topic

What’s the most effective Intent data point? Companies comparing the products of other vendors in your category

In fact, seven of the top eight most effective Intent data points all involved competitor research and comparison. If a company is comparing vendors in your space—to each other or to your solution—they’re not far from making a purchase. And at that point in the buyer’s journey, the choices have been narrowed down to a small handful.

The information collected by marketing automation systems for an organization is one level of Intent data, but many organizations expand that layer to vast networks of sites and partners that gather intent data from numerous places.

Intent data offers something Fit data cannot: It signals interest, demand, or urgency related to a particular topic or need.

Overall, just 15% of respondents use Fit AND Opportunity AND Intent data.

The sequential, piecemeal nature of the Fit + Opportunity + Intent scoring combination is not always well understood by sales and marketing professionals.

Predictive Analytics in Action

Now that we’ve learned the fundamentals of predictive intelligence, let’s explore some application using a fictional business as an example:


Fit: Kelly is a sales development rep at a company that sells applicant-tracking software. Her best-fit clients are enterprise-size companies in the retail industry—which is always hiring due to a high rate of turnover.

Opportunity: Kelly learns that one of her target accounts is opening 23 new stores in her territory…and Christmas season is just 3 months away.

Intent: Kelly can see that someone from that same account has visited her company’s website several times, downloaded a datasheet of the integration capabilities of her product, and signed up for a weekly recruitment-tech news round-up. Through third-party intent data, Kelly can see a recent spike in activity and interest in content related to applicant tracking systems and recruiting.

… NOW, there’s a very good chance that prospects will be happy to take Kelly’s call!

5 Ways to Use Predictive Intelligence in Sales & Marketing

Let’s look at five practical ways you can apply predictive intelligence to your sales and marketing efforts.

1. Facilitate more accurate lead scoring.

Lead scoring is the process or system used to rank the sales-readiness of each lead you generate. Marketers score leads based on a set of predetermined criteria so they can better route, nurture, or sell to them.

The criteria involved in B2B lead scoring may involve data points like industry, company size, or a completed action such as a form submission or content download. Although manual lead scoring can be effective, this process has its flaws.

For one, lead scoring relies heavily on assumption—a lead with the job title of Marketing Manager might receive a high score, but that doesn’t guarantee they’re qualified or ready to make a purchase.

Predictive intelligence allows for a much more comprehensive approach to lead scoring. In fact, the latest technology can identify patterns in a prospect’s entire digital footprint, from the terms they search to the web pages they visit the products they’ve purchased from you in the past.

Predictive algorithms can analyze these behavioral patterns and accurately predict when each lead will be ready to make a purchase and what actions will accelerate them through the sales cycle—instantly!

2. Offer recommendations in real-time.

If you’ve used the internet at any time in the past decade, it’s safe to assume you’re familiar with e-commerce sites like Amazon. And, you know that when you visit Amazon’s homepage, you’ll likely find custom recommendations based on previous purchases or searches.  This is a perfect example of predictive intelligence at work.

Predictive intelligence allows marketers to create hyper-targeted and dynamic web experiences. When a customer visits a specific page or views a certain item, your site can process this data in real time and offer personalized recommendations based on that person’s actions.

Rather than creating a static website, predictive intelligence makes your site intuitive and unique to each individual customer—which can drastically improve conversion rates.

3. Improve your content marketing strategy.

Content is an increasingly important part of the buyer’s journey—and predictive intelligence helps marketers tailor their content marketing strategy to fit each prospect’s needs and preferences.

Predictive intelligence analyzes prospect data and provides insights into the subjects, tones, and content types that your target customers respond to. Given the time and energy that it takes to create content, these insights are essential to a streamlined, efficient content creation strategy.

4. Improve and scale your SEO efforts.

Search engine optimization is a never-ending process of adjustments and reactions. As Google algorithms update throughout the year, marketers often fall behind the current trends and their website quickly becomes outdated. Predictive intelligence allows marketers to collect and analyze data to help them anticipate and react to trends quickly.

5. Upgrade your email marketing strategy.

Email marketing is one of the oldest marketing tactics around, yet it remains one of the most effective. But, the standards for email marketing have changed. One-size-fits-all campaigns are no longer effective and must be replaced by personalized emails that pertain to the recipient’s preferences.

Predictive intelligence has made it possible for email marketers to personalize their B2B marketing emails to meet the needs of each individual subscriber.

The most common example is an abandoned cart email. A customer places an item in their virtual cart and subsequently leaves the page. After a predetermined amount of time passes,  modern marketing automation platforms can send this prospect a follow-up email, reminding your prospect to complete their purchase.

Final Thoughts on Predictive Intelligence in B2B Sales & Marketing

As we unpack the proverbial “predictive black box,” the most surprising takeaway is not that any data point is a magic bullet. There is no single data point that can do the work of good salesmanship.

What is surprising is that when all three types of data – Fit, Opportunity, and Intent – are present, they are tremendously effective.

Over 95% of survey respondents can link growth to predictive indicators. The positive result with the highest correlation is most often higher conversion rates of prospect to qualified lead.

At the end of the day, business won is truly the #1 data point for sales professionals and marketers alike.

If we’ve piqued your interest and you’d like to explore a solution, be sure to check out ZoomInfo’s Intent Data & Alerts feature. It leverages our best-in-class B2B database and sales intelligence tools to predict buying intent from the prospects you care most about. You can try a free trial today!