”Have you ever heard the saying, “The best predictor of future behavior is past behavior?”

At Zoominfo, we’re always hypothesizing, testing, and optimizing programs so we’re working with best-in-class go-to-market standards. To help our reps maximize their time and focus on opportunities that are likely to close, we put our historical data under the microscope to see what we could learn to build an effective lead scoring model. 

We started by enriching prior leads and closed-won opportunities against dozens of data points, which created a mosaic of data that underlined the people and companies we engaged with. We took that data and identified the characteristics that led to higher conversion and win rates.

(Pictured: Example of what a ZoomInfo SDR sees when a lead comes in via Salesforce)

The process described above is the foundation of a predictive lead scoring model. Running leads and demos against this scoring model—and prioritizing leads and routing demos to sales reps best equipped to close them—raised ZoomInfo’s win rate by 130 percent.

To help you achieve the same impact, we’ll walk you the roadmap to how we did it. 

Enrich past leads and opportunities to find patterns.

When we found that typical data points—such as the size of company or industry—were neither deep or complete enough for us to build a valuable scoring model, we turned to the company attribute and technographic data that was available through our API and Enrich product to do it.

“One of the most predictive elements for good outcomes for us is the number of salespeople an organization has,” says Chris Hays, chief revenue officer at ZoomInfo. “Because that data point is directly correlated to win rate, as well as the average selling price (ASP) within the account.”

We took all past wins and ran them through ZoomInfo Enrich to bring back normalized, consistent data fields for those accounts.

“One of the most predictive elements for good outcomes for us is the number of salespeople an organization has,” Hays explains. 

After enriching the closed-won opportunities against dozens of data points in ZoomInfo’s platform, we found the data that best predicted a high probability for future successes. We did this by examining historical win rates across certain company attributes, including:

  • Title tier
  • Industry tiering
  • Year founded
  • Number of sales and marketing employees
  • Revenue
  • Number of employees
  • Technologies (such as CRM and marketing automation systems)
  • Number of company office locations
  • Has 5+ data scientists
  • Has a vice president of sales
  • Runs NetSuite
  • Uses AWS, Microsoft Azure, Google Cloud
  • Attended Dreamforce

Tip: You can build this in an automated way in ZoomInfo’s ICP module.

Here’s an interesting fact that we learned about its data and scoring: “A lead that doesn’t exist in the ZoomInfo platform has a very low win rate,” says Hays. 

To put that thought into numbers, prospects that are in the ZoomInfo platform have two times the win rate of prospects that do not show up in the platform. They also have a 55 percent higher ASP.  

This drives our ability to focus on the right accounts.

Narrow Down Stronger Leads With Sales Intelligence

The nuts and bolts of how this comes together look like this:

When a lead comes through, the ZoomInfo API matches the lead against records in the database and adds all the data points that were found to be predictive in the scoring model.

From there, the ZoomInfo platform rolls enriched leads into a predictive model that forecasts the likelihood of a sales win, the overall average contract value potential, and the ASP potential.

That looks something like this:

  • Has 5+ sales people: +5 pts
  • Uses Salesforce.com: +2 pts
  • Uses Eloqua: +1 pt
  • Uses Marketo: +1 pt
  • Matches the platform: +5 pts
  • Has 2+ locations: +3pts
  • Has a CFO: +2 pts

Based on the accumulated data, the lead receives a score.

Develop Routing Rules That Interact With the Scoring Model

The scoring affects how we route and prioritize leads and demos. For example, scores that predict a potential opportunity with a high ASP get routed to top sales reps, and scores associated with a lower win rate go to junior sellers.

“In a way that’s as automated as possible, I want to get my best people on the best leads, and the scoring model tells me which leads are the best,” Hays explains. 

Lou Wolf, vice president of new business sales at ZoomInfo, recalled a situation in which two of his reps closed the same amount each month. The first rep closed a small percentage of leads but sold a high ASP, while the second rep closed a high percentage of leads with a lower ASP.

Once Wolf noted those patterns, he used ZoomInfo’s lead scoring model to send the first rep high-level leads, but fewer of them, while also sending the second rep many lower-level leads. Their performance increased as a result, which provided important feedback to the company’s sales leaders on how to best route leads.

Other companies can refine territories based on different observations. For example, instead of ASP potential, a sales manager might assign lead routing that centers on specialized functions in an industry, such as medical imaging technicians versus radiologists in health care settings.

Get an Objective Definition of “Opportunities”

Determining win rates—particularly the ratio of closed-won opportunities to created opportunities—is complicated because there is no industry standard to measure it.

All companies agree about how to define closed-won opportunities: a customer signed an agreement and paid. The definition of created opportunities, however, is less scientific from company to company—and even sometimes from sales rep to sales rep.

To create an objective definition for “sales opportunities” at ZoomInfo, Chris Hays took this approach:

“If a demo is of a certain quality according to the scoring model, the demo is considered an opportunity,” he says. An opportunity is a qualified prospect with a high chance of closing, meaning they have a problem your product can fix, they can afford the product you’re offering, and they’re in a position to make a purchase decision.

This definition has remained consistent across ZoomInfo.

And this is important because the best way to understand account executive (AE) performance is to understand how they perform against leads that are provided to them. If two account executives both get 10 A+ demos and one closes two of them for $1M of ACV, and the other closes eight for $400K, then you know how to coach (your second AE needs to learn how to sell value and your first rep needs to figure out velocity).

Focus on Leads With the Highest Likelihood to Convert to Demos

ZoomInfo generates 15,000 hot leads per month, and many of those are not routed to reps. However, low-scoring leads are not entirely abandoned. Instead, some of the low scorers get picked up by the outbound demand generation team, which puts those leads into persona buckets (for example, VPs of sales, directors of marketing, regional sales managers, directors of sales operations, directors of talent acquisition), wraps campaigns around them and proactively markets out to that group.

“We’re continually going back and taking data from the ZoomInfo platform, augmenting it, supplementing it, rescoring it, and doing re-engagement campaigns,” Hays remarks.

Results: ZoomInfo’s Win Rate Boosted by Data

When sales leaders use a scoring model to prioritize leads, and additionally, develop demo routing rules that interact with the same model, account executives increase their productivity. 

“For ZoomInfo, the sales intelligence behind its scoring model led to a 130 percent increase in our win rate,” says Hays.

More Success Stories

Get your free trial today

Free Trial
Try ZoomInfo Free