If you are in a position to implement a lead scoring system, take a moment to pat yourself on the back.
You’ve built a business that can help many people, potential customers are flocking to you, and you need a way of separating the good leads from the bad. That’s an impressive accomplishment.
Lead scoring is the fun part. It’s a good problem to be tasked with.
It’s what allows you to go after the best of the best leads, pursuing those who will help your business the most, while reducing the time and energy it takes to chase low-value prospects. If you want to boost company-wide efficiency by aligning marketing and sales teams, there’s no better way to do so than through lead scoring.
What is lead scoring?
Lead scoring is the process of analyzing everything you know about a lead and coming up with a number that represents both how likely they are to become a customer and how valuable they will be to your business as a customer. In short, lead scores tell you with a quantitative element which prospects are the best fit for your business.
Sales and marketing teams score leads as part of the qualification process, and the scores are used to determine when, how, and by whom the leads will be approached by the sales team. The scores are determined by actions that prospects take in the sales and marketing funnels (e.g. filling out a form, visiting a specific page, starting a free trial, downloading an ebook) as well as through the use of predictive analytics and 3rd party data.
When should you score your leads?
You should start scoring leads after:
1. You have enough leads coming through your door that you need to strategically approach accounts
2. You’ve taken the time to figure out your ideal customer profile (ICP)
Your ICP is determined by analyzing your top performing customers and figuring out what they have in common. Dig into the details of these customers, analyzing how many employees they have, where they are located, what tech stacks they use, and any other relevant information.
Once you have an ideal customer profile, you can structure your lead scoring framework so that high scores are determined by the criteria that matter most to your bottom line. You’ll want to start scoring leads sooner rather than later — in fact, lead scoring can boost close rates by 30%.
Lead scoring and personalization
The ultimate goal of lead scoring is to figure out how to build a personalized experience for each prospect, from the emails you send down to the CTAs they see on your site. If you can do that, conversion rates and user satisfaction will skyrocket.
After all, a recent survey from Adobe shows that 42% of consumers get annoyed when their content isn’t personalized, and 66% of consumers will not purchase from a site when they feel the content does not speak to their specific interests.
The structure of a lead scoring model
In order to build a solid lead scoring system, it’s important to understand the component parts that make up a robust lead scoring model. You can only accurately score leads once you have done your best to understand who your leads are, where they work, and the behaviors they engage in online.
Don’t overlook the basics. Demographic data, such as age, location, job title, and salary should play a large role in any lead scoring system. 30-year-olds from the Bay Area should probably be given higher priority than 60-year-olds form the Rust Belt (though there are exceptions to every rule.)
You can use a reverse IP lookup tool, such as Clearbit Reveal, in order to learn detailed firmographic data — company size, location, even tech stack. Imagine your ideal customer profile is an engineer at a FinTech company with 500-1000 employees. If you notice that someone from a company in NYC that fits those specs has visited your site five times in the last month, you can leverage the firmographic data to assign them a high lead score.
Behaviors are the measurable actions a prospect takes while on your site or engaging with your services. This can be anything from reading a blog post to filling out a survey. Each action gives you insight into what they want, and your lead scoring can be tailored appropriately based on the information.
You can weight different actions that are more likely to demonstrate intent. For instance, reading a blog post (3 points) might be less indicative of intent than visiting a sales page on the site (5 points).
Contextual behavior covers actions taken that aren’t directly related to your product but are clearly related. For instance, this could include the type of device, the IP address, or browser type. While on first glance, these things might not seem like huge indicators, with a big enough data set they can reveal a lot about your visitors.
For instance, the hour a prospect visits could demonstrate whether they are thinking of the purchase outside of traditional work hours. Or for your product, Mac users might be more likely to convert than PC or Linux users.
Track signups to your newsletter, email open rates, and click-through rates. These are all key indicators of how engaged a prospect is with your business, and thus how likely they are to become a customer.
Just like how your closest friends are the ones who like most of your pictures on Instagram, your biggest fans are the ones who are liking, commenting and sharing content from your brand. Monitor this social behavior and bake it into your lead score. This gives a good indication of off-site behavior that
What type of behavior is important?
A typical company should place a high value on the following actions in order to build the most relevant lead scoring system. These prospect behaviors tell you a lot, so monitor them closely.
If a prospect has visited your site multiple times in a single day, that’s telling that they could be ready to buy. This is especially true if they are visiting your pricing page.
Or if they’re visiting your product pages day after day.
What are your prospects interested in learning about? A lead that searches for “monthly pricing” is probably hotter than one who searches for “frequently asked questions.”
Both are showing intent on researching your product, but once a prospect has started to debate your product or service on price, your lead is red hot.
The frequency and type of download activity on your site can tell you a lot. Bump up the lead scores of people who download meatier content such as whitepapers and ebooks, especially if this occurs at the bottom of the funnel, with tactical content about how to use your service.
Webinar viewers are pure gold. They have shown the intent to engage with long-form, detailed content, which places them far above most casual browsers.
How do you build a lead scoring model?
Now that we’ve covered the nuts and bolts of lead scoring, let’s dive into the different ways of creating a lead scoring model.
Manual lead scoring
With manual lead scoring, start by setting up your lead scoring criteria. This will vary by company — there is no one size fits all formula. The key is to assess your sales and marketing process in order to determine what attributes and actions made for a high-quality lead. Look at what actions led to the highest close rates.
Then, assign points based on who the lead is and the actions they’ve taken.
Let’s take a simple example of how you might assign points based on firmographic, behavioral, and social media data, assuming once again that your ICP is an engineer at a FinTech company.
Visitor 1 —> A senior engineer at FinTech company (+50) visits your site and reads a top of funnel blog post (+2). Total score = 52
Visitor 2 —> An entry-level engineer at a sales firm (+5) visits your site, signs up for a webinar (+15), goes to your pricing page (+5), likes one of your LinkedIn posts (+3), and downloads an ebook (+4.) Total score = 32
In this case, it’s appropriate that the perfect fit for your ICP gets a higher score, but you can see how the multiple positive actions taken by the other engineer can make her a promising target as well.
Both could be qualified leads depending on your threshold, but all else equal, visitor 1 should be pursued first. This might not seem like a huge issue when comparing two leads, but when you have thousands of visitors flocking to your site each and every day, micro-actions can make a difference.
Probability-based lead scoring
This system builds off the work you’ve done with manual lead scoring. You input customer attributes into a spreadsheet, weight the importance of each, and then create a formula that crunches the numbers and spits out a probability that the lead will become a customer. It monitors past behavior and attempts to formulaically apply learnings to every new customer that enters the funnel.
You can work off these probabilities to sort out which types of leads to pursue.
Predictive lead scoring
This lead scoring model uses machine learning to assimilate first and third party data in order to automatically assign scores to leads based on demographics, firmographics, behavior, and intent. Think of it like probability based lead scoring on steroids. The differentiator is that a predictive lead scoring model assesses all your past opportunities and identifies what attributes lead to a close, so it can constantly update your lead scoring system without you having to do so manually.
How you’ll see lead scores arranged in Salesforce. Source: Salesforce
Whether you choose one of these models or a blend of all three, the key, as with any other aspect of growth hacking, is to test and iterate. Try one out, see what works best for your unique circumstances, and adjust.
Data is king
B2B lead scoring has never been more important. According to Forrester Research, only 8% of companies have sales and marketing departments that are fully aligned. Misaligned departments mean lost revenue and wasted time.
Couple that with the fact that companies that use lead scoring show a 77% boost in lead ROI over those with no lead scoring in place, and you’re left with the conclusion that building a good lead scoring system should be a high priority for any company with traction.
Interested in personalizing your website for high-quality visitors? Take a look at our new SaaS personalization software, Experiences.