So, your inbound marketing is a success. The leads are flowing in, and the phones won’t stop ringing. Traditionally, sales professionals have no choice but to make deliberate and equal efforts for all leads that express interest. Many consider this ‘pot luck’ approach an occupational norm, but, it no longer has to be.
This is where lead scoring comes in.
Yes, you caught me. This topic has been around for many years. However, the world of machine learning and artificial intelligence continues to develop our ability not just score/qualify leads on their actions, but do so in a predictive manner. But first, let us consider the basics.
Before we start, I must point out that not every business requires a lead scoring model. Consider the following before reading on:
If ‘yes’ has been your answer to any of these questions, this blog is probably for you.
Lead scoring is the process of assigning numerical ‘scores’ to prospects based on their perceived value to an organisation. It requires the use of a logical model based on data collected from previous lead/customer interactions. Lead scoring can maximise your sales teams’ time while also producing more detailed metrics on your sales pipeline.
Like many things in marketing, one size does not fit all. The type of attribute a business might use to determine a leads worth varies… A valuable trait to a hair salon will differ significantly from a legal firm.
These attributes, actions and characteristics can be captured across many platforms, four of which can be found below:
Most businesses use email as a form of marketing. Whether they are intensive email campaigns or monthly newsletters, we can use the data to capture user interest. Examples of metrics might include open and clickthrough rates. Tracking such data can provide a general overview of a lead’s interest in your product or service. Now, some might stop there, but how can we use this information to even better effect?
Email campaigns often have different objectives; these might include bringing awareness to a new product, rewarding subscribers with fantastic deals or update your audience on a re-brand. Each email the subscriber opens is going have varying importance to the organisation. For example, a lead that opens and clicks through to a product demonstration or books onto a webinar is likely to be more valuable than one that has opened a newsletter.
Email Automation tip – Automate your emails based on a lead’s score, it’s a great way to reward those that have already engaged in your email content.
Every business that operates online should track some form of analytics. It allows us to assess the outcomes of our activities and whether they are in line with our marketing objectives.
In the past, demographic and firmographic information was retrieved through a tangled network of half-complete forms. However, advances in digital capabilities now allow us to use reverse IP lookup to find information without it being directly provided! (Similar to a sonar detector tracking fish). As more data is collected, you can fit together the puzzle-like pieces of your audience’s interactions. Based on these interactions, your audience accumulates a lead score. If over time a particular viewer reduces their activity, the score decreases to represent their lessening engagement.
The extent to which a lead engages with your social networks can also give information about their possible interest. How many times did they click on your companies tweets or LinkedIn posts? How many times have they followed a CTA? The value of social media engagement varies depending on the business. Generally at this stage, actions a lead takes will receive a low number of points. Social media engagement is usually near the start of the buyer cycle; again, this is very business dependent.
As we all know, fake leads are difficult to avoid. Giving negative scores to leads that haven’t engaged with content properly can be useful in filtering out spam. For example, were the first name, last name, and/or company name not capitalised? Did the lead use random characters to fill any fields?
Interestingly, we can also look at the types of email your leads use. If your company is B2B, there is a high likelihood they using a business domain. If visitors fill our your forms with ‘@gmail.com’ or ‘Yahoo.com’ you can remove points from their lead score!
Dip a toe in the water with manual lead scoring, it can be a great way to understand the value of a small number of leads and their respective variables.
The most straightforward way to do this is as follows:
Calculate the lead to customer conversion rates of all your leads
Decide on which characteristics contribute to becoming a high-value lead
Calculate the individual close rates for each characteristic
Compare the close rates of each characteristic with your overall close rate and assign point values accordingly.
As you can see from the manual method featured above, lead scoring is a time-consuming task. Furthermore, the implementation of such systems can be prone to inconsistencies, diluting their otherwise distinct benefits. Establishing a set of criteria to score leads is a rather involved process with many iterations and revisions, so how can a time-poor business reap the rewards of lead scoring?
Machine learning has developed use cases across a range of industries, providing new ways to automate in smarter, almost human-like ways.
Machine learning can analyse thousands of data points to understand what your customers have in common, painting a clear picture of what a high-value lead might look like. Furthermore, data from non-converting leads is also considered, filling in any gaps and making the overall criteria more accurate. This allows you and your sales team to spend more time engaging with prospects who will be most receptive to your product offering.
Want to find out more about predictive lead scoring or any other topic mentioned in this blog? Contact the author directedly (email@example.com).