more than frustrating. Multiply that by the number of sales people on your team, and you’ll get a mind-boggling amount of wasted productivity.
In a similar vein, thinking that you have a healthy pipeline, only to discover that none of the prospects are ready to buy your solution can have disastrous consequences for your sales that quarter.
Industry surveys show that up to 50% of your prospects might not be a good fit in the end. But how can you find and eliminate those early in the process? The answer is having a good lead qualification model.
Quick reminder - what are lead qualification models?
Even though most sales teams already qualify their leads, it doesn’t automatically guarantee high-quality results.
Historically, a lot of lead qualification was subjective: sales people would often (and still do) use their prior knowledge and experience to pick what they think are the best opportunities at the time. Sometimes it works, as experienced salespeople do develop a knack for determining good leads - but you can see why this is not a very reliable approach, especially for junior members of your sales team (or at scale).
Another approach is to qualify leads in the order they come in. While this increases the qualifying consistency, it risks losing potentially high-value leads who won’t wait long enough for you to get to them. Plus, your sales team still has to manually reject every low-quality lead.
Lead qualifications models help solve these problems because they introduce a quick and repeatable workflow into your sales process. Instead of sorting leads after reaching out to them, you filter them right at the root.
The key to modern qualification models is this: You come up with criteria by which you can judge whether a lead is a good fit and you apply it to all incoming leads automatically.
However, when criteria are left up to human hypothesis, do they really serve as a reliable foundation to a lead qualification model?
Let’s find out.
How to use lead qualification models
There are generally give steps when using traditional lead qualification models:
- You define your ICP (ideal client profile) and its characteristics
- Set expected values for each characteristic to find an ICP fit
- Build a source of explicit data (when a prospect “gives” you their information). Traditionally, this could be having a form on your website or collecting data in person at conferences.
- Find a source of implicit data (also called third-party data). Some CRMs can source data for you as well as lead qualification tools.
- Use a tool to add up the information from #3 and #4 to determine which leads are hot or cold, integrating them into one pipeline ready for the sales team to act on.
However, this commonly used process sets a foundation for low volume, low quality lead scoring.
Defining your ICP - helpful, but only for a handful
For example, if you run a company that provides B2B finance analytics software in the US, your ICP might be a middle-aged senior investment banker with a house in the Hamptons.
You can also classify characteristics into must-haves, should-haves, nice-to-haves, etc.
Some characteristics to consider are location, income, job title, age, income, interests, etc. Your must-haves could be living around New York and making over $500,000 a year. Your should-haves could be working in finance and being over 40 years old. Your nice-to-haves might be enjoying golf. These characteristics and their respective values will vary depending on your company and the type of customer you’re looking for.
The problem here is that focusing too narrowly on an ICP results in a small handful of targets amongst a sea of possibilities. A narrow focus can exclude other prospects who could or would be willing to buy from you, but your program hasn’t considered them.
The other major problem here is determining how someone fits these criteria, which brings us to lead scoring.
Finding your ICP fit - lead scoring based on hypothesis and third party data
Note that not having a certain characteristic shouldn’t automatically mark your leads as unqualified, because it depends how well they fit your other criteria. That’s why a lot of traditional lead qualification is a scoring process, known simply as “lead scoring”.
The way that most lead scoring tools work involves the discovery of necessary values for your ICP characteristics. These tools try to combine explicit with implicit data.
Explicit data is what your leads tell you (e.g. their name and email address). Implicit data is what you or your tools help you find out (e.g. a third party database of contact information).
Here’s the big problem with lead scoring: they rely on hypotheses to weigh the characteristics of a prospect. Think about it - a human somewhere in your organization has determined who a good target is (a great start, but excluding thousands of potential other buyers) and now a lead scoring tool is attempting to add up scores for those characteristics based on manually-inputted or estimated weights.
In an attempt to modernize lead scoring software, some tools will include a few basic website actions in their calculations. For example, if a visitor goes to your pricing page it may suddenly increase their lead score. That’s a good start, but doesn’t truly represent a visitor’s intent. For example, 64% of ready-to-buy website visitors don’t even go to a pricing page.
The other big problem is that lead scoring tools only work with identifiable visitors - meaning the small handful of visitors that can be matched to a database of contact details based on an IP address or similar. However, some studies show that up to 98% of website visitors can be anonymous - that’s a tiny portion of your website traffic. That’s not to mention that those databases of contact details are often outdated or inaccurate.
How to design a better lead qualification model - real-time behavior and buyer intent
So, we know that traditional lead qualification methods such as lead scoring can be ineffective. But how do you get around their inherent challenges?
As with most breakthroughs in traditional thinking, we turn to leading-edge technology for the solution. Specifically, machine-learning models that can determine how likely a prospect is to buy based on their real-time behavior, represented as “buyer intent”.
Think about this - it’s important to determine who your target market is and attempt to catch them, but would you turn away a prospect who is ready to buy right now while you’re out chasing those ideal clients? No, surely not.
The only tool that can perform this function in real-time and repeatable accuracy is Lift AI.
Lift AI is a buyer intent technology that is able to evaluate the behavior of website visitors in real-time as they navigate your website, assigning buyer intent scores to each visitor that gets updated constantly. This works even if a website visitor is completely unknown to you (not revealed by your lead scoring and ABM tools).
Visitors can be segmented into high, medium or low buyer intent then acted on in real-time.
For example, you could connect the highest-rated visitors directly to your BDRs through chat (e.g. Drift). Lower-scoring visitors aren’t then discarded, but rather delegated to a nurturing bot or a self-help guide - saving your sales team precious time being spent with the wrong visitors.
You also have the option of integrating Lift AI’s results with other tools you already use, including your traditional lead qualification tools! Imagine a scenario where your ABM and lead scoring tools find an ICP fit on your website, then combine that data with a high intent Lift AI score, meaning the visitor is sales-ready right now. Suddenly, you reveal the best, red-hot leads on your website and can act while the visitor is still on the site, rather than following up in the future where there’s a lower chance of conversion.
After just 90 days, most Lift AI customers report up to 10 times improvements in chat conversions. Formstack, for example, increased its lead pipeline by 88%, while PointClickCare improved its own by 400%.