Lead Scoring Points: Why The 100-Point Rule of Scoring is Wrong

If you’ve used lead scoring before, you may not have a particularly high opinion of it. That’s because most lead-scoring tools have implicitly led you astray. 

The traditional approach to scoring suggests a finite scale of 100 points.  

That scale is then divided into as many cohorts as you would like, implying that the higher the score, the better the lead quality.  

Clearly, a threshold is required to trigger actions or processes; otherwise, what would be the point?  

However, there are several reasons why the path to that threshold should accommodate a much more elastic model, going beyond the conventional 100-point system and adopting a co-dynamic approach. 

How the Conventional 100-Point Lead Scoring System Works 

As a quick recap: Conventional lead scoring systems use a 100-point system. Certain traits, actions, or other factors will cause a lead to “gain” a specific number of points, and some will cause users to “lose” points. Each action or trait is assigned with a numeric value, and ideally all of the factors will add up to a numeric score compared to a 100-point ideal total. 

You might get 10 points for coming as a referral, or 5 points if you sign up for a free trial from Google search, but only 1 point if you enter as a lead through an ebook. 

So a lead that scores 50 out of 100 points may be a warm but not hot lead, and one who scores 70 or above is one that the sales team should (in theory) pursue most aggressively.

Most of these systems rely almost exclusively on either fit—like demographic traits of a company—or a contact’s activity. They rarely take both into account, and if they do, they almost certainly neglect either the recency or frequency of different desired actions (or both!). 

Why You Need a Co-Dynamic Lead Scoring Tool 

Conventional lead scoring relies on a single, static score. You get one number between zero and a hundred, and that’s that.

Here at Breadcrumbs, however, we firmly believe in what we call co-dynamic scoring. This is a fancy term for a model that scores both the fit (which assesses how well a lead aligns with your ICP or buyer personas) in addition to actions a lead is taking. We’re going to talk more about what this means and give you a preview of how we can help a little later on. 

How This is Different From Conventional Lead Scoring Tools?

We’ve mentioned how co-dynamic models account for both fit and activity; we also take the recency and frequency of different actions into consideration, so someone opening an email today is considered a warmer lead than someone who opened an email six months ago. 

Unfortunately, many tools end up focusing on just one or the other (usually just fit), or they treat them independently of each other–both of which are a mistake, but that’s a topic for a different day.  

Regardless, the typical thinking manifests itself in something like the below:

Essentially, this simplified example demonstrates that Acme Inc. is interested in speaking to CEO’s of early-stage tech companies with revenue between zero and $1 Million dollars.  They also believe that if that persona visits their website four times, clicks on 30 emails, and downloads 3 assets, they are “sales-ready.” 

Don’t get me wrong: This model, depending on circumstances, is much better than nothing.  Ultimately, lead scoring is intended to drive a more efficient organization by positively impacting Value, Volume, Velocity and Conversion. 

You may recognize this as the 3VC framework espoused by our friends at GoNimbly, but the basic idea is that the following factors play a crucial role in your overall sales potential:

• Volume of leads, and how many leads are being managed at each stage
• The velocity of your sales cycle—aka how long the average sales cycle takes
• The value of your deals, and how that’s being impacted 
• The percentage of your opportunities that are converting; what’s causing leakage at each stage? 

However, it’s missing out on two additional elements that can take this simple manual model from “better than nothing” to kick ass, which we’ve briefly touched on: Recency and Frequency. There’s also a solid chance that there’s a problem with poor data quality or hygiene. 

Recency and Frequency

Recency and frequency of actions taken are a crucial part of lead scoring, so you should never use a tool that doesn’t take them into account.

Is a user who booked a demo a year ago as valuable as one who booked a demo today? No, of course not—we know that there’s a natural decay based on time, and more recent actions mean a higher chance of conversion. 

Similarly, frequency is also important. Someone who is logging into their free trial daily is going to be much more likely to become a customer than someone who logged in the day they created it and hasn’t touched the tool since. 

We’ve been preaching about the importance of time as an element of lead scoring since day one. It is plainly self-evident that the recency of action impacts its value. 

Layer on frequency, so they’ve visited that pricing page three times in the last week is more interesting than they’ve visited the pricing page three in the last year. All that information is relevant and should be tracked, but it’s essential in a lead-scoring model. 

In our example above and in practice, most models make no distinction. In fact, many models obfuscate the data by leveraging some arbitrary decay or reset rule. Having some sort of decay after six months, for example, isn’t useful when your lead intent declines significantly one week after initial contact. 

The specific time frames for different actions matter, and you need to be able to customize that with the models that you’re using. 

Breadcrumbs treats recency and frequency as native variables for each scoring category allowing for the fine-tuning of signal to noise. You can dive deeper into this concept by reading about RFM analysis myths, which looks at how models based on recency, frequency, and monetary value can be invaluable when used correctly. 

Limitations of Data 

In addition to recency and frequency, we also need to talk about the potential limitation that comes with poor-quality data.

We’ve found that many organizations have an absolute plethora of data… but it’s often messy and disorganized, spread across disparate systems. 

A recent article from Forbes suggests that the average company may have as many as 130 SaaS applications, with large enterprises having as many as 400, most of which will have their own data existing in a silo. 

So what does that mean practically? Well, it means that potentially meaningful data is distributed across many data sources. You don’t want a lead scoring tool that only draws data from your email account or a single analytics platform, which, unfortunately, many tools will resort to.

Instead, it’s essential to choose a tool that integrates data from across multiple tools into a single lead-scoring tool that will look at all of your data and take it into account. 

How Breadcrumbs Can Help 

Going beyond a 100-point lead scoring model is automatically going to give you a much more complex and accurate look at the potential of every contact in your pipeline. 

Below is a simplified example of a “beyond 100” model:

As you can see, both the fit and activity side of the model allow for scores above 100. This is to accommodate for lack of data completeness or capturing of signals that are independently contributory to the prediction.  

Ultimately the threshold is still 100, and any cohort up to and including those who score 100 can be handled through various workflows or processes. By allowing more contributing categories, you don’t miss out on relevant signals based on an arbitrary scale of 100.

That being said, we strongly recommend opting for a true co-dynamic approach that goes beyond just numbers. Here at Breadcrumbs, we use an alpha-numeric score to give you a much better understanding of the potential of different leads. 

We look at fit, activity, recency, and frequency. That high-value ICP user that booked a free demo (a high-intent activity) six months ago will—correctly—not be valued as highly as a user who isn’t quite an ICP match but who booked a demo this week and has logged into their trial multiple times and come close to maxing out their existing plans.

There’s a great deal of nuance in sales, and that’s where we can help, especially since we have plenty of templates that we’ll suggest but that you can fully customize (and split test!) as needed. 

Final Thoughts

In reality, most lead scoring models will be better than none for a number of reasons, including the following: 

1. The sooner you start, the sooner you begin to learn
2. Although every lead seems valuable (especially during the early days), logic dictates this isn’t true, and the focus should be on the 3VC framework mentioned above
3. A strong collaboration between sales and marketing is endlessly touted as a key ingredient to exceptional outcomes, and lead scoring lays the foundation for a shared vernacular and data-driven foundation for the relationship between sales and marketing

That being said, opting for a strong and co-dynamic lead-scoring tool is your best bet. You want to have the most actionable, detailed data possible, and that’s only going to happen when you have a tool looking at fit and activity and recency and frequency. Any tool that neglects any of the above in any capacity will give limited (and sometimes inaccurate) data.

And when you’re looking at data that can impact sales performance, the one thing you want is up-to-date, accurate data that can help them. Choose a tool that can help with that.

Ready to get started with a co-dynamic lead scoring tool? Create a free account to get started now, or book your free demo here.