What makes the difference between a qualified and unqualified lead? That’s the question on everyone’s lips who runs a website and wants to be sure of getting a satisfactory return on their investment. Now, how do you score leads to decide on their quality? There are two main models, one rather “traditional” and the other “predictive”. Let’s take a closer look at what they look like, and how choosing the right lead scoring model can help you considerably.
The importance of setting priorities
First of all, let’s bear in mind that the usefulness of lead scoring will depend on your stage of development. Do you currently have a sufficient number of leads? How is your sales team dealing with these leads to get them to buy your products? Do you have enough data to assign them a score and rank them according to their quality? To be able to assign a score to your leads, you still need to have data on leads that have failed, as well as on leads that have resulted in sales. Then you can analyze them to set your strategic priorities (not forgetting the risks of collecting non-representative samples). If you’re just starting out, you’ll be short of data, so it’ll be difficult to set a definitive course at this stage. Instead, your priority should be to collect more leads.
The traditional lead scoring model
Traditionally, most marketers use the “BANT” model for lead scoring: Budget Authority Need Timeframe This model has proved its worth in the past, but is probably no longer sufficient. For example, it doesn’t take into account a factor such as the type of addresses used: while you’re looking for B2B customers to sell them a relatively expensive SaaS solution, you receive a request from an @gmail.com or @yahoo.com address? On the face of it, this is a pretty bad start, but the classic BANT system won’t allow you to quantify this. Keeping score against current data is therefore necessary, but the criteria need to be refined according to your industry. On the other hand, for leads with high potential but still needing to mature a little, it’s a good idea to employ a predictive model.
The predictive lead scoring model
With a predictive lead scoring model, we rely on an algorithm that cross-references a much larger number of data points, and no longer simply assigns points to individual criteria. We no longer rely solely on the information provided on a form, but also on actions taken on the site, behavior on social networks, demographic profile, etc. While this model is much more complex than the traditional BANT model, it’s also much more accurate: by comparing your leads with the profiles of customers who have already purchased from you, an algorithm can much more easily predict the chances of a lead translating into subsequent sales or not. The same applies to comparisons with leads and personas that haven’t converted into sales in the past.
How do you set up these lead scoring models?
Depending on your current objectives and stage of development, it’s up to you to decide whether the BANT model is sufficient for now, or whether you’d like to adopt a more powerful predictive model. In any case, there are a number of applications that can help you set up or improve your lead scoring protocol. But above all, never forget to collect enough data to be able to come up with a relevant lead scoring!
