A lead scoring model is the set of rules or statistical logic used to turn lead data into a score that helps sales and marketing decide who to prioritise. The model defines which signals count, how they are weighted and how the resulting score should be interpreted.

Start with the business decision, not the points

A useful model starts by defining what the score is meant to help decide. That might be which inbound leads deserve faster follow-up, which accounts should be prioritised, or when a marketing-qualified lead should be reviewed by sales. The scoring logic should then be built around the data that is relevant to that decision.

Fit and behaviour answer different questions

Many B2B models combine profile data with behavioural data, but the two are easier to manage when their roles are clear. Fit criteria such as company size, industry or role indicate whether a prospect resembles the target customer. Behavioural criteria such as product-page visits, return activity or form submissions indicate what the prospect is doing now.

SalesWings can keep fit grading and behavioural scoring visible separately in Salesforce, so a rep can distinguish a highly active poor-fit lead from a high-fit prospect who has shown little recent activity.

A simple rule-based model can be easier to trust

Rule-based scoring assigns explicit weights to selected attributes or events. Its main advantage is explainability: sales can see why a score changed. Predictive scoring instead uses historical outcomes to infer which combinations of signals matter. It can uncover patterns that are difficult to define manually, but it also requires suitable historical data and more care when explaining the result to users.

How to validate a scoring model

The score should be checked against actual outcomes. Useful questions include whether higher-scoring leads convert at a higher rate, whether the signals being rewarded still match the current sales process, and whether reps can understand the activity behind the score. A model that produces neat numbers but does not improve prioritisation is not doing its job.

There is also no universal handoff threshold. The right threshold depends on the model scale, sales capacity and observed conversion behaviour.

For the wider implementation options, see the Salesforce lead scoring guide. For the commercial platform, see SalesWings lead scoring software. Related terms: Lead Score Threshold and Predictive Lead Scoring.