Lead scoring assigns each lead a score that estimates how ready and how suitable they are to buy. Rules-based scoring adds points for attributes such as job title and company size and for actions such as visiting the pricing page. Predictive scoring uses a statistical model trained on past wins and losses. Either way, the score decides who becomes a marketing-qualified lead and who sales contacts first.
Fit and engagement
Most scoring models combine two dimensions. Fit asks whether this person and company could buy. Engagement asks whether they are showing interest now. A high-fit, low-engagement lead may suit outbound or nurture. A low-fit, high-engagement lead may be a student or competitor and should not take up sales time.
- Fit: role, seniority, company size, industry, technology used.
- Engagement: pricing page visits, demo requests, webinar attendance, email clicks, product usage.
- Negative signals: personal email domains, competitor domains, unsubscribes, careers page visits.
Weighting high-intent pages
Not all page views are equal. Visiting a comparison, pricing or security page usually signals more buying intent than reading a top-of-funnel article. Many teams give extra points for these pages. Validate weights against real conversion data rather than intuition, and remember that privacy settings and consent choices mean behavior is not tracked for every visitor.
Keeping the model honest
Compare scores with outcomes on a schedule. If high-scoring leads do not convert better than low-scoring ones, the model is not working. Predictive models need enough historical wins to learn from; with little data, a simple rules-based model that sales understands is usually more useful.
Example
A simple points model
Hypothetical company Kestrelbank uses an illustrative scheme: points for a target job title, points for company size in range, more points for a pricing or comparison page visit, the most for a demo request, and a deduction for a personal email domain. Leads above a threshold become MQLs. After a quarter, the team checks conversion by score band and lowers the weight on webinar attendance, which turned out to predict little.
Common mistakes
Scoring only engagement
Very active leads who could never buy end up at the top. Combine fit and engagement.
Set and forget
Scores drift as products, pricing and audiences change. Recalibrate regularly.
Questions
Yes, with product usage as the main engagement signal. That version is usually called PQL scoring.
