Lead Scoring

Most Leads Sent to Sales Are Not Actually Qualified

The majority of leads passed to sales are not truly qualified — and lead scoring is the mechanism that should prevent that. Vendor research suggests lead generation ROI roughly doubles with proper scoring, and conversion rates improve meaningfully through better alignment between marketing and sales. The problem is that most lead scoring implementations are broken. They score what is easy to measure instead of what actually predicts buying.

Lead scoring assigns numerical values to prospects based on two dimensions: who they are (firmographic fit) and what they do (behavioral signals). Both matter, but only one predicts timing.

Firmographic vs. Behavioral Scoring

The most common mistake is scoring based on profile fit alone. A VP of Sales at a 500-person SaaS company matches your ICP perfectly — but if they have not visited your site in three months, they are not ready to buy. Behavioral scoring catches that distinction.

Scoring Dimension	What It Measures	Strengths	Weaknesses
Firmographic	Company size, industry, title, revenue	Identifies potential fit	No indication of timing or intent
Behavioral	Page visits, email engagement, content downloads	Indicates active interest and timing	Can trigger on non-buyers (researchers, competitors)
Combined	Fit plus behavior	Most accurate prioritization	Requires both data sets and ongoing calibration

The highest-performing models weight behavioral signals more heavily for routing decisions (who gets called now) and firmographic signals more heavily for territory and segment assignment (who owns the account long-term). See engagement scoring for a deeper dive on behavioral signal hierarchy.

Why Scoring Models Decay

A lead scoring model that worked last year is probably wrong today. Buyer behavior changes. New channels emerge. Product positioning evolves. The actions that predicted buying intent twelve months ago may not predict it now. Teams that build a scoring model and never revisit it end up with a system that routes leads based on outdated assumptions.

Calibrate quarterly at minimum. Pull a sample of won and lost opportunities, review the scores they received at handoff, and adjust weights accordingly. If high-scoring leads are not converting, your model is rewarding the wrong behaviors. If low-scoring leads are closing, your model is missing important signals.

Connecting Scores to Action

A scoring model that does not trigger specific actions is a reporting exercise, not a sales tool. Define clear thresholds: above a certain score, the lead routes to sales immediately. Below that threshold but above a second, the lead enters an accelerated nurture track. Below both, the lead stays in general marketing. Every score range should have a corresponding action and owner.

The connection between lead scoring and marketing qualified lead definitions should be explicit and agreed upon by both marketing and sales. When both teams own the scoring criteria, the MQL-to-SQL handoff stops being a source of friction and becomes a system both sides trust.