Marketing Operations
Lead Scoring That Actually Predicts Revenue
Most lead scoring rewards engagement. Engagement does not predict revenue. Here is what does.
Most B2B lead scoring models are built on engagement: page views, email opens, form fills, content downloads. Engagement correlates with attention, not buying intent. The result is a high MQL volume that converts poorly to revenue, and a sales team that learns to ignore the score.
The predictive model that works is fit plus intent. Fit is the firmographic and persona match to your ICP (industry, company size, role, geography). Intent is the behavioral signal that the buyer is actively evaluating (multiple visits to the pricing page, comparison searches, intent data spikes from 6sense or Bombora, repeated visits from the same account in a 14-day window).
Combine fit and intent into a simple matrix. High fit + high intent = sales priority. High fit + low intent = nurture. Low fit + high intent = qualify before passing. Low fit + low intent = monitor. This matrix is more useful than a numeric score because it tells the sales rep what to do, not just how excited to be.
Iterate the model quarterly using closed-won data. Look at which leads converted to revenue and reverse-engineer the score. Drop signals that do not predict. Add signals that do. After 12 months, the model is doing the work for you, and the sales team trusts it because the conversion rate proves it out.