A survey cited in Spona's guide to lead scoring best practices found that 68% of sales professionals say lead quality improved over the past year. But "better" is a feeling reps report, not something anyone measured. Without a scoring model behind it, the improvement is a hunch.
A quick refresher
Lead scoring assigns point values to leads based on who they are and what they do, so sales knows who to call first. Explicit data covers job title, company size and industry. Implicit data covers behaviour: pricing-page visits, demo requests, the signals that show intent without anyone saying it.
The mistake most teams make
They build the model backwards. They pick criteria that sound right, assign points and launch. Then they spend six months wondering why high scorers do not close.
The fix
Start with closed-won deals. Pull the accounts that converted and look for what they share: company size, industry, the pages visited before booking a demo, how many touches it took. Then do the same for leads that never converted. The gap between the two groups is your scoring criteria. Not a guess about what a good lead looks like, but what one looked like before they bought.
Two conditions
• Data-driven from day one. Building on instinct and fixing later means months of chasing the wrong leads while marketing insists the scoring is fine.
• Bring sales in early. Reps on the phones every day spot patterns, such as a title that never buys or a behaviour that always signals urgency, long before the data shows them.
What comes next
Combine fit and intent signals, weight high-intent actions more heavily, subtract points for poor fit, set thresholds that trigger an automated handoff, and review quarterly.
Where the data comes from
Closed-won analysis needs complete firmographic records. If company size and industry are blank on half your deals, the pattern is invisible. Spona's guide to lead enrichment covers filling those fields, and how lead scoring software works covers running the model. CRM enrichment from Spona completes the records the analysis depends on.
Views: 5 · 30s+ reads: 2