Lead Scoring

Lead scoring is triage for the pipeline. It assigns a numeric value to every inbound lead based on two axes — how well they fit your ideal customer (firmographics, role, company size) and how engaged they are (page views, demo requests, email clicks) — so reps work the prospects most likely to buy first instead of dialing top to bottom. A well-tuned model can lift sales productivity meaningfully: when 50% of leads in a typical funnel are never sales-ready, scoring is what keeps an AE from spending Tuesday on tire-kickers. The output is usually a number, a grade, or a threshold that promotes a lead from marketing's hands to sales'.

How Lead Scoring Is Calculated

Most models are additive point systems with two scoring dimensions, sometimes combined into a grid (A1, B3, etc.).

A lead crossing a set threshold — say 50 points — converts to a marketing qualified lead and routes to sales. Modern stacks increasingly replace hand-set point values with predictive models trained on which past leads actually closed.

A Worked Lead Scoring Example

A VP of Operations at a 500-person target-ICP company downloads a whitepaper (+10), visits the pricing page twice (+30), and requests a demo (+30). Title and firmographics add another +35. Total: 105, well past the 50-point threshold, so the lead becomes an MQL and hits an SDR's queue within minutes. Compare that to a student using a personal email who reads one blog post: +5 behavior, −10 for the free domain, net −5. The model parks the second lead in nurture and never wastes a rep's call on it. Multiply that decision across 2,000 monthly inbounds and the model is effectively allocating the entire SDR team's calendar.

When Sales Teams Use Lead Scoring

Lead scoring lives at the marketing-to-sales handoff, so it's owned by demand gen and RevOps but felt hardest by the SDR who works the queue. Marketing leaders use it to define what counts as an MQL and to report pipeline generation; sales leaders use it to defend rep capacity, since every junk lead routed to a rep is a real call not made. The score also governs SLA: many orgs commit to working any lead above the threshold within a fixed window, which makes the model a contract between two teams, not just a sorting tool.

Common Lead Scoring Misconceptions

A lead score predicts engagement, not intent, and the two diverge constantly. The most common failure is a model that rewards activity without fit — a competitor's analyst clicking every email racks up behavioral points and lands on a rep's desk looking like a hot buyer. The most common manipulation is structural: marketing tunes the scoring threshold to hit its MQL quota, lowering the bar near quarter-end so the lead count clears target while quality quietly drops, which is why MQL volume and sales accepted lead conversion have to be read together. A model also decays the moment your ICP shifts and nobody retrains it, so last year's point weights keep promoting the wrong companies. Lead scoring tells you who looks ready. It does not tell you who has budget, authority, or a real problem — that's still discovery's job, and a high score that converts to a stalled SQL is just a faster route to a dead deal.