What Is Lead Scoring?
What lead scoring is
Lead scoring is a system that assigns a numeric value to every lead so you can rank a large pool by how sales-ready each one is. Instead of treating all new leads equally, you award points for the traits and actions that historically precede a closed deal, and subtract points for signals that predict the opposite. The output is a single score, or a pair of scores, that tells sales who to call first.
The goal is prioritization, not gatekeeping. With finite sales capacity, scoring concentrates follow-up on the leads most likely to convert and prevents good opportunities from sitting untouched in a long queue. It also creates a shared, objective definition of "qualified" that marketing and sales can both stand behind.
Explicit vs implicit scoring
Lead scoring blends two kinds of signals: explicit (who the lead is) and implicit (what the lead does). Explicit scoring rates fit using attributes a lead provides or that enrichment fills in. Implicit scoring rates interest using behavior tracked over time.
| Type | Answers | Example signals |
|---|---|---|
| Explicit (fit) | Are they the right buyer? | Job title, seniority, company size, industry, region, budget |
| Implicit (behavior) | Are they interested now? | Pricing-page visits, demo requests, email opens/clicks, content downloads, event attendance |
Why many teams keep them on separate axes
A single combined number can hide important nuance: a perfect-fit executive who has barely engaged scores the same as a highly active student who will never buy. Splitting fit and behavior into two axes (often a letter grade for fit and a number for interest, such as "A2" or "D4") lets you route only leads that are both a good fit and engaged, and treat the others differently. You can nurture good-fit, low-interest leads, for example, rather than sending them to sales.
An example point model
A simple model lives on a 0–100 scale, awarding positive points for strong fit and engagement and deducting points for negative signals. The numbers below are illustrative starting points, not benchmarks. Calibrate them against your own conversion history.
| Signal | Type | Points |
|---|---|---|
| Title contains "Director" or above | Explicit | +15 |
| Company size 200–2,000 employees | Explicit | +10 |
| Target industry | Explicit | +10 |
| Visited the pricing page | Implicit | +15 |
| Requested a demo | Implicit | +25 |
| Opened or clicked a nurture email | Implicit | +3 each |
| Personal email domain (gmail, etc.) | Explicit | −10 |
| Competitor or current customer | Explicit | −20 |
| No activity in 90 days | Implicit | −10 |
Notice the negative scores. Deductions are what stop poor-fit or disengaged leads from drifting over the line on volume of low-value activity alone. A model without them tends to inflate every lead toward "qualified."
How MQL thresholds work
An MQL threshold is the score at which a lead is judged ready to hand to sales, for example, 50 of 100. Cross the line and the lead is flagged as a Marketing Qualified Lead and routed for follow-up; stay below it and the lead keeps nurturing until its score grows. The threshold is the operational link between a scoring model and the MQL → SQL handoff.
How to set the number
Set the threshold by working backward from two things: how many leads your sales team can realistically work, and what your historical data says about which scores actually convert. Pick the cutoff that passes roughly that volume while keeping the conversion rate of accepted leads healthy. There is no universal value. A frequently cited starting point is "half of your maximum possible score," but treat that as a hypothesis to validate, not a rule.
Tuning it over time
Review the threshold on a regular cadence with sales. If reps complain MQLs are weak, raise the bar or add negative scoring; if sales is starved or great leads slip through, lower it or re-weight the strongest predictive signals. Scoring is a living model, and the threshold should move as your data and capacity change.
Common mistakes to avoid
- Skipping negative scores. Without deductions, models inflate and pass low-quality leads to sales.
- Overweighting low-value clicks. A single opened email should not move a lead much; reserve big points for high-intent actions like demo requests.
- Setting it and forgetting it. Thresholds and weights drift out of date; review them with sales on a schedule.
- Ignoring decay. Without decay on behavioral points, a stale burst of activity keeps cold leads artificially hot.
- No agreed MQL definition. If marketing and sales do not jointly own the threshold, the handoff breaks down.
Frequently asked questions
What is the difference between explicit and implicit lead scoring?
Explicit scoring rates who the lead is: job title, company size, industry, and other fit attributes they tell you. Implicit scoring rates what the lead does: page views, email opens, demo requests, and other engagement signals. Most models combine both, often on separate fit and interest axes.
What score makes a lead an MQL?
An MQL is any lead whose score crosses a threshold your team agrees on, such as 50 of 100 points. The right number is set by working backward from sales capacity and historical conversion data, then tuned so the leads passed actually convert. There is no universal cutoff.
Should I use negative scores in lead scoring?
Yes. Negative points keep poor-fit and disengaged leads from reaching your MQL threshold. Common deductions include personal email domains, competitors, students, unsubscribes, and long inactivity. Negative scoring is as important as positive scoring for keeping the bar honest and protecting sales time.
What is lead score decay?
Decay reduces a lead's behavioral score over time when they stop engaging, so an old burst of activity does not keep a cold lead artificially hot. It is usually applied only to the implicit/behavioral side, since fit attributes like company size do not fade the way interest does.
Last updated: 14 June 2026