Lead scoring is a methodology for assigning numerical values to leads based on how closely they match the ideal customer profile and how actively they have engaged with the organization’s content and outreach. The resulting score helps sales and marketing teams prioritize which leads deserve immediate attention and which need further nurturing.
Combining fit and engagement into a composite score
Lead scoring combines two dimensions of evaluation: fit (who the lead is) and engagement (what the lead has done). Together, these dimensions produce a composite score that indicates sales readiness.
Fit scoring
Fit scoring evaluates whether the lead matches the attributes of the ideal customer. Common fit criteria include:
- Company size: Measured by revenue or employee count, with leads from companies in the target size range scoring higher.
- Industry: Leads from target verticals receive higher scores.
- Job title and role: Decision-makers and economic buyers score higher than individual contributors or researchers.
- Geography: Leads in served regions score higher than those in regions without sales coverage or partner presence.
- Technology stack: For technology vendors, leads from companies running compatible or complementary platforms receive a score boost.
Engagement scoring
Engagement scoring measures how the lead has interacted with the organization. Each action is assigned a point value reflecting its intent signal:
- High-intent actions: Requesting a demo, visiting the pricing page, or attending a live product webinar carry the highest point values.
- Medium-intent actions: Downloading a solution brief, attending an educational webinar, or visiting the product page multiple times.
- Low-intent actions: Opening an email, visiting the blog, or following on social media.
- Negative signals: Unsubscribing from emails, visiting the careers page (indicating a job seeker rather than a buyer), or prolonged inactivity can subtract points.
Composite scoring
The fit score and engagement score are combined (through addition, weighted formulas, or matrix models) to produce an overall lead score. When the score crosses a predefined threshold, the lead is classified as a marketing qualified lead (MQL) and routed to sales or distributed to a partner.
Prioritizing sales effort through scoring
Without scoring, all leads are treated equally. A VP of IT at a mid-market company who requested a demo sits in the same queue as a student who downloaded a white paper for a class project, and sales reps waste time sorting through unqualified contacts while high-potential leads go cold because they were not prioritized.
Lead scoring solves this by creating a triage system. High-scoring leads get immediate attention, medium-scoring leads enter lead nurturing tracks, and low-scoring leads are deprioritized or disqualified.
In channel programs, lead scoring determines which leads are distributed to partners and when. Distributing under-scored leads frustrates partners and erodes trust in the vendor’s lead distribution program, while holding leads too long while waiting for higher scores delays follow-up and risks losing the prospect. A well-calibrated scoring model balances these competing pressures.
Scoring models, calibration, and maintenance
Scoring model example
| Action or attribute | Points |
|---|---|
| Job title: VP or C-level | +20 |
| Company size: 200-5,000 employees | +15 |
| Industry: target vertical | +10 |
| Requested a demo | +30 |
| Visited pricing page | +20 |
| Downloaded a solution brief | +10 |
| Opened a nurture email | +2 |
| No activity for 30 days | -15 |
| Visited careers page | -20 |
| MQL threshold | 75 points |
Scoring model types
- Points-based models: The most common approach, where each attribute and action contributes a fixed number of points. These are straightforward to build and explain.
- Grade-and-score models: Fit and engagement are tracked separately, with fit producing a letter grade (A through D) and engagement producing a numerical score. A lead must meet minimum thresholds on both dimensions to qualify.
- Predictive models: Machine learning analyzes historical conversion data to identify patterns that predict which leads are most likely to close. These models can surface non-obvious correlations but require sufficient data volume to train.
Calibration and maintenance
Lead scoring models degrade over time if they are not maintained. Calibration involves:
- Conversion analysis: Regularly comparing the scores of leads that converted vs. those that did not. If high-scoring leads convert at the same rate as low-scoring leads, the model needs adjustment.
- Threshold review: The MQL threshold should be validated quarterly. If sales teams report that MQLs are not qualified, the threshold may need to increase; if partners complain about insufficient lead flow, it may be too high.
- Score decay: Adding time-based decay so that old engagement signals lose value. A prospect who visited the pricing page six months ago is less likely to be in-market than one who visited yesterday.
- Feedback loops: Sales and partner feedback on lead quality should feed back into scoring model updates. This closed-loop process is the single most important factor in maintaining a useful scoring model.