A lead scoring framework for multi-market B2B SaaS

Lead scoring promises a number that summarizes how good a lead is. Most implementations produce a number that summarizes how often the rep updated the lead. The gap between the promise and the practice comes from the framework: without a clear structure, scoring drifts into theater.

Here is a framework that holds. It is not magic; it is structure that the team can apply, calibrate, and trust.

The four dimensions

Every workable lead score has four dimensions. Each contributes a weighted sub-score; the total is the lead's score.

Dimension 1: Firmographic fit. How well does the lead's company match your ICP (Ideal Customer Profile)? Inputs: industry, employee count, revenue tier, tech stack signals, geographic market.

Dimension 2: Behavioral engagement. How actively has the lead engaged with you? Inputs: form submissions, webinar attendance, content downloads, repeat visits, demo requests.

Dimension 3: Intent signal. External signals of buying intent. Inputs: third-party intent data (Bombora, G2 intent, content engagement on review sites), competitive-research patterns, hiring signals.

Dimension 4: Recency. How recent is the engagement? Inputs: time since last interaction, time since first touch, velocity of recent activity.

The four dimensions cover the meaningful axes. Adding a fifth dimension usually produces redundancy without information.

Initial weights

Default weights for a typical B2B SaaS:

• Firmographic fit: 30%
• Behavioral engagement: 30%
• Intent signal: 25%
• Recency: 15%

These are starting points, not gospel. Calibrate based on your own data.

For organizations heavily reliant on outbound, behavioral engagement weight goes down (less inbound behavior to score) and firmographic fit goes up.

For organizations in mature, saturated categories where intent signals are strong, intent weight goes up.

For organizations with long sales cycles, recency weight goes down (a touch six months ago can still be relevant).

Computing each sub-score

Each dimension's sub-score is computed on a 0 to 100 scale.

Firmographic fit (0-100):

• 100 if all ICP criteria match (industry, size, revenue, geography).
• 70 if most match.
• 40 if some match.
• 0 if none.

Behavioral engagement (0-100):

• 100 if recent demo request or pricing-page visit.
• 70 if multiple content downloads or webinar attendance.
• 40 if single content download.
• 0 if no engagement.

Intent signal (0-100):

• 100 if third-party intent data shows strong category interest.
• 70 if moderate intent signal.
• 40 if minor signal.
• 0 if none.

Recency (0-100):

• 100 if engagement within last 7 days.
• 80 if within 30 days.
• 50 if within 90 days.
• 20 if older.
• 0 if no engagement.

The weighted sum produces the final score, 0 to 100.

Routing by score

The score drives prioritization, not assignment:

• 80+: hot leads. Immediate routing to senior reps. SLA at the tighter end.
• 60-79: warm leads. Standard routing. Normal SLA.
• 40-59: cool leads. Routed to nurture motion. Longer SLA, may not warrant active outbound.
• Below 40: cold leads. Marketing-nurture only. No sales engagement until score rises.

The thresholds calibrate against your actual conversion rates. Adjust based on what historical data shows.

Multi-market variation

In multi-market organizations, the same framework applies but the weights differ:

• Markets with mature inbound motion (US, UK): behavioral engagement weight higher.
• Markets with outbound-heavy motion (India, GCC): firmographic fit weight higher.
• Markets with limited intent-data coverage (some APAC markets): intent signal weight lower.

The structure (four dimensions) is the same across markets. The weights vary. This lets the framework adapt to local conditions without diverging into separate frameworks.

The calibration loop

Lead scoring is not set-and-forget. Quarterly calibration:

1. Pull closed-won deals from the last quarter. What was their lead score at the moment they entered QUALIFIED?
2. Pull closed-lost deals. What was theirs?
3. Compare distributions. Are closed-won leads scoring higher than closed-lost? By how much?
4. Identify mismatches. Which dimensions correlate most with closed-won? Adjust weights.
5. Identify gaps. Which leads closed despite low scores? What did the score miss?

This is 2-3 hours of analysis per quarter. It keeps the framework calibrated against reality.

What the framework explicitly does not do

A few honest non-features:

No ML model. Sophisticated lead scoring uses ML to learn weights from data. ML is great but requires data volume most organizations do not have. The four-dimension framework with quarterly calibration produces 80% of the value of ML scoring with 5% of the operational complexity.

No magic formula. The framework is structure, not formula. The team has to understand the structure and apply it. A black-box score the team does not understand produces theater.

No score-locked routing. The score informs prioritization but does not lock routing. Other factors (territory, capacity, specialization) still apply. Routing rules consult the score among other inputs.

Operationalizing the framework

Three things to do:

1. Document the framework. Write down the four dimensions, the inputs to each, the scoring scale, the current weights. Share with sales operations and the reps who will see the scores.

2. Configure the score in the platform. The lead intelligence platform computes the score from the inputs. The score is a lead attribute, visible on the lead record and queryable in reports.

3. Schedule the calibration meeting. Quarterly, 90 minutes. Pull the data, review the distributions, adjust weights, document the changes.

This is operational discipline. The framework alone does nothing; the discipline of using and calibrating it produces the value.

For how MegatronLead supports score-as-an-attribute and score-informed routing, see workflow automation.