lead-scoring

Lead Scoring

Quantify how likely a lead/account is to buy (fit) and how actively they show intent (engagement), then route the highest-probability records to sales. This skill covers the scoring math, qualification frameworks, CRM + warehouse implementation, calibration against real outcomes, and privacy/compliance.

For the broader lifecycle (stages, conversion-rate analysis, pipeline math) see the sibling sales-funnel skill. Use this skill for the scoring/qualification layer that feeds those stages.

Scoring prioritizes outreach; it does not replace human qualification. A high score means "call sooner," not "close the deal." Discovery (BANT/MEDDIC) confirms what the score predicted.

Scoring Model Design

Two-Axis Model

Score on two independent axes so a great-fit-but-cold account isn't confused with a poor-fit tire-kicker who clicks everything:

1. Fit Score (0–100): how well they match your ICP (firmographic/demographic). Mostly static; changes on enrichment or job change.
2. Engagement Score (0–100): how actively they show buying intent (behavioral). Time-sensitive; decays.

Both axes are hard-capped at 100. Compute raw points, then clamp:

fit        = min(100, sum(fit_points))
engagement = min(100, sum(engagement_points_after_decay_and_dedup))
total      = round(0.4 * fit + 0.6 * engagement)   # 0–100

The 40/60 weighting favors intent over fit — flip toward fit (e.g., 60/40) in long, committee-driven enterprise sales where firmographics predict more than clicks. Calibrate the weights against won/lost data (see Calibration); do not ship the default blindly.

Use a grade × score matrix, not a single number, for routing. Collapsing fit and engagement into one total hides the most important quadrant. Route on the 2x2 below and keep total only as a tiebreaker/sort key.

	Low engagement (<40)	High engagement (≥60)
High fit (≥60)	Nurture, account-based ads (A2 / "right fit, not ready")	Hot — alert AE, SLA timer (A1)
Low fit (<40)	Disqualify / low-touch (D)	Reroute or self-serve; investigate why low-fit is so active (could be a competitor, student, or job seeker)