ICP lead scorer

Scores inbound leads and accounts against the written ICP definition with a stated reason code per score — firmographic fit, use-case signal, engagement quality — and routes by band. Use when the user mentions lead scoring, ICP fit, inbound qualification, MQL routing, "is this lead any good," or wants the inbound queue prioritized.

Every scoring model eventually becomes a number nobody trusts attached to a lead nobody routes right. The fix is not a better black box — it's reasons. This skill scores against the ICP you actually wrote down, shows its work on every lead, and routes by band, so sales argues with the criteria instead of the mystery.

When to run

Continuous on new inbound (recommended); batch: "rescore this quarter's MQLs"
After an ICP definition update: full-book rescore with delta report

Computer scopes

Docs: the written ICP definition (the rulebook — no rulebook, no scoring; halt and say so)
CRM + marketing sync: lead/account records, source, engagement history
Enrichment sync: firmographics where connected
Web analytics sync: on-site behavior signals where connected

Workflow

Load the ICP rulebook version; every score is stamped with the rulebook version it used.
Score three lanes per lead: firmographic fit (segment, size, industry, geo), use-case signal (source content, stated pain, role), engagement quality (depth and recency).
Emit the composite with band (A/B/C/D) and the reason codes: which criteria passed, failed, or were unknown. Unknown is a real state — never silently score missing data as bad data.
Route by band per the routing table: A to owner assignment, B to nurture-fast, C/D to nurture-slow; enrichment-gap leads route to an enrichment queue instead of a grave.
Log every score for the calibration loop: monthly, compare bands to actual conversion and propose rulebook adjustments to a human — the model suggests, the ICP owner decides.

Guardrails

Autonomy: L2 for scoring and queue routing (reversible, audited). Owner assignment notifications follow the org's existing rules; this skill never emails leads.
Scores only on criteria in the written ICP — no invented signals, no demographic proxies.
The rulebook is versioned and human-owned; the skill proposes changes with evidence, never self-modifies.

Output

Per lead: score, band, reason codes, rulebook version, route taken. Monthly: calibration report (band vs. conversion, proposed adjustments with evidence).

Eval fixtures

Set: 150 labeled leads against a sandbox ICP rulebook, including 20 with missing fields and 10 designed edge cases; 6-month conversion outcomes for calibration test.
Rubric: band agreement with gold; unknown-handling correctness on missing fields; reason codes cite real criteria; calibration report math correct.
Ship gate: band agreement ≥ 0.85; missing-data mishandling 0.

Run icp lead scorer on your own data.

MORE SKILLS

GTM Engineering & Marketing Ops · L0

ABM account brief

Builds a target-account research brief for ABM plays by merging internal history from GBrain (prior touches, closed-lost context, existing relationships) with external signals (news, hiring, tech stack, filings). Use when the user mentions account research, ABM targets, "brief me on this prospect," territory planning, or prepping outbound to a named account.

GTM Engineering & Marketing Ops · L0

ABM account brief

Builds a target-account research brief for ABM plays by merging internal history from GBrain (prior touches, closed-lost context, existing relationships) with external signals (news, hiring, tech stack, filings). Use when the user mentions account research, ABM targets, "brief me on this prospect," territory planning, or prepping outbound to a named account.

GTM Engineering & Marketing Ops · L0

Competitor diff watch

Monitors competitor websites, changelogs, pricing pages, and sitemaps for changes and produces a weekly diff digest — new pages, messaging shifts, feature launches, pricing moves — with before/after evidence. Use when the user mentions competitor monitoring, competitive intel, "what did [competitor] just launch," pricing change alerts, or tracking competitor content strategy.

GTM Engineering & Marketing Ops · L0

Competitor diff watch

Monitors competitor websites, changelogs, pricing pages, and sitemaps for changes and produces a weekly diff digest — new pages, messaging shifts, feature launches, pricing moves — with before/after evidence. Use when the user mentions competitor monitoring, competitive intel, "what did [competitor] just launch," pricing change alerts, or tracking competitor content strategy.