Identify 'Hot' Prospects from Metadata

Predictive Lead Intelligence Engine Multi-Signal Scoring Model AI builds a composite score from: Fit Score (firmographic match to ICP: industry, company size, job title) + Engagement Score (behavioral: page views, content downloads, email engagement) + Intent Score (3rd-party: G2 research, review site visits, competitor evaluations). Buying Committee Detection When 3+ people from the same company show elevated engagement within a 14-day window, AI flags it as 'Buying Committee Activity' (BCA). This is the strongest buying signal — individual engagement

is noise, coordinated engagement is intent. Journey Stage Mapping Based on content consumption patterns, AI classifies each lead: Awareness (blog readers), Consideration (case study/webinar), Decision (pricing/demo page), Evaluation (technical docs/API reference). Sales receives leads with stage context for appropriate outreach. Score Decay & Reactivation Lead scores decay over time: -5 points per week of inactivity. AI auto-moves decayed leads back to nurture sequences. If a previously cold lead re-engages, it triggers a 'Re-activation Alert' to Sales

with the specific trigger event. Closed-Loop Attribution AI tracks what marketing touchpoints preceded every closed-won deal. Uses this data to continuously refine scoring weights: if 'Technical Webinar' attendance correlates with 3x higher close rates, it gets weighted more heavily in future scoring. Define your MQL (Marketing Qualified Lead) threshold jointly with Sales — AI can optimize the threshold based on historical conversion data to maximize SQL acceptance rates. Track 'negative signals' equally: unsubscribes, competitor job titles, student

email domains, and repeat demo registrations with no-shows should subtract points. Implement a 'Speed to Lead' dashboard — leads contacted within 5 minutes of reaching MQL threshold convert at 9x the rate of those contacted after 24 hours. Don't score based solely on email opens — bots and preview panes inflate open rates. Weight clicks and time-on-page far more heavily. Don't treat all page views equally — a visit to 'Pricing' carries 10x the intent signal of a visit to 'About Us'. Don't set scores and forget — recalibrate the scoring model quarterly

using actual won/lost deal data to keep it predictive. The 'Dark Funnel' Layer 75% of the B2B buying journey happens in channels you can't track: peer conversations, private Slack groups, podcasts, and dark social. AI can layer intent data from providers like Bombora or 6sense to capture 'off-property' research signals that supplement your first-party engagement data — illuminating the dark funnel.