Scoring Inbound vs Outbound Leads

Inbound and outbound leads arrive with fundamentally different evidence, so scoring them identically distorts both. An inbound lead has already raised a hand and carries fresh intent, while an outbound lead was chosen by you and starts with fit but little expressed intent. A single blended score buries a high-intent inbound demo request next to a cold prospect who merely matches your ICP. Treating scoring like code means modeling fit and intent as separate, observable components rather than one opaque number.

The cleaner model separates fit from intent and recombines them per motion. Fit comes from firmographics enriched by Clearbit, Apollo, or Cognism, while intent comes from behavior like product usage in Koala, a pricing-page visit caught by Warmly, or an email reply. For inbound, intent is already present so fit becomes the prioritizer; for outbound, you are manufacturing intent so fit and signal timing drive sequencing. Owning both dimensions explicitly lets you reason about why a lead scored the way it did.

The mistake is building two disconnected scoring systems that never reconcile. Allbound works because inbound, outbound, paid, and content all read from one shared signal and identity graph, even if they apply different scoring lenses. A contact enriched in Clay and tracked across product, email, and web should carry one identity whether marketing or an SDR touched them first. HubSpot or Salesforce becomes the single source where both the fit score and the intent score live on the same record.

With a shared graph, the two lenses stay coherent. An outbound prospect who suddenly visits your pricing page should have their intent score jump and flip into a fast-follow play, using the same signals an inbound lead would trigger. Smartlead and Instantly can adapt outbound sequencing based on intent the prospect starts showing mid-campaign. Because the scoring logic is versioned, you can test threshold changes for each motion independently while keeping the underlying graph unified.

Routing is where the split pays off, so make it deterministic. A high-fit, high-intent inbound lead routes straight to an AE with full context, while a high-fit, low-intent outbound account enters a nurturing sequence until intent appears. Low-fit inbound leads should not consume AE time regardless of intent, and that rule belongs in code, not in a rep's discretion. Clear, versioned routing rules keep both motions consistent and auditable.

Keep the system honest and compliant. Review where deals actually come from and rebalance the scoring weights, since a model that never gets tuned drifts away from reality. For EU contacts, ensure the behavioral signals feeding intent scores have a lawful basis under GDPR before they drive outreach. Measure conversion by score band for each motion so you can prove the split works and adjust thresholds with evidence rather than opinion.