a6e5b969-be6e-4756-ae53-47d3cb899ca4.png

23f5031c-b88b-4ce2-8795-61273b79b17e.jpg

Not all leads are created equal. Without a consistent scoring model, teams risk wasting time on low-quality leads while high-potential prospects slip through the cracks. The Lead Scoring Copilot solves this by analyzing lead attributes and behaviors against historical MQLs and closed-won opportunities, producing both a numerical score and a clear next step.

The agent accepts `lead_json` including attributes such as source, channel, company size, industry, geography, recent touches, and web intent. It calculates a conversion **score** (0–1) and assigns a **segment** (e.g., A, B, or C) based on thresholds. It also outputs a **next_action** such as *route to SDR*, *add to nurture*, or *request enrichment*. Optionally, it can generate a plain-English **reason string** to provide additional context for the score, while keeping thresholds and logic coded for consistency.

By embedding intelligence into lead routing, the Lead Scoring Copilot improves conversion rates, accelerates response times, and ensures sales teams focus on the leads most likely to close.

The Lead Scoring Agent is designed to evaluate inbound leads, assign a conversion score, and recommend the next best action for sales and marketing teams.