Evidence-Based Lead Prioritizer

Convert a mixed CSV of authorized B2B inquiries into a transparent sales review queue. Every point is tied to an input field and a configurable rule.

Three-Step Workflow

1. 1. Define the ideal customer profile in examples/icp.json: target industries, regions, company sizes, decision roles, need keywords, score weights and tier thresholds.
2. Run the local script or import the n8n workflow. The pipeline normalizes fields, removes duplicate organizations, calculates deterministic scores, lists evidence and missing fields, and proposes a next action.
3. Let an authorized salesperson verify the evidence, lawful contact basis and CRM record before approving follow-up. The included workflow does not scrape contacts or send messages.

Quick Start

CODEBLOCK0

Outputs:

• - scored_leads.csv: ranked review queue with score, tier, confidence and next action.
• INLINECODE2: complete factor-by-factor scoring evidence.
• INLINECODE3: human-readable priority queue.
• INLINECODE4: retained and skipped organization records.

Rules

• - Process only inquiries the organization received lawfully or is authorized to review.
• Do not scrape email addresses, phone numbers, social profiles or personal data.
• Do not infer protected traits, wealth, health, beliefs or other sensitive attributes.
• Treat missing information as missing; never invent enrichment or certainty.
• A score ranks review priority only. It must not create contracts, reject regulated applicants, set prices or make eligibility decisions.
• Require human approval before outreach, CRM stage changes, deal creation, scheduling or any external action.

Read references/usage.md for schemas and recovery. Read references/scoring-rubric.md before changing weights.