AI Lead Qualification & CRM Entry
n8n workflow that scores inbound property management leads 1–10 using Claude AI — then routes hot leads directly into HubSpot CRM with a personalized email and Slack alert, while warm and cold leads enter a nurture sequence. Zero manual triage.
Overview
A webhook-triggered n8n automation that eliminates manual lead triage for property management companies. The moment a lead submits a form, Claude AI evaluates their portfolio size, budget, urgency, and service fit against a defined scoring rubric — returning a structured JSON response with a 1–10 score, tier classification (hot/warm/cold), reasoning, key signals, and a personalization hook for outreach. Hot leads (score ≥7) are automatically created as contacts in HubSpot CRM, receive a personalized email referencing their specific pain points, and trigger a Slack alert to the sales team. Warm and cold leads receive a value-add nurture email. All leads regardless of tier are logged to a Google Sheets audit trail with their score, urgency, and recommended next step.
Problem
Property management companies spend 20–30 minutes manually reviewing and qualifying each inbound lead — reading form submissions, estimating fit, deciding who to call first, and drafting follow-up emails. At 20–40 leads per month, that's 8–20 hours of repeatable, judgement-intensive work that follows the same criteria every time. The best leads often wait hours or days for a response while the team prioritizes manually. Fast follow-up is one of the strongest predictors of conversion — and most PM companies are losing deals to slower processes.
Approach
Webhook trigger receives the lead payload from any form tool (Typeform, Webflow, HubSpot Forms). A Parse Lead Data Code node normalizes and extracts all fields. A Build Scoring Prompt Code node formats the lead as a structured context table and constructs the full Claude messages array with a precise 10-point scoring rubric. An HTTP Request node sends to OpenRouter (Claude 3.5 Sonnet) and receives a structured JSON response. A Parse AI Score Code node extracts the score, tier, reasoning, key signals, and personalization hook — with JSON parse error fallback. An IF node branches on score ≥7. Hot path: HTTP Request creates a HubSpot contact via the CRM v3 API with lead data and AI-generated fields (hs_lead_status, lead_score, lead_tier, reasoning as notes). Gmail sends a warm, personalized email referencing Claude's personalization hook. Slack posts a rich alert with the score, signals, and recommended next step. Google Sheets logs the full run. Cold/warm path: Gmail sends a value-add nurture email with a soft CTA. Google Sheets logs independently. Error Trigger node routes failures to a Slack error alert.
Tech Stack
Results
3/3 test scenarios passing end-to-end: Sarah Chen (28 units, $4,500/mo) scored 9/10 hot — HubSpot contact created, personalized email sent, Slack alerted, Sheets logged. Marcus Reeves (9 units, $1,800/mo) scored 6/10 warm — nurture email delivered, Sheets logged. Tom Nguyen (2 units, $500/mo) scored 3/10 cold — nurture sequence triggered. Claude reasoning was accurate and detailed across all three tiers. Execution time: 6.6–8.2 seconds per lead. HubSpot integration is live and ready — swap placeholder token for a real Private App key to activate CRM sync. Eliminates 20–30 minutes of manual triage per lead for property management companies processing 20–40 leads/month.