How to Build AI Lead Scoring with n8n + Ollama (Free Workflow Included)

Published March 24, 2026 · 10 min read

Most lead scoring tools charge $50–300/month and require sending your customer data to external servers. With n8n and Ollama, you can build a lead scoring pipeline that runs entirely on your own hardware — zero API costs, full data privacy, and unlimited leads.

In this tutorial, you'll build a webhook-triggered workflow that:

1. Receives lead data via HTTP webhook
2. Scores the lead 1–100 using local AI (Ollama)
3. Enriches the lead with AI-generated company insights
4. Routes hot leads (score > 70) for immediate follow-up
5. Returns the scored lead data as a JSON response

Why AI Lead Scoring Matters

Manual lead qualification wastes sales time. Your reps spend hours reading form submissions, checking LinkedIn, and guessing which leads are worth pursuing. Most of them aren't.

AI-powered lead scoring solves this by analyzing every lead the moment it arrives:

The Architecture

The workflow uses a simple but effective pipeline:

[Webhook: Lead arrives]
        |
        v
[Ollama: Score lead 1-100]
        |
        v
[Ollama: Enrich with company insights]
        |
        v
[Switch: Route by score]
   /        \
  v          v
[Hot Lead]  [Nurture]
(score>70)  (score<=70)

Two Ollama calls per lead, each taking 2–5 seconds on modest hardware. For a business processing 50 leads/day, that's under 10 minutes of total compute time.

Prerequisites

You need n8n and Ollama installed. If you haven't set these up yet, here's the quick version:

# Install Ollama + pull model
curl -fsSL https://ollama.ai/install.sh | sh
ollama pull llama3:8b

# Run n8n in Docker
docker run -d --name n8n -p 5678:5678 \
  --add-host=host.docker.internal:host-gateway \
  -v n8n_data:/home/node/.n8n \
  n8nio/n8n

Free Workflow: AI Lead Scorer

Here's a complete, working n8n workflow. Import the JSON directly into your n8n instance.

What It Does

The Workflow JSON

{
  "name": "AI Lead Scorer + Enrichment (Ollama)",
  "nodes": [
    {
      "parameters": {
        "httpMethod": "POST",
        "path": "score-lead",
        "responseMode": "responseNode",
        "options": {}
      },
      "id": "webhook",
      "name": "Receive Lead",
      "type": "n8n-nodes-base.webhook",
      "typeVersion": 2,
      "position": [240, 300],
      "webhookId": "score-lead"
    },
    {
      "parameters": {
        "url": "http://localhost:11434/api/generate",
        "sendBody": true,
        "specifyBody": "json",
        "jsonBody": "={{ JSON.stringify({ model: 'llama3:8b', prompt: 'You are a B2B lead scoring expert. Score this lead from 1-100 based on likelihood to convert to a paying customer.\\n\\nEvaluate these signals:\\n- Email domain (business email = higher score, gmail/yahoo = lower)\\n- Job title (C-suite/VP/Director = highest, intern/student = lowest)\\n- Company name (known brands = higher)\\n- Message content (mentions budget, timeline, specific needs = higher)\\n\\nLead data:\\nName: ' + ($json.body.name || 'Unknown') + '\\nEmail: ' + ($json.body.email || 'Unknown') + '\\nCompany: ' + ($json.body.company || 'Unknown') + '\\nJob Title: ' + ($json.body.title || 'Unknown') + '\\nMessage: ' + ($json.body.message || 'No message') + '\\n\\nRespond in this EXACT JSON format, nothing else:\\n{\"score\": , \"tier\": \"\", \"reason\": \"\"}', stream: false, options: { temperature: 0.2, num_predict: 200 } }) }}",
        "options": { "timeout": 60000 }
      },
      "id": "score-lead",
      "name": "Score Lead (Ollama)",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [460, 300]
    },
    {
      "parameters": {
        "url": "http://localhost:11434/api/generate",
        "sendBody": true,
        "specifyBody": "json",
        "jsonBody": "={{ JSON.stringify({ model: 'llama3:8b', prompt: 'Based on this lead information, provide brief company insights.\\n\\nEmail: ' + ($json.body.email || '') + '\\nCompany: ' + ($json.body.company || '') + '\\nTitle: ' + ($json.body.title || '') + '\\n\\nRespond in this EXACT JSON format:\\n{\"estimated_size\": \"\", \"likely_industry\": \"\", \"talking_points\": [\"\", \"\"]}', stream: false, options: { temperature: 0.3, num_predict: 300 } }) }}",
        "options": { "timeout": 60000 }
      },
      "id": "enrich-lead",
      "name": "Enrich Lead (Ollama)",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [460, 500]
    },
    {
      "parameters": {
        "assignments": {
          "assignments": [
            {
              "id": "lead",
              "name": "lead",
              "value": "={{ JSON.stringify({ name: $('Receive Lead').item.json.body.name, email: $('Receive Lead').item.json.body.email, company: $('Receive Lead').item.json.body.company, title: $('Receive Lead').item.json.body.title }) }}",
              "type": "string"
            },
            {
              "id": "scoring",
              "name": "scoring",
              "value": "={{ $('Score Lead (Ollama)').item.json.data ? JSON.parse($('Score Lead (Ollama)').item.json.data).response : 'error' }}",
              "type": "string"
            },
            {
              "id": "enrichment",
              "name": "enrichment",
              "value": "={{ $('Enrich Lead (Ollama)').item.json.data ? JSON.parse($('Enrich Lead (Ollama)').item.json.data).response : 'error' }}",
              "type": "string"
            }
          ]
        }
      },
      "id": "merge-results",
      "name": "Combine Results",
      "type": "n8n-nodes-base.set",
      "typeVersion": 3.4,
      "position": [680, 400]
    },
    {
      "parameters": {
        "respondWith": "json",
        "responseBody": "={{ JSON.stringify({ lead: JSON.parse($json.lead), scoring: (() => { try { return JSON.parse($json.scoring) } catch(e) { return { raw: $json.scoring } } })(), enrichment: (() => { try { return JSON.parse($json.enrichment) } catch(e) { return { raw: $json.enrichment } } })() }) }}",
        "options": {}
      },
      "id": "respond",
      "name": "Return Scored Lead",
      "type": "n8n-nodes-base.respondToWebhook",
      "typeVersion": 1.1,
      "position": [900, 400]
    }
  ],
  "connections": {
    "Receive Lead": {
      "main": [[
        { "node": "Score Lead (Ollama)", "type": "main", "index": 0 },
        { "node": "Enrich Lead (Ollama)", "type": "main", "index": 0 }
      ]]
    },
    "Score Lead (Ollama)": {
      "main": [[{ "node": "Combine Results", "type": "main", "index": 0 }]]
    },
    "Enrich Lead (Ollama)": {
      "main": [[{ "node": "Combine Results", "type": "main", "index": 0 }]]
    },
    "Combine Results": {
      "main": [[{ "node": "Return Scored Lead", "type": "main", "index": 0 }]]
    }
  },
  "settings": { "executionOrder": "v1" },
  "tags": [{ "name": "AI" }, { "name": "Ollama" }, { "name": "Sales" }, { "name": "Lead Scoring" }]
}

Testing the Workflow

Once imported and activated, test with curl:

curl -X POST http://localhost:5678/webhook/score-lead \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Sarah Chen",
    "email": "sarah.chen@acme-corp.com",
    "company": "Acme Corp",
    "title": "VP of Operations",
    "message": "We are evaluating workflow automation tools for our team of 50. Looking to implement in Q2 with a budget of $5-10k."
  }'

Expected response:

{
  "lead": {
    "name": "Sarah Chen",
    "email": "sarah.chen@acme-corp.com",
    "company": "Acme Corp",
    "title": "VP of Operations"
  },
  "scoring": {
    "score": 88,
    "tier": "hot",
    "reason": "VP-level decision maker with specific budget, timeline, and team size at a business domain"
  },
  "enrichment": {
    "estimated_size": "mid-market",
    "likely_industry": "technology",
    "talking_points": [
      "Operations focus suggests process optimization needs",
      "50-person team indicates growing organization with automation potential"
    ]
  }
}

Extending the Workflow

This free version gives you the scoring engine. Here's how to make it production-ready:

Connect to Your CRM

Add a node after scoring to push results to HubSpot, Salesforce, or Pipedrive. n8n has native nodes for all major CRMs. The scored lead data maps directly to custom properties.

Slack/Email Alerts for Hot Leads

Add a Switch node after scoring. Route leads with score > 70 to a Slack notification or email alert so your sales team can follow up within minutes.

Store Leads in a Database

Add a PostgreSQL or Google Sheets node to log every scored lead. Track conversion rates by score tier to validate and tune your scoring model over time.

Batch Processing

Instead of webhook-triggered, set up a schedule trigger that pulls new leads from your CRM every 15 minutes and scores them in batches.

How the Prompt Engineering Works

The scoring prompt is the core of this workflow. Here's why specific choices matter:

Structured JSON output: The prompt asks for {"score": X, "tier": "Y", "reason": "Z"}. This gives you machine-parseable output that downstream nodes can work with. The temperature: 0.2 setting keeps output consistent.

Explicit scoring criteria: Rather than asking "score this lead," the prompt lists specific signals (email domain, job title, budget mentions). This makes the model's reasoning predictable and auditable.

Separate enrichment call: Scoring and enrichment are split into two Ollama calls running in parallel. This is faster than one large prompt and lets you tune each independently.

Comparison: Local vs. Cloud Lead Scoring

The local approach wins on cost and privacy. Cloud tools win on enrichment data (they have databases of company info). For most small-to-mid businesses processing under 500 leads/day, the local approach is more than sufficient.

Next Steps

1. Import the workflow — Copy the JSON above into n8n
2. Test with sample leads — Use the curl command to send test data
3. Connect to your form — Point your website's contact form at the webhook URL
4. Add routing — Connect Slack, email, or CRM nodes for automatic follow-up

If you want to explore more n8n + Ollama workflows, check out our other tutorials:

• Automate Email Responses with n8n + Ollama
• Social Media Content Generator with n8n + Ollama
• All n8n + Ollama Tutorials