Claude Code n8n Lead Generation Engine

This workflow connects Claude Code to n8n via the n8n-mcp MCP server so that Claude Code builds and deploys a complete lead generation pipeline in minutes. Claude Code uses its MCP mode to call create_workflow, update_workflow, and execute_workflow tools against the n8n instance. The resulting n8n workflow runs on a schedule, scraping Reddit and Twitter for buying-intent signals using keyword matching on phrases like 'recommend', 'alternatives to', and 'looking for'. Raw mentions flow into an Apify actor for enrichment — Apify pulls company profile data, contact details, and recent activity for each detected lead. Enriched leads enter a Google Sheet where a Code node deduplicates against existing entries by email, username, and company domain. New leads trigger a Slack notification with a structured card showing the lead name, source, key intent signal, and enrichment summary. Claude Code uses a 2-phase model: reasoning mode to generate the workflow JSON definition and MCP mode to deploy it live to n8n. Build time drops from 45 minutes of manual drag-and-drop node configuration to 10 minutes of AI-assisted creation.

BUSINESS PROBLEM

Sales development representatives spend 40-50% of their time on lead research — finding prospects, verifying fit, and gathering context before outreach. For a team of 5 SDRs, that is 80-100 hours per week of research time that could go toward actual selling. According to HubSpot's 2025 Sales Efficiency Report, sales teams using intent data see 2.3x higher conversion rates than teams relying solely on demographic targeting. The core problem is that buying-intent signals are scattered across Reddit threads, Twitter conversations, industry forums, and review sites. Manually monitoring these sources at scale is impractical. Most teams settle for outdated lead lists with 10-20% accuracy. Claude Code connected to n8n via the MCP protocol solves this by generating a complete workflow that monitors multiple social sources simultaneously, enriches each lead with company and contact data, deduplicates automatically, and alerts the team in Slack — all without dragging a single node in the n8n editor.

WHO BENEFITS

FOR SDR team leads at B2B SaaS companies with 5-20 reps SITUATION: Your team spends 15+ hours weekly on manual lead research from social signals. PAYOFF: Automated pipeline captures 100+ buying-intent leads per week. Reps focus on outreach, not research.

FOR outbound sales managers at agencies running multi-channel prospecting SITUATION: Reddit mentions, tweets, and forum signals are impossible to monitor manually across 5+ channels. PAYOFF: n8n scrapes all sources in parallel. Apify enriches every lead. Slack delivers ready-to-contact leads.

FOR freelance sales consultants managing their own pipeline SITUATION: You cannot afford a full SDR team but need consistent lead flow. PAYOFF: One 10-minute Claude Code session builds a 24/7 lead gen pipeline.

HOW IT WORKS

1. Source Configuration (Claude Code MCP — 2 min) Input: Natural language prompt with target keywords, subreddits, and Twitter accounts Action: Claude calls create_workflow tool in n8n, adds HTTP Request nodes for Reddit and Twitter API Output: Workflow skeleton with Reddit and Twitter source nodes

2. Intent Filter Setup (Claude Code MCP — 30 sec) Input: Intent keyword list including 'recommend', 'looking for', 'alternatives to', 'evaluating' Action: Claude adds IF node with keyword matching logic against post titles and bodies Output: Filtered mention stream passing only posts containing buying-intent phrases

3. Apify Enrichment (Claude Code MCP — 30 sec) Input: Raw mention with username, source URL, and post content Action: Claude adds Apify node configured with LinkedIn profile scraper or company enricher actor Output: Enriched lead record with company name, contact title, employee count, recent activity

4. Google Sheets Logging (Claude Code MCP — 30 sec) Input: Enriched lead data schema with 12 fields Action: Claude adds Google Sheets node with column mapping and dedup key configuration Output: Append-only lead sheet with status tracking columns and timestamps

5. Deduplication Logic (Claude Code MCP — 1 min) Input: New lead fields compared against existing sheet rows Action: Claude writes a Code node that checks email, username, and company domain for matches Output: Lead tagged as new or duplicate with match reference ID

6. Slack Notification (Claude Code MCP — 30 sec) Input: New lead object with enrichment data Action: Claude adds Slack node with structured message block — name, source, intent signal, enrichment summary Output: Slack message posted to #leads channel with actionable card

7. Schedule Setup (n8n UI manual — 1 min) Input: n8n workflow timer node Action: Set cron schedule in n8n UI for every 4 hours Output: Automated execution on timer trigger

TOOL INTEGRATION

n8n v1.80+ Role: Workflow execution engine Install: npx n8n for self-hosted or n8n.cloud account Config step: Enable MCP via Settings > MCP > toggle Instance-level MCP. Generate an access token. Copy it immediately — n8n shows it only once. Gotcha: The MCP server will not load until the n8n instance is fully restarted. A stop and start is required, not just a reload. This is the most common cause of first-time setup failure.

Claude Code v2.1.154+ Role: AI workflow builder — generates, deploys, and iterates on n8n workflows using natural language Install: npm install -g @anthropic-ai/claude-code Config step: Connect to n8n via: claude mcp add n8n-mcp -e MCP_MODE=stdio -e LOG_LEVEL=error -e DISABLE_CONSOLE_OUTPUT=true -e N8N_API_URL=https://your-instance.com -e N8N_API_KEY=your-key -- npx n8n-mcp MCP tools: list_workflows, get_workflow, create_workflow, update_workflow, delete_workflow, execute_workflow, get_execution Gotcha: Claude Code cannot set credentials in n8n API key fields — Google Sheets, Slack, and Apify credentials must be configured manually in the n8n UI before the workflow runs.

Apify Role: Data enrichment — pulls company profiles, contact details, and social data Config step: Apify account, API token stored in n8n credentials Gotcha: Free tier limited to 5 actor runs per day. Production volumes require a paid plan.

Slack Role: Alert delivery channel Config step: Slack workspace, n8n Slack app integration with OAuth

Google Sheets Role: Lead database with dedup Config step: Google API credentials in n8n Gotcha: Row limits apply at 10M cells. Archive leads monthly to keep sheet performant under 10K rows.

ROI METRICS

1. Workflow build time: 45 minutes manual node dragging to 10 minutes with Claude Code MCP (Ability.ai, 2026)
2. Lead capture throughput: 10-15 manual leads per week to 100+ automated leads per week
3. SDR research recovery: 15 hours per week per rep to zero hours on research
4. Conversion uplift: Intent-signal leads convert at 2.3x versus demographic-only targeting (HubSpot 2025 Sales Efficiency Report)
5. First-7-day win: First 20 enriched leads delivered to Slack within 24 hours of deployment

CAVEATS

1. (moderate risk) Credential configuration required: Claude Code creates all nodes but cannot inject API keys into n8n credential fields. Budget 5 extra minutes for manual credential entry.
2. (moderate risk) MCP server restart requirement: The n8n MCP server will not load until the instance is fully restarted — stop, start, not reload.
3. (minor risk) False positive intent signals: Keyword matching catches noise alongside real signals. Review Slack alerts before first contact.
4. (minor risk) API rate limits: Twitter API v2 and Reddit API enforce rate caps at high scraping volumes.