How We Built a Full Lead Intelligence Pipeline Using n8n + Zoho + LLMs

When we needed to scale SingularRarity Labs' own outreach, we didn't hire a sales team. We built a fully autonomous lead intelligence pipeline in 2 weeks for ₹0 recurring cost.

This is the complete technical breakdown of how it works — architecture, code, results, and everything you need to replicate it.

The Problem

Manual outreach doesn't scale:

615 leads → Research each manually: 3 min/lead = 30 hours
Personalize 615 emails: 5 min/email = 51 hours
Track opens + follow-ups: 2 hours/day
Total: 83+ hours for one outreach cycle

We needed this to run itself while we built client systems.

The Full Architecture

[Google Maps Scraper] ──────────────────────────┐
[LinkedIn Sales Nav Export] ──────────────────── →  [Master CSV: 615 leads]
[Manual Additions] ─────────────────────────────┘
                                ↓
                    [n8n Workflow Orchestrator]
                         ↓           ↓
              [LLM Enrichment]  [Segment Classifier]
                         ↓           ↓
              [Email Personalizer] [Score Engine]
                         ↓
              [Zoho SMTP Sender: info@singularraritylabs.com]
                         ↓
              [Zoho CRM: Lead created + scored]
                         ↓
              [n8n Follow-up Scheduler]

Phase 1: Lead Collection (Week 1)

Source 1: LinkedIn (130 verified emails)

Target: AI/Tech/SaaS/VC decision makers
Tools: Sales Navigator CSV export
Fields: Name, Title, Company, Email, LinkedIn URL
Cost: ₹0 (existing subscription)

Source 2: Google Maps (485 companies)

Target: BPO, fintech, regional enterprises
Tools: Custom Python scraper
Fields: Company, Phone, Website, Category
Cost: ₹0 (self-built)

Python scraper (core logic):

import googlemaps

gmaps = googlemaps.Client(key=MAPS_API_KEY)
results = gmaps.places(
    query="BPO company Delhi",
    type="establishment"
)

for place in results:
    details = gmaps.place(place["place_id"])
    leads.append({
        "company": details["name"],
        "phone": details.get("formatted_phone_number"),
        "website": details.get("website"),
        "category": details["types"][0]
    })

Combined master list:

Total: 615 companies
Segment breakdown:
├── AI/Tech/SaaS: 478 (78%)
├── VC/Investors: 20 (3%)
├── BPO/Enterprise: 48 (8%)
└── Other: 69 (11%)

Phase 2: LLM Enrichment (The Intelligence Layer)

n8n HTTP Request node → OpenAI API:

{
  "model": "gpt-4o-mini",
  "messages": [{
    "role": "system",
    "content": "Research {{company}} and return JSON: 
    {industry, pain_points, recent_news, 
    tech_stack, decision_maker_title}"
  }]
}

Output per lead:

{
  "company": "TechCorp India",
  "industry": "SaaS",
  "pain_points": ["manual lead qualification", "scaling support"],
  "recent_news": "Series A raised ₹20 Cr, Feb 2026",
  "tech_stack": ["React", "Node.js", "AWS"],
  "decision_maker": "CTO/VP Engineering",
  "score": 8.2
}

Cost: ₹0.03/lead × 615 = ₹18.45 total enrichment

Phase 3: Segmentation + Scoring

n8n IF nodes (scoring logic):

Score 9–10 (Priority):
├── Series A/B funded → +3 points
├── Tech stack matches → +2 points  
├── Pain point match → +3 points
└── Decision maker identified → +2 points

Score 6–8 (High):
├── Matches ICP → +3 points
├── Website has contact form → +2 points
└── Active LinkedIn → +3 points

Score 1–5 (Nurture):
└── Keep in pipeline → Monthly newsletter

Template selection (4 variants):

Score 9–10 → Template A (technical, direct CTA)
Score 7–8 → Template B (problem-focused)
Score 5–6 → Template C (educational)
BPO segment → Template D (voice AI focus)

Phase 4: Email Personalization + Send

n8n Code node (personalization engine):

const lead = $input.first().json;
const subject = `AI Automation for ${lead.company} — Quick Intro`;

const body = `Hi ${lead.firstName},
Saw ${lead.company} ${lead.recent_news || "is scaling"} — 
congratulations on the momentum.

Most teams at your stage struggle with ${lead.pain_points[0]}.

We built an autonomous system at SingularRarity Labs that:
• ${lead.pain_points[0]} → 85% automated  
• Deployed in 4 weeks
• ₹0 recurring cost on your existing stack

Worth a 15-minute call this week?

Best,
[Name]
SingularRarity Labs | info@singularraritylabs.com`;

return { subject, body };

Zoho SMTP config (n8n Send Email node):

Host: smtp.zoho.in
Port: 465 (SSL)
From: info@singularraritylabs.com
Throttle: 90 seconds between sends
Batch: 30/day (warm-up phase)

Phase 5: CRM Sync + Lead Tracking

Zoho CRM node (auto-creates lead):

Fields mapped:
├── First Name / Last Name
├── Company
├── Email  
├── Lead Source: "n8n Outreach"
├── Lead Score: {{score}}
├── Segment: {{industry}}
├── Last Contacted: {{timestamp}}
└── Email Status: "Sent"

Follow-up automation:

Day 0: Initial email sent
Day 3: No reply? → Send follow-up #1 (value add)
Day 7: No reply? → Send follow-up #2 (case study)
Day 14: No reply? → Move to "Newsletter" segment
Reply received? → Immediately notify via Zoho + Slack

Full n8n Workflow (12 Nodes)

[⏰ Cron: Mon-Fri 9AM IST]
      ↓
[📂 Read CSV: leads.csv]
      ↓
[🔀 Split In Batches: 1]
      ↓
[✅ IF: Has Email?]
      ↓
[🤖 LLM: Enrich + Score]
      ↓
[🔀 Switch: Template Select]
      ↓
[✍️ Code: Personalize Body]
      ↓
[📧 Zoho SMTP: Send]
      ↓
[⏳ Wait: 90s]
      ↓
[📊 Zoho CRM: Create Lead]
      ↓
[📝 Google Sheets: Log]
      ↓
[🔔 Slack: Notify on Reply]

Real Results (Week 1 Live)

Leads processed: 130 (LinkedIn segment first)
Emails sent: 130
Open rate: 42% (industry avg: 21%)
Reply rate: 8.2% (industry avg: 2.4%)
Discovery calls booked: 3
Proposals sent: 2
Cost: ₹485 total (₹18 enrichment + ₹467 Zoho)

vs manual alternative:
Time: 83 hours → 0 hours
Cost: ₹25,000 SDR → ₹485
ROI: 51x on cost alone

Scaling to 485 Phone Leads

WhatsApp/call workflow (separate n8n pipeline):

[CSV: 485 companies] → [Segment by score]
      ↓
[Template: WhatsApp message]  
      ↓
[Manual send trigger] → [CRM log]

Phone lead script (LLM-generated per company):

"Hi, is this [Company]? I'm calling from SingularRarity Labs.
We build AI automation for [industry] companies.
I noticed [specific observation about their business].
Are you the right person to speak about tech partnerships?"

The Complete Cost Breakdown

Infrastructure:
├── n8n: ₹0 (self-hosted Docker)
├── Postgres: ₹0 (existing)
├── Ollama/OpenAI: ₹18.45 (enrichment)
└── Zoho Mail: ₹1,200/month (existing)

Development:
├── Pipeline build: ₹0 (in-house)
├── Scraper build: ₹0 (in-house)
└── CRM setup: ₹0 (existing Zoho)

Total monthly: ₹1,200
vs SDR hire: ₹40,000/month
Savings: ₹38,800/month (97%)

What This Means for Your Business

If you have:

✅ A lead list (or we'll build one)
✅ Zoho/Gmail/Outlook email
✅ n8n (self-hosted or cloud ₹1,500/month)
✅ ₹5 Lakh budget (we build + hand over)

You get:

✅ 500+ personalized emails/week (automated)
✅ Live CRM with scored leads
✅ 3x open rates vs generic blasts
✅ Auto follow-up sequences
✅ Full lead intelligence dossier per company
✅ ₹0 recurring cost on self-hosted stack

At SingularRarity Labs, we deliver this exact pipeline for ₹5 Lakh, in 2 weeks, fully documented. You own the workflow, the data, and the results.

Ready to build your autonomous lead pipeline? We'll hand you the n8n JSON, the scraper, and the CRM integration — production ready.

SingularRarity Labs builds what others can't imagine — where singular ideas become rare realities.