Airtable Lead Enricher

Fill missing lead data automatically. Email, phone, AI scoring. Works with Airtable or standalone or in any workflow!

Documentation • Try on Apify • Support

---

Demo

---

What It Does

Reads leads from Airtable → Finds missing data → Writes back.

Before	After
Acme Corp	Acme Corp contact@acme.com +1 555 0100 Lead Score: Good ICP Score: Excellent

Key Stats:

• ⚡ 3-5 seconds per lead enrichment
• 💰 $0.03 per lead (vs competitors: $0.15-$0.36)
• 📊 85%+ success rate for contact info
• 🎯 AI-powered scoring with 7 quality labels
• 🔄 3 processing modes: Airtable batch, single record, or standalone API
• 📈 Up to 1000 companies per run in API mode

---

Features

• Contact Info - Email, phone
• Verified Emails - Hunter.io
• Social Links - LinkedIn, Facebook, Twitter
• AI Scoring - 0-100 score with quality labels (Excellent, Good, Fair, Poor, Bad)
• Batch Mode - 1000 companies per run
• API Mode - No Airtable needed

---

Quick Start

1. Get Airtable Token

airtable.com/create/tokens → Create token → Copy

2. Get IDs

• Base ID: airtable.com/appXXXX → appXXXX
• Table ID: Table menu → Copy table ID → tblXXXX

3. Run

{
  "airtable": {
    "apiKey": "patXXXXXXXX",
    "baseId": "appXXXXXX",
    "tableId": "tblXXXXXX",
    "inputFields": {
      "companyName": "Company Name"
    }
  }
}

---

Pricing

$0.03 per lead (vs Clearbit $0.36, Apollo $0.15)

Leads	Cost
100	$3
1,000	$30

---

Data Sources

Source	Data	API Key?
Website	Email, phone, tech	No
Google Maps	Phone, address, rating	No
Hunter.io	Verified email, socials	Yes
AI	Score, ICP match	Yes

---

Configuration

Basic

{
  "airtable": {
    "apiKey": "patXXX",
    "baseId": "appXXX",
    "tableId": "tblXXX",
    "inputFields": {"companyName": "Company Name"}
  }
}

With AI

{
  "airtable": {...},
  "llm": {
    "enabled": true,
    "provider": "openai",
    "apiKey": "sk-XXX"
  },
  "scoring": {
    "icpCriteria": "B2B SaaS, 50-500 employees, US"
  }
}

With Hunter.io

{
  "enrichment": {
    "sources": ["google_maps", "website", "hunter"],
    "hunter": {
      "enabled": true,
      "apiKey": "YOUR_KEY"
    }
  }
}

---

Field Mapping

Map your Airtable columns to internal fields.

Input Fields (What to Read)

{
  "inputFields": {
    "companyName": "Company Name",  // REQUIRED
    "website": "Website",           // REQUIRED
    "location": "City"              // Optional
  }
}

Internal → Your Column Name (case-sensitive)

Output Fields (What to Write)

{
  "outputFields": {
    "email": "Contact Email",
    "phone": "Phone Number",
    "leadScore": "Lead Score",
    "icpScore": "ICP Match",
    "summary": "AI Summary",
    "techStack": "Tech Stack",
    "linkedinUrl": "LinkedIn"
  }
}

If not specified, uses default names: email, phone, leadScore, etc.

Available Output Fields:

• Contact: email, phone, address
• Google Maps: rating, reviewCount, category
• Company: description, techStack, industry, employeeCount, foundingYear, companyStage
• Social: linkedinUrl, facebookUrl, twitterHandle
• AI: leadScore, icpScore, summary, targetCustomers, valueProposition, keyProducts, outreachAngles

---

Modes

Mode	Use Case	Needs Airtable?	Max Records
batch	Scheduled Airtable sync	✅ Yes	100
single	Specific record IDs	✅ Yes	100
api	Standalone (no Airtable)	❌ No	1000

API Mode (No Airtable Required)

Perfect for integrations, workflows, and pipelines.

{
  "mode": "api",
  "companies": [
    {"companyName": "Acme", "website": "https://acme.com"}
  ]
}

No airtable.apiKey, baseId, or tableId needed. Just provide companies.

---

LLM Providers

Supports OpenAI, Anthropic, and AWS Bedrock for AI-powered lead scoring.

Tested models: Claude 3 Haiku, GPT-4o, GPT-3.5 Turbo work well.

Get API keys:

• OpenAI
• Anthropic
• AWS Bedrock

---

Examples & Use Cases

Example 1: Enrich 100 Leads in Airtable

Scenario: You have a table with company names. You want to fill in missing emails, phones, and AI scores.

Input:

{
  "airtable": {
    "apiKey": "patXXXXXXXXXXXXXXXX",
    "baseId": "appABC123DEF456",
    "tableId": "tblXYZ789GHI012",
    "maxRecords": 100,
    "inputFields": {
      "companyName": "Company Name",
      "website": "Website"
    },
    "outputFields": {
      "email": "Email",
      "phone": "Phone",
      "leadScore": "Lead Score",
      "summary": "Company Summary"
    }
  },
  "llm": {
    "enabled": true,
    "provider": "anthropic",
    "apiKey": "sk-ant-XXXXXXXXXXXXX"
  }
}

Result: All 100 records get enriched with contact info and AI-generated scores. Takes ~5-8 minutes.

---

Example 2: Qualify Leads with ICP Scoring

Scenario: You want to score leads against your ideal customer profile to prioritize outreach.

Input:

{
  "mode": "api",
  "companies": [
    {"companyName": "Stripe", "website": "https://stripe.com"},
    {"companyName": "Shopify", "website": "https://shopify.com"},
    {"companyName": "Twilio", "website": "https://twilio.com"}
  ],
  "llm": {
    "enabled": true,
    "provider": "openai",
    "apiKey": "sk-proj-XXXXXXXXXXXXX"
  },
  "scoring": {
    "enabled": true,
    "icpCriteria": "B2B SaaS companies, 100-1000 employees, API-first products, US market"
  }
}

Output:

{
  "companyName": "Stripe",
  "icpScore": "Excellent",
  "leadScore": "Excellent",
  "summary": "Stripe is a technology company providing payment processing software and APIs...",
  "outreachAngles": [
    "Their API-first approach aligns perfectly with your developer-focused solution",
    "Target their payment engineering team for infrastructure optimization",
    "Reference their recent Series H funding for budget availability"
  ]
}

---

Example 3: Daily Scheduled Enrichment

Scenario: Automatically enrich new leads every day at 2 AM.

Setup:

1. In Apify Console → Schedules → Create new schedule
2. Cron: 0 2 * * *
3. Input:

{
  "airtable": {
    "apiKey": "patXXXXXXXXXXXXXXXX",
    "baseId": "appABC123DEF456",
    "tableId": "tblXYZ789GHI012",
    "filterFormula": "AND({Status} = 'New', NOT({enrichedAt}))",
    "maxRecords": 100,
    "inputFields": {
      "companyName": "Company Name",
      "website": "Website"
    }
  }
}

Result: Only processes records with Status="New" and no enrichedAt timestamp. Runs automatically daily.

---

Example 4: Lambda Function Integration (Python)

Scenario: Enrich leads from your application backend via AWS Lambda.

import json
import requests
import os

def lambda_handler(event, context):
    """
    Enrich a single company when a new lead is created
    """
    company_name = event.get('company_name')
    website = event.get('website')

    # Call Apify Actor in API mode
    response = requests.post(
        "https://api.apify.com/v2/acts/datahq~airtable-lead-enricher/run-sync-get-dataset-items",
        params={"token": os.environ['APIFY_TOKEN']},
        json={
            "mode": "api",
            "companies": [
                {
                    "companyName": company_name,
                    "website": website
                }
            ],
            "enrichment": {
                "sources": ["website", "google_maps"]
            },
            "llm": {"enabled": False}  # Fast mode, no AI
        },
        timeout=60
    )

    results = response.json()
    enriched_data = results[0] if results else {}

    return {
        'statusCode': 200,
        'body': json.dumps({
            'email': enriched_data.get('email'),
            'phone': enriched_data.get('phone'),
            'techStack': enriched_data.get('techStack')
        })
    }

Deployment:

# Add to requirements.txt
requests

# Deploy
zip function.zip lambda_function.py
aws lambda create-function \
  --function-name enrich-lead \
  --runtime python3.11 \
  --handler lambda_function.lambda_handler \
  --zip-file fileb://function.zip \
  --environment Variables={APIFY_TOKEN=apify_api_XXXXX}

---

Example 5: Webhook Integration with Make.com/Zapier

Scenario: Get enriched data pushed to your CRM automatically.

Setup:

1. Create webhook endpoint (Make.com/Zapier/n8n)
2. Configure Actor input:

{
  "mode": "api",
  "companies": [
    {"companyName": "Acme Corp", "website": "https://acme.com"}
  ],
  "webhookUrl": "https://hook.us1.make.com/xxxxxxxxxxxxx"
}

3. Webhook receives:

{
  "type": "RUN_STATS",
  "stats": {
    "companiesProcessed": 1,
    "enrichmentSuccessful": 1
  },
  "results": [
    {
      "companyName": "Acme Corp",
      "email": "contact@acme.com",
      "phone": "+1-555-0100",
      "leadScore": "Good"
    }
  ]
}

4. Make.com automation:
   • Trigger: Webhook received
   • Action 1: Parse JSON
   • Action 2: Update Salesforce/HubSpot record
   • Action 3: Send Slack notification for "Excellent" leads

---

Example 6: Airflow DAG for ETL Pipeline

Scenario: Enrich leads as part of your nightly data pipeline.

from airflow import DAG
from airflow.operators.http_operator import SimpleHttpOperator
from airflow.operators.python_operator import PythonOperator
from datetime import datetime, timedelta
import json

default_args = {
    'owner': 'data-team',
    'depends_on_past': False,
    'start_date': datetime(2025, 1, 1),
    'retries': 2,
    'retry_delay': timedelta(minutes=5),
}

dag = DAG(
    'lead_enrichment_pipeline',
    default_args=default_args,
    schedule_interval='0 2 * * *',  # 2 AM daily
    catchup=False
)

# Task 1: Extract new leads from warehouse
extract_leads = PythonOperator(
    task_id='extract_new_leads',
    python_callable=lambda: [...],  # Your extraction logic
    dag=dag
)

# Task 2: Enrich via Apify
enrich_leads = SimpleHttpOperator(
    task_id='enrich_leads',
    http_conn_id='apify_api',
    endpoint='/v2/acts/datahq~airtable-lead-enricher/run-sync-get-dataset-items',
    method='POST',
    data=json.dumps({
        "mode": "api",
        "companies": "{{ task_instance.xcom_pull(task_ids='extract_new_leads') }}",
        "enrichment": {"sources": ["website", "google_maps"]}
    }),
    headers={'Content-Type': 'application/json'},
    dag=dag
)

# Task 3: Load enriched data to warehouse
load_enriched_data = PythonOperator(
    task_id='load_to_warehouse',
    python_callable=lambda: [...],  # Your load logic
    dag=dag
)

extract_leads >> enrich_leads >> load_enriched_data

---

Example 7: Batch Process 1000 Companies (API Mode)

Scenario: You have a CSV with 1000 companies to enrich. No Airtable needed.

Input:

{
  "mode": "api",
  "companies": [
    {"companyName": "Company 1", "website": "https://example1.com"},
    {"companyName": "Company 2", "website": "https://example2.com"},
    ...
    {"companyName": "Company 1000", "website": "https://example1000.com"}
  ],
  "enrichment": {
    "sources": ["website"],
    "concurrency": 20
  }
}

Processing time: ~25-40 minutes for 1000 companies (parallel processing)

Output: Dataset with 1000 enriched records. Download as JSON/CSV from Apify Console.

---

Example 8: Hunter.io Integration for Verified Emails

Scenario: Get verified professional emails and social profiles.

Input:

{
  "mode": "api",
  "companies": [
    {"companyName": "Anthropic", "website": "https://anthropic.com"}
  ],
  "enrichment": {
    "sources": ["website", "hunter"],
    "hunter": {
      "enabled": true,
      "apiKey": "YOUR_HUNTER_API_KEY"
    }
  }
}

Output:

{
  "companyName": "Anthropic",
  "email": "contact@anthropic.com",
  "linkedinUrl": "https://linkedin.com/company/anthropic-ai",
  "twitterHandle": "@AnthropicAI",
  "emailVerified": true,
  "emailScore": 92
}

---

Integration Examples

More examples in examples/ directory:

• lambda_python.py - AWS Lambda (Python)
• lambda_node.js - AWS Lambda (Node.js)
• gcp_function_python.py - GCP Cloud Functions
• glue_job.py - AWS Glue ETL
• ecs_task_definition.json - AWS ECS/Fargate
• docker_run.sh - Docker (EC2, anywhere)
• github_actions.yml - GitHub Actions
• airflow_dag.py - Apache Airflow

CRM Integrations

Salesforce, HubSpot, Pipedrive, Zoho

Use webhooks or automation platforms:

1. Webhook: Set webhookUrl to receive enriched data
2. Zapier/Make: Trigger on Airtable update or schedule
3. Direct API: Call from your CRM's automation (see examples above)

---

Output Fields

Field	Source
email	Website/Hunter
phone	Maps/Website
linkedinUrl	Hunter
leadScore	AI (Excellent/Good/Fair/Poor/Bad)
icpScore	AI (Excellent/Good/Fair/Poor/Bad)
summary	AI
techStack	Website

Full list

---

Scheduling

Run daily at 2 AM:

{
  "airtable": {
    "filterFormula": "NOT({enrichedAt})",
    "maxRecords": 100
  }
}

Cron: 0 2 * * *

---

Best Practices & Tips

🎯 Optimize for Speed

Fastest configuration (website only, no AI):

{
  "enrichment": {
    "sources": ["website"],
    "concurrency": 20
  },
  "llm": {"enabled": false},
  "scoring": {"enabled": false}
}

Speed: ~2 seconds per lead

---

Balanced configuration (website + Google Maps):

{
  "enrichment": {
    "sources": ["website", "google_maps"],
    "concurrency": 10
  },
  "llm": {"enabled": false}
}

Speed: ~3-4 seconds per lead

---

Full enrichment (all sources + AI):

{
  "enrichment": {
    "sources": ["website", "google_maps", "hunter"],
    "concurrency": 5
  },
  "llm": {"enabled": true}
}

Speed: ~8-12 seconds per lead (includes AI analysis)

---

💰 Optimize for Cost

Tips to reduce costs:

1. Use API mode instead of Airtable mode when you don't need live sync
2. Disable AI scoring when you only need contact info (saves LLM API costs)
3. Start with website-only enrichment, then add Hunter.io/AI if needed
4. Use filterFormula to avoid re-processing enriched records:

   "filterFormula": "NOT({enrichedAt})"
   

5. Batch strategically - Process 100-200 records per run instead of one-by-one

Cost breakdown:

• Website scraping: Free
• Google Maps: Free
• Hunter.io: $0.005 per request (their pricing)
• OpenAI GPT-4o: ~$0.01 per lead
• Anthropic Claude 3 Haiku: ~$0.001 per lead (recommended)

---

🔍 Improve Data Quality

1. Always provide website URLs

{
  "inputFields": {
    "companyName": "Company Name",
    "website": "Website"  // Critical for best results
  }
}

2. Use ICP criteria for better scoring

{
  "scoring": {
    "icpCriteria": "Specific criteria here - industry, size, location, tech stack"
  }
}

More specific = more accurate scores.

3. Enable multiple sources

{
  "sources": ["website", "google_maps", "hunter"]
}

More sources = more complete data.

4. Handle updates correctly

{
  "updateMode": "append"  // Recommended: only fill empty fields
}

Options:

• append (default) - Only fill empty fields, preserve existing data
• overwrite - Replace all fields with new data
• skip - Skip record entirely if any field has data

---

⚡ Common Patterns

Pattern 1: New Lead Qualification

{
  "filterFormula": "AND({Status} = 'New', NOT({enrichedAt}))",
  "llm": {"enabled": true},
  "scoring": {
    "icpCriteria": "Your ICP here",
    "minScore": 60
  }
}

Pattern 2: Re-enrich Old Leads

{
  "filterFormula": "AND({enrichedAt} < DATEADD(TODAY(), -90, 'days'))",
  "updateMode": "overwrite"
}

Pattern 3: Enrich Only High-Value Segments

{
  "filterFormula": "AND({Deal Size} > 10000, NOT({Email}))",
  "enrichment": {
    "sources": ["website", "hunter"]
  }
}

Pattern 4: Webhook-Driven Real-Time Enrichment

{
  "mode": "api",
  "companies": [{"companyName": "...", "website": "..."}],
  "webhookUrl": "https://your-webhook.com/endpoint",
  "enrichment": {
    "sources": ["website"]
  }
}

---

Performance Guide

Expected Processing Times

Records	Mode	Config	Time
1	API	Website only	~2-3 sec
1	API	Website + Maps	~4-5 sec
1	API	Full (Website + Maps + Hunter + AI)	~10-15 sec
10	Batch	Website only	~30-45 sec
100	Batch	Website + Maps	~5-8 min
100	Batch	Full enrichment	~15-25 min
1000	API	Website only	~25-40 min

Factors affecting speed:

• Number of sources enabled
• AI/LLM enabled (adds 3-5 sec per lead)
• Concurrency setting (1-20, default: 10)
• Website response time (some sites are slower)
• Network latency

Memory usage:

• Base: 512 MB
• Recommended: 2048 MB (default)
• For 1000+ records with AI: 4096 MB

---

Troubleshooting

Problem	Fix
Invalid API key	Use PAT (starts with pat)
Unknown field	Check case-sensitive names
No records	Check filter/view
No social links	Enable Hunter.io
Timeout errors	Reduce concurrency or increase timeout
Out of memory	Increase memory to 4096 MB in Actor settings
Low success rate	Ensure website URLs are provided
Empty AI scores	Check LLM API key is valid

Common Issues:

Q: Why are some fields empty? A: Not all data is available for every company. Results depend on:

• Website structure (some sites hide contact info)
• Google Maps presence
• Hunter.io database coverage

Q: How to handle rate limits? A: Reduce concurrency:

{
  "enrichment": {
    "concurrency": 5  // Lower = fewer simultaneous requests
  }
}

Q: Why is AI scoring not working? A: Check:

1. LLM provider API key is valid
2. llm.enabled: true and scoring.enabled: true
3. Provider supports your region (OpenAI, Anthropic widely available)

---

Links

• Documentation
• Apify Actor
• Report Issues

---

Lead enrichment made simple.