utm-referrer-attribution-parser 0.1.4

Modern Python library combining referrer parsing with tracking parameter extraction for web analytics

utm-referrer-attribution-parser

A modern Python library that combines referrer parsing with tracking parameter extraction for comprehensive web analytics attribution.

✨ Super Simple API

from utm_referrer_parser import webmetic_referrer

# Just pass the URL and optional referrer - that's it!
result = webmetic_referrer(
    url="https://example.com/page?utm_source=google&utm_medium=cpc&gclid=abc123",
    referrer="https://www.google.com/search?q=analytics"
)

print(result)
# {
#     'source': 'google',
#     'medium': 'cpc',
#     'click_id': 'abc123',
#     'click_id_type': 'gclid',
#     'term': 'analytics'
# }

🚀 Features

• Ultra-Simple API: Just webmetic_referrer(url, referrer) - that's it!
• Unified Click Tracking: Clean click_id and click_id_type fields instead of 15+ individual parameters
• 25+ Tracking Parameters: UTM, Google Ads, Facebook, TikTok, LinkedIn, email platforms, and more
• Smart Referrer Analysis: Uses Snowplow's referrer database for accurate source/medium classification
• Advanced Domain Parsing: Uses tldextract for robust international domain handling (.co.uk, .com.au, etc.)
• Auto-updating Database: Weekly updates of referrer database with local fallback
• High Performance: In-memory caching and optimized parsing
• Framework Agnostic: Works with any Python web framework
• Production Ready: 99%+ accuracy validated with 150+ real-world test cases
• International Support: Handles global search engines (Google, Bing, Baidu, Yandex, Naver, etc.)

📦 Installation

pip install utm-referrer-attribution-parser

🎯 Quick Examples

Google Ads Click

result = webmetic_referrer(
    url="https://site.com/landing?utm_source=google&utm_medium=cpc&gclid=abc123"
)
# Returns: {'source': 'google', 'medium': 'cpc', 'click_id': 'abc123', 'click_id_type': 'gclid'}

Facebook Ad

result = webmetic_referrer(
    url="https://site.com/product?fbclid=fb123",
    referrer="https://www.facebook.com/"
)
# Returns: {'source': 'facebook', 'medium': 'cpc', 'click_id': 'fb123', 'click_id_type': 'fbclid'}

Organic Search

result = webmetic_referrer(
    url="https://site.com/blog",
    referrer="https://www.google.com/search?q=analytics+guide"
)
# Returns: {'source': 'Google', 'medium': 'search', 'term': 'analytics guide'}

Direct Traffic

result = webmetic_referrer("https://site.com/")
# Returns: {'source': '(direct)', 'medium': '(none)'}

Internal Navigation

result = webmetic_referrer(
    url="https://shop.example.com/products",
    referrer="https://example.com/"
)
# Returns: {'source': '(internal)', 'medium': 'internal'}

The library automatically detects internal navigation between subdomains using advanced TLD parsing, correctly handling complex domains like .co.uk, .com.au, .org.br, etc.

🎯 Unified Click Tracking

Instead of tracking 15+ individual click ID fields, we provide a clean unified structure:

Old Approach (Complex)

# Multiple individual fields to check
result = {
    'gclid': 'abc123',
    'fbclid': None,
    'ttclid': None,
    'msclkid': None,
    # ... 15+ more fields
}

New Approach (Clean)

# Just 2 unified fields
result = {
    'click_id': 'abc123',        # The actual tracking value
    'click_id_type': 'gclid'     # Which parameter it came from
}

Benefits

• Cleaner API: 2 fields instead of 15+
• Easier Logic: Simple if result['click_id'] checks
• Platform Detection: Still get source/medium attribution automatically
• Priority Handling: Google Ads → Facebook → Microsoft → Other platforms

Supported Parameters

Standard UTM

• utm_source, utm_medium, utm_campaign, utm_term, utm_content, utm_id

Click Tracking (Unified)

• click_id - The actual click tracking value
• click_id_type - Which parameter provided it (gclid, fbclid, ttclid, etc.)

Google Ads Metadata

• gclsrc, gad_source, srsltid

Social Media

• igshid (Instagram), sccid (Snapchat)

Email Marketing

• mc_cid, mc_eid (Mailchimp)
• ml_subscriber_hash (MailerLite)

Other Platform Parameters

• epik (Pinterest), ttd_uuid (Trade Desk), obOrigUrl (Outbrain), and more

🧪 Validation & Testing

This library has been extensively tested with:

• 150+ real database cases from production environments
• 50+ diverse internet scenarios covering global platforms
• 99%+ accuracy rate in attribution detection
• 100% error handling - no crashes on malformed inputs

Supported Platforms

• Search Engines: Google, Bing, Baidu, Yandex, DuckDuckGo, Naver, Yahoo, Ecosia
• Social Media: Facebook, Instagram, TikTok, Twitter, LinkedIn, Pinterest, Reddit, Snapchat
• Email Marketing: Mailchimp, MailerLite, Constant Contact, SendGrid, ConvertKit
• Business Tools: Slack, Microsoft Teams, Calendly, Notion, Zoom
• E-commerce: Amazon, eBay, Shopify, Etsy, AliExpress

🔄 Migration from Complex Systems

Replace complex tracking data dictionaries with simple function calls:

# OLD: Complex dictionary approach
tracking_data = {
    "dl": "https://site.com/?utm_source=google&gclid=abc123",
    "dr": "https://www.google.com/search?q=analytics", 
    "bu": "https://site.com"
}
result = parse_attribution(tracking_data)

# NEW: Ultra-simple API
result = webmetic_referrer(
    url="https://site.com/?utm_source=google&gclid=abc123",
    referrer="https://www.google.com/search?q=analytics"
)

📊 What Makes This Different

• Intelligent Priority: UTM parameters → Click IDs → Referrer analysis → Direct traffic
• Unified Click Tracking: Clean click_id/click_id_type structure instead of 15+ individual fields
• Click ID Detection: Automatically identifies 25+ types of advertising click IDs
• International Ready: Built-in support for global search engines and platforms
• Real-world Tested: Validated against actual production analytics data
• Future Proof: Auto-updating referrer database keeps up with new platforms

License

MIT License - see LICENSE file for details.