LinkForty Documentation

Deferred Deep Linking Explained

Deferred deep linking is a mobile app technique that routes users to specific content after they install your app, remembering the context of the original link they clicked before installing. It's the "magic" that makes mobile user acquisition seamless.

The Problem Without Deferred Deep Linking

Standard scenario (no deferred deep linking):

User sees Instagram ad for a specific product (winter jacket)
User clicks ad → Redirected to App Store (app not installed)
User downloads and installs app
User opens app → Lands on home screen
❌ User must search for winter jacket manually
❌ Context lost, poor experience
❌ User likely abandons (70% drop-off)

Result: You paid for the ad, but lost the sale.

With deferred deep linking:

User sees Instagram ad for winter jacket
User clicks ad → System "remembers" this was for winter jacket
User downloads and installs app
User opens app → Automatically routed to winter jacket product page
✅ Seamless experience
✅ User sees exactly what they clicked
✅ 3-5x higher conversion rate

Result: You paid for the ad, and made the sale.

How Deferred Deep Linking Works

The Complete Flow

1. User clicks link
   → Link: https://go.yourapp.com/products/winter-jacket

2. Platform detects: App not installed
   → Creates "fingerprint" of user's device
   → Stores: {link: "products/winter-jacket", fingerprint: "abc123"}
   → Redirects to App Store

3. User installs app from App Store

4. User opens app for first time
   → App SDK contacts attribution platform
   → Sends device fingerprint: "abc123"

5. Platform matches fingerprint
   → Match found! Original link was "products/winter-jacket"
   → Returns link data to app

6. App receives link data
   → Routes user to winter jacket product page
   → User sees the product they originally clicked

Time between click and install: Can be seconds to days (within attribution window)

Device Fingerprinting: The Technology Behind It

What is a Device Fingerprint?

A device fingerprint is a combination of device characteristics that, together, create a unique identifier without using persistent device IDs (like IDFA or GAID).

Fingerprint Components

Data collected at click time:

{
  "ip": "192.168.1.100",
  "userAgent": "Mozilla/5.0 (iPhone; CPU iPhone OS 17_2 like Mac OS X)",
  "screen": {
    "width": 1179,
    "height": 2556,
    "density": 3
  },
  "language": "en-US",
  "timezone": "America/Los_Angeles",
  "platform": "iOS",
  "os_version": "17.2",
  "device_model": "iPhone 15 Pro",
  "carrier": "Verizon",
  "timestamp": "2024-01-15T10:30:00Z"
}

Data collected at app open (install time):

{
  // Same data points collected when app opens
  "ip": "192.168.1.100",  // Same WiFi/cellular network
  "userAgent": "...",     // Same device
  "screen": {...},        // Same screen size
  "language": "en-US",    // Same settings
  "timezone": "America/Los_Angeles",  // Same timezone
  // etc.
}

Matching Algorithm

Step 1: Compare fingerprints

Click fingerprint vs Install fingerprint

Matches:
✅ IP address: 192.168.1.100 (same)
✅ Screen: 1179x2556 (same)
✅ Language: en-US (same)
✅ Timezone: America/Los_Angeles (same)
✅ OS version: 17.2 (same)
✅ Carrier: Verizon (same)

Match confidence: 95%

Step 2: Check attribution window

Click time: 2024-01-15 10:30:00
Install time: 2024-01-15 11:45:00
Time diff: 1 hour 15 minutes

Attribution window: 7 days (168 hours)

✅ Within window → Match valid

Step 3: Return result

Match found!
Original link: https://go.yourapp.com/products/winter-jacket
Link data: {
  "product_id": "winter-jacket",
  "utm_source": "instagram",
  "utm_campaign": "winter-sale",
  "discount_code": "WINTER20"
}

App receives this data → Routes to product + applies discount

Accuracy Expectations

Industry Standard Accuracy

Fingerprinting accuracy: 70-80%

Why not 100%?

Reasons for false negatives (missed matches):

IP address changed between click and install
- User switched from WiFi to cellular
- User left home/office
- Dynamic IP reassigned
Shared IP addresses
- Office WiFi (100+ devices on same IP)
- Public WiFi (coffee shops, airports)
- Mobile carrier NAT (multiple users share IP)
Attribution window expired
- User clicked 8 days ago (7-day window)
Device settings changed
- User changed language/timezone
- User updated OS between click and install

Reasons for false positives (wrong matches - rare):

Two users with identical device characteristics on same IP
- Extremely rare (1-2% of matches)
- More common on shared WiFi networks

Accuracy by Scenario

Scenario	Accuracy	Explanation
Same WiFi, short time (<1 hour)	90-95%	Ideal conditions
Same cellular network, <24 hours	80-85%	Cellular IPs more unique
Different network (WiFi→Cellular)	60-70%	IP changed, other factors match
Shared WiFi (office, cafe)	50-60%	Multiple devices, same IP
Long time (3-7 days)	65-75%	IP may have changed
Beyond attribution window	0%	Expired, labeled "organic"

LinkForty achieves: 70-80% average accuracy (industry standard)

Privacy & Compliance

Why Fingerprinting is Privacy-Friendly

No persistent device IDs:

✅ No IDFA (iOS Identifier for Advertisers)
✅ No GAID (Google Advertising ID)
✅ No permanent tracking identifiers

No personal data:

✅ No names, emails, phone numbers
✅ No cross-app tracking
✅ No long-term user profiling

Temporary matching:

✅ Fingerprints used only for attribution window (7 days)
✅ After window expires, data can be purged
✅ No persistent user identity

Why fingerprinting is GDPR-compliant:

Not personal data under GDPR definition
- Device characteristics ≠ personally identifiable information
- Cannot identify a specific individual
- Temporary, used only for attribution
Legitimate interest legal basis
- Attribution is necessary business function
- Users benefit from better experience
- Privacy impact is minimal
User rights supported
- Right to deletion (delete click/install records)
- Right to data export (export attribution data)
- Right to object (can opt-out)

LinkForty's approach:

GDPR-compliant out of the box
Supports data deletion via API
Clear privacy policy
Self-hosted option (full data ownership)

Deferred Deep Linking Use Cases

1. Product-Specific Advertising

Scenario: E-commerce app running ads for specific products

Setup:

Instagram ad for "Winter Jacket"
→ Link: https://go.store.com/products/winter-jacket-2024
→ User without app clicks
→ Redirected to App Store
→ User installs, opens app
→ Deferred deep link routes to winter jacket page

Benefits:

✅ User sees exact product they clicked
✅ 3-5x higher purchase rate
✅ Better ad ROI

Without deferred deep linking:

User lands on home screen
Must search for winter jacket
70% abandon before finding it

2. Referral Programs

Scenario: User invites friend to app

Setup:

User A shares referral link
→ Link: https://go.app.com/invite/userA?code=FRIEND20
→ User B (friend) clicks, installs app
→ Deferred deep link knows:
  - Referred by userA
  - Discount code: FRIEND20
→ App automatically:
  - Credits userA with referral bonus
  - Applies FRIEND20 discount to User B

Benefits:

✅ Automatic referral tracking
✅ Rewards applied instantly
✅ No manual code entry needed

3. Email Campaigns (New User Acquisition)

Scenario: Email campaign targeting non-users

Setup:

Email: "Get 20% off your first order"
→ Link: https://go.app.com/signup?promo=EMAIL20&source=newsletter
→ Recipient clicks (doesn't have app)
→ Redirected to App Store
→ Installs app
→ Deferred deep link:
  - Applies EMAIL20 discount automatically
  - Tracks install came from email newsletter

Benefits:

✅ Seamless onboarding
✅ Discount applied automatically
✅ Track email campaign ROI

4. Influencer Marketing

Scenario: Each influencer gets unique link

Setup:

Influencer X posts: https://go.app.com/influencer-x
Influencer Y posts: https://go.app.com/influencer-y

User clicks Influencer X link → Installs
Deferred deep link attributes install to Influencer X

Analytics show:
- Influencer X: 1,000 installs
- Influencer Y: 400 installs
→ Know which influencer drives more installs

Benefits:

✅ Precise attribution per influencer
✅ Calculate ROI per partnership
✅ Optimize influencer spend

5. QR Code Campaigns

Scenario: QR codes on product packaging, posters, business cards

Setup:

QR code on product packaging
→ Encodes: https://go.app.com/product/xyz?source=package
→ User scans QR code
→ Doesn't have app → Sent to App Store
→ Installs app
→ Deferred deep link routes to product xyz
→ Tracks install came from product packaging

Benefits:

✅ Offline-to-online attribution
✅ Measure packaging campaign effectiveness
✅ Seamless product discovery

Implementation Approaches

1. DIY (Build Your Own)

What you need to build:

Link generation system
- Create short links
- Store destination URLs
- Track UTM parameters
Fingerprinting logic
- Collect device characteristics at click time
- Store fingerprints with link data
- Set TTL (e.g., 7 days)
SDK for mobile apps
- Collect device characteristics at app open
- Send to server for matching
- Receive matched link data
- Route user to destination
Matching algorithm
- Compare click fingerprint vs install fingerprint
- Calculate confidence score
- Return best match
Analytics system
- Track clicks, installs, conversions
- Attribution reporting
- Campaign performance

Time investment: 3-6 weeks of development

Challenges:

Fingerprinting accuracy optimization
Handling edge cases (shared IPs, network changes)
Cross-platform SDK (iOS, Android, React Native)
Scale (millions of clicks/installs)

2. Use an Attribution Platform

What platforms provide:

✅ Pre-built fingerprinting (optimized for accuracy)
✅ Mobile SDKs (iOS, Android, React Native)
✅ Link management UI
✅ Analytics dashboard
✅ API access

Platform options:

Enterprise Platforms

AppsFlyer, Branch, Adjust
Pricing: $500-$5,000+/month
Best for: Large enterprises, complex integrations

Open Source & Privacy-First

LinkForty
Pricing: Free (self-hosted) or $29/month (cloud)
Best for: Privacy-conscious teams, cost-effective scaling

Time investment: 1-2 days of SDK integration

LinkForty's Deferred Deep Linking

How LinkForty Implements It

1. Link Creation

Create link in LinkForty dashboard:
- Destination: /products/winter-jacket
- iOS URL: App Store link
- Android URL: Google Play link
- UTM params: source=instagram, campaign=winter-sale

Short link generated: https://go.yourapp.com/abc123

2. User Clicks (App Not Installed)

User clicks https://go.yourapp.com/abc123

LinkForty:
1. Creates device fingerprint
2. Stores: {
     link: "/products/winter-jacket",
     fingerprint: {...},
     utm_params: {...},
     timestamp: "2024-01-15T10:30:00Z"
   }
3. Redirects to App Store

3. User Installs and Opens App

App opens, LinkForty SDK initializes:

LinkForty.init({
  baseUrl: 'https://api.linkforty.com',
  apiKey: 'your-api-key'
});

const attribution = await LinkForty.getAttributionData();

4. Matching Happens

SDK sends device fingerprint to LinkForty
LinkForty matches: Click fingerprint → Install fingerprint
Confidence: 85% (within 7-day window)

Returns:
{
  "matched": true,
  "destination": "/products/winter-jacket",
  "utm_source": "instagram",
  "utm_campaign": "winter-sale",
  "link_id": "abc123"
}

5. App Routes User

if (attribution.matched) {
  // User came from a link, route to destination
  navigation.navigate(attribution.destination);

  // Track the conversion
  analytics.track('deferred_deep_link_matched', {
    campaign: attribution.utm_campaign,
    source: attribution.utm_source
  });
}

LinkForty Accuracy

LinkForty achieves 70-80% attribution accuracy:

Fingerprint data points used:

IP address
User agent (device model, OS version)
Screen resolution
Language settings
Timezone
Carrier information (mobile)

Optimization techniques:

IP address proximity matching (if exact IP changed but in same subnet)
Confidence scoring (weight factors differently)
Time decay (clicks closer in time weighted higher)
Deduplication (prevent double-attribution)

Result: Industry-standard 70-80% accuracy, privacy-compliant

Getting Started with LinkForty Deferred Deep Linking

Cloud (5-Minute Setup)

1. Create account at linkforty.com

2. Create your first link

Destination: /products/winter-jacket
iOS URL: https://apps.apple.com/app/yourapp/id123456
Android URL: https://play.google.com/store/apps/details?id=com.yourapp
Attribution window: 7 days

3. Install SDK

React Native:

npm install @linkforty/react-native-sdk

import LinkForty from '@linkforty/react-native-sdk';

// Initialize in App.tsx
await LinkForty.init({
  baseUrl: 'https://api.linkforty.com',
  apiKey: 'your-api-key',
  debug: __DEV__,
});

// Check for deferred deep link on app launch
const attribution = await LinkForty.getAttributionData();
if (attribution && attribution.matched) {
  console.log('Install attributed to:', attribution.utm_campaign);

  // Route to destination
  navigation.navigate(attribution.destination);

  // Apply discount code if present
  if (attribution.discount_code) {
    applyDiscount(attribution.discount_code);
  }

  // Award referral credit if present
  if (attribution.referrer_id) {
    creditReferrer(attribution.referrer_id);
  }
}

4. Share your link

Use https://go.yourapp.com/abc123 in:
- Instagram ads
- Email campaigns
- Influencer posts
- QR codes

5. Track conversions

// Track purchase event
await LinkForty.trackEvent('purchase', {
  amount: 49.99,
  product_id: 'winter-jacket',
  currency: 'USD'
});

// View attribution in LinkForty dashboard:
// - Which campaigns drive installs
// - Which campaigns drive purchases
// - ROI per campaign

View Complete SDK Integration Guide →

Self-Hosted (Free)

1. Install LinkForty Core

npm install @linkforty/core

2. Set up infrastructure

# PostgreSQL (required)
# Redis (optional, improves performance)

3. Configure

DATABASE_URL=postgresql://...
REDIS_URL=redis://...
JWT_SECRET=your-secret-key
ATTRIBUTION_WINDOW=168  # hours (7 days)

4. Deploy

Docker / Kubernetes / any Node.js hosting

5. Integrate SDK (same as cloud)

View Self-Hosting Guide →

Best Practices for Deferred Deep Linking

1. Always Provide Fallback Content

Don't assume attribution always works (70-80% match rate):

const attribution = await LinkForty.getAttributionData();

if (attribution && attribution.matched) {
  // Matched! Route to specific content
  navigation.navigate(attribution.destination);
} else {
  // No match (organic install or match failed)
  // Show onboarding flow
  navigation.navigate('Onboarding');
}

Never: Crash or show error if attribution fails Always: Handle both matched and unmatched scenarios gracefully

2. Set Appropriate Attribution Windows

Choose based on your product:

✅ 24 hours: Flash sales, time-sensitive offers
✅ 7 days: Standard (most products)
✅ 14-30 days: Consideration products (research required)

Don't:

❌ Use 90+ day windows (too much false matching)
❌ Use <1 hour windows (users need time to install)

3. Test Both Scenarios

Test with app installed:

Have app installed
Click link
App should open directly (no install)
Route to content

Test without app installed:

Uninstall app
Click link
Redirected to App Store/Google Play
Install app
Open app
Verify deferred deep link works (routes to content)

Test edge cases:

Switch from WiFi to cellular between click and install
Wait several hours between click and install
Click multiple links before installing

4. Track Everything

Don't just track installs, track the full funnel:

// Click event (tracked automatically by platform)
// Install event (tracked automatically by SDK)

// Track when deferred deep link matched
if (attribution.matched) {
  analytics.track('deferred_deep_link_matched', {
    campaign: attribution.utm_campaign,
    time_to_install: attribution.click_to_install_seconds
  });
}

// Track when user reaches destination content
analytics.track('deferred_deep_link_content_viewed', {
  destination: attribution.destination
});

// Track conversions
analytics.track('purchase', {
  attributed_to: attribution.campaign
});

Calculate:

Click → Install rate
Install → Match rate (should be 70-80%)
Match → Content view rate
Content view → Conversion rate
Overall click → Purchase rate

5. Use Custom Data for Personalization

Pass custom data through deferred deep links:

Link: https://go.app.com/signup?
  discount=WINTER20&
  referrer=user123&
  segment=vip&
  product_id=winter-jacket

App receives:

const attribution = await LinkForty.getAttributionData();

// Personalize experience
if (attribution.matched) {
  // Apply discount
  if (attribution.discount) {
    applyDiscount(attribution.discount);
  }

  // Credit referrer
  if (attribution.referrer) {
    creditReferrer(attribution.referrer);
  }

  // Personalize UI
  if (attribution.segment === 'vip') {
    showVIPWelcome();
  }

  // Pre-fill shopping cart
  if (attribution.product_id) {
    addToCart(attribution.product_id);
  }
}

Alternatives to Deferred Deep Linking

1. Manual Code Entry

User experience:

User clicks ad
Redirected to App Store
Installs app
Opens app
Asked to enter code: "WINTER20"
Manually types code
Gets routed to content

Problems:

❌ High friction (50% abandon)
❌ Typos common
❌ Poor user experience

When to use: Never (deferred deep linking is always better)

2. Email/SMS Verification

User experience:

User clicks ad
Asked to enter email before redirect
Email stored with link data
Redirected to App Store
Installs app
Asked to enter same email
Email matched → Routed to content

Problems:

❌ Two-step process (high friction)
❌ Users may use different email
❌ Privacy concerns (collecting email before install)

When to use: High-value flows where extra friction acceptable (enterprise B2B)

3. Universal Links (iOS Only)

User experience:

User clicks Universal Link (app already installed)
App opens directly (no deferred deep linking needed)
Routes to content

Limitations:

✅ 100% accurate
❌ Only works if app already installed
❌ Doesn't work for new user acquisition

When to use: Combine with deferred deep linking:

If app installed → Universal Links (100% accurate)
If app NOT installed → Deferred deep linking (70-80% accurate)

Summary

Deferred deep linking is critical for mobile user acquisition. It enables seamless experiences by routing users to specific content after they install your app, even though the installation process breaks the direct link context.

Key Takeaways:

Deferred deep linking "remembers" the link through the install process
Fingerprinting achieves 70-80% accuracy without privacy concerns
Essential for product-specific ads, referrals, influencer marketing
LinkForty provides production-ready deferred deep linking at a fraction of enterprise costs

Ready to implement deferred deep linking?

LinkForty offers:

✅ 70-80% attribution accuracy (fingerprinting)
✅ 5-minute SDK integration
✅ Privacy-first (GDPR compliant)
✅ Open source & affordable

Get Started with LinkForty →

View SDK Integration Guide →

Next Steps

What is Deep Linking? - Understand basic deep linking
What is Mobile Attribution? - Learn about attribution in general
Attribution Without IDFA - Privacy-first attribution guide
SDK Integration Guide - Implement LinkForty SDK

Questions?

The Problem Without Deferred Deep Linking​

How Deferred Deep Linking Works​

The Complete Flow​

Device Fingerprinting: The Technology Behind It​

What is a Device Fingerprint?​

Fingerprint Components​

Matching Algorithm​

Accuracy Expectations​

Industry Standard Accuracy​

Accuracy by Scenario​

Privacy & Compliance​

Why Fingerprinting is Privacy-Friendly​

GDPR Compliance​

Deferred Deep Linking Use Cases​

1. Product-Specific Advertising​

2. Referral Programs​

3. Email Campaigns (New User Acquisition)​

4. Influencer Marketing​

5. QR Code Campaigns​

Implementation Approaches​

1. DIY (Build Your Own)​

2. Use an Attribution Platform​

Enterprise Platforms​

Open Source & Privacy-First​

LinkForty's Deferred Deep Linking​

How LinkForty Implements It​

LinkForty Accuracy​

Getting Started with LinkForty Deferred Deep Linking​

Cloud (5-Minute Setup)​

Self-Hosted (Free)​

Best Practices for Deferred Deep Linking​

1. Always Provide Fallback Content​

2. Set Appropriate Attribution Windows​

3. Test Both Scenarios​

4. Track Everything​

5. Use Custom Data for Personalization​

Alternatives to Deferred Deep Linking​

1. Manual Code Entry​

2. Email/SMS Verification​

3. Universal Links (iOS Only)​

Summary​

Next Steps​

The Problem Without Deferred Deep Linking

How Deferred Deep Linking Works

The Complete Flow

Device Fingerprinting: The Technology Behind It

What is a Device Fingerprint?

Fingerprint Components

Matching Algorithm

Accuracy Expectations

Industry Standard Accuracy

Accuracy by Scenario

Privacy & Compliance

Why Fingerprinting is Privacy-Friendly

GDPR Compliance

Deferred Deep Linking Use Cases

1. Product-Specific Advertising

2. Referral Programs

3. Email Campaigns (New User Acquisition)

4. Influencer Marketing

5. QR Code Campaigns

Implementation Approaches

1. DIY (Build Your Own)

2. Use an Attribution Platform

Enterprise Platforms

Open Source & Privacy-First

LinkForty's Deferred Deep Linking

How LinkForty Implements It

LinkForty Accuracy

Getting Started with LinkForty Deferred Deep Linking

Cloud (5-Minute Setup)

Self-Hosted (Free)

Best Practices for Deferred Deep Linking

1. Always Provide Fallback Content

2. Set Appropriate Attribution Windows

3. Test Both Scenarios

4. Track Everything

5. Use Custom Data for Personalization

Alternatives to Deferred Deep Linking

1. Manual Code Entry

2. Email/SMS Verification

3. Universal Links (iOS Only)

Summary

Next Steps