Multi-Touch Attribution

Understand which marketing channels and touchpoints drive revenue with sophisticated attribution modeling.

Overview

Attribution analysis assigns credit to the marketing touchpoints that contributed to a conversion. Unlike simple last-click attribution, multi-touch models recognize that customers interact with multiple channels before purchasing.

Attribution Models

First-Touch Attribution

Assigns 100% of credit to the first touchpoint in a customer's journey.

Best for: Understanding awareness channels and initial discovery sources.

Example: User sees a Facebook ad (first touch) → reads blog posts → signs up via Google search. Facebook gets 100% credit.

Last-Touch Attribution

Assigns 100% of credit to the final touchpoint before conversion.

Best for: Understanding which channels close deals and drive immediate action.

Example: User sees a Facebook ad → reads blog posts → signs up via Google search (last touch). Google gets 100% credit.

Linear Attribution

Distributes credit equally across all touchpoints.

Best for: Recognizing all channels that contributed to the conversion journey.

Example: 3 touchpoints (Facebook, Blog, Google) each get 33.3% credit.

Time-Decay Attribution

Assigns exponentially more credit to recent touchpoints using a 7-day half-life decay function.

Best for: Balancing awareness and conversion channels, with emphasis on recent interactions.

Example: Touchpoint from 7 days ago gets 50% of the weight of today's touchpoint. Touchpoint from 14 days ago gets 25%.

How It Works

Touchpoint Tracking

Attribution requires tracking UTM parameters and referrer information in your event metadata:

{
  "eventName": "page_view",
  "userId": "user_123",
  "metadata": {
    "utm_source": "google",
    "utm_medium": "cpc",
    "utm_campaign": "winter_sale",
    "utm_content": "ad_variant_a",
    "utm_term": "analytics software",
    "referrer": "https://google.com/search"
  }
}

Revenue Events

Attribution analyzes events with revenue data to calculate ROI:

{
  "eventName": "purchase",
  "userId": "user_123",
  "metadata": {
    "revenue": {
      "amount": 99.00,
      "currency": "USD"
    }
  }
}

Attribution Window

Serla uses a 30-day attribution window, meaning touchpoints up to 30 days before a conversion are considered.

Using Attribution Analysis

Via UI

Navigate to Attribution in the sidebar
Select an attribution model (first-touch, last-touch, linear, time-decay)
Choose a time period (7, 14, 30, or 90 days)
View revenue attribution by source
See metrics:
- Total revenue attributed
- Total conversions
- Average revenue per conversion

Via API

GET /api/analytics/attribution?projectId={id}&model=time-decay&days=30

Response:
{
  "model": "time-decay",
  "period": "30 days",
  "totalRevenue": 24850.50,
  "totalConversions": 347.25,
  "attribution": [
    {
      "source": "google / cpc",
      "revenue": 12400.00,
      "conversions": 156.5,
      "percentage": 49.9
    },
    {
      "source": "facebook / social",
      "revenue": 6200.00,
      "conversions": 98.25,
      "percentage": 24.9
    }
  ]
}

Interpreting Results

Source Attribution

Sources are identified from UTM parameters and referrers:

utm_source / utm_medium: "google / cpc", "facebook / social"
utm_campaign: If no source/medium, campaign name is used
referrer domain: If no UTM params, referring site is used
direct: No UTM params or referrer

Revenue vs Conversions

Revenue: Total monetary value attributed to this source
Conversions: Number of conversions attributed (may be fractional for multi-touch models)
Avg Revenue/Conversion: Revenue divided by conversions

Model Comparison

Compare models to understand channel roles:

High first-touch, low last-touch → Awareness channel
Low first-touch, high last-touch → Conversion channel
High across all models → Multi-stage contributor

Use Cases

Marketing Budget Allocation

Identify which channels drive the most revenue and allocate budget accordingly. Compare cost per acquisition across channels using attributed revenue.

Campaign Performance

Track individual campaigns to see which creative, messaging, or targeting drives conversions.

Channel Mix Optimization

Understand how channels work together. If Facebook drives awareness but Google converts, maintain both investments.

Customer Journey Analysis

Identify common touchpoint sequences that lead to high-value conversions. Replicate successful patterns.

Best Practices

UTM Tagging

Tag all marketing links with UTM parameters
Use consistent naming conventions:
- utm_source: google, facebook, twitter, newsletter
- utm_medium: cpc, social, email, affiliate
- utm_campaign: descriptive campaign name
Document your UTM taxonomy
Use a URL builder tool for consistency

Revenue Tracking

Track revenue on conversion events (purchase, subscription, upgrade)
Include currency for multi-currency businesses
Use actual transaction amounts, not estimated values
Track both initial and recurring revenue

Model Selection

First-touch: For top-of-funnel optimization
Last-touch: For bottom-of-funnel optimization
Linear: For overall channel contribution
Time-decay: For balanced, recency-weighted analysis (recommended)

Analysis Strategy

Run reports across multiple models to get complete picture
Compare attributed revenue to actual costs (ROAS)
Segment by customer type (new vs returning)
Track changes over time to measure optimization impact

Common Scenarios

Scenario 1: Awareness vs Conversion

First-touch: Facebook (60%), Google (30%), Direct (10%)
Last-touch: Google (70%), Facebook (20%), Direct (10%)

Insight: Facebook drives awareness, but Google converts. Maintain Facebook for top-of-funnel, optimize Google for conversion.

Scenario 2: Email Nurture Value

Linear: Email (40%), Organic (35%), Paid (25%)
Last-touch: Organic (50%), Paid (35%), Email (15%)

Insight: Email contributes significantly mid-funnel but doesn't often close deals. Continue email nurture; don't expect last-click credit.

Scenario 3: Long Sales Cycles

Time-decay: Webinar (45%), Email (30%), Organic (25%)
First-touch: Organic (60%), Webinar (25%), Email (15%)

Insight: Users discover via organic, but webinars drive recent engagement before conversion. Invest in webinar content.

Limitations

Attribution Window

The 30-day window may not capture the full customer journey for products with longer sales cycles. Consider extending analysis periods.

UTM Persistence

UTM parameters are tracked per event. If users don't persist UTMs across sessions, some touchpoints may be missed. Implement client-side UTM persistence.

Cross-Device Tracking

Attribution requires userId to connect touchpoints. Anonymous users or cross-device journeys may show fragmented attribution.

Offline Touchpoints

Offline interactions (billboards, TV ads, word of mouth) aren't captured in digital attribution. Supplement with surveys or promo codes.

Advanced Tips

Custom Attribution Windows

For B2B or high-consideration products, track touchpoints over 60-90 days by adjusting the days parameter in API calls.

Channel Grouping

Group similar sources for clearer insights (e.g., all social channels, all paid search).

Incremental Testing

Pause specific channels temporarily to measure incremental lift in attributed revenue. True incrementality differs from attribution.

Cost Integration

Import ad spend data and calculate ROAS (Return on Ad Spend) by dividing attributed revenue by channel costs.