What is Custom Weight?

Custom weights allow you to override default attribution distribution across user touchpoints by assigning specific percentages to first, early, middle, late, and last interactions in the customer journey.

Instead of using standard attribution models that apply fixed credit distribution, custom weights give you control over how much credit each position in the user journey receives, enabling attribution models tailored to your specific business context.

Understanding custom weights helps marketers create attribution models that reflect their actual customer behavior patterns rather than relying on generic models that may not match their reality.

In attribution modeling, the customer journey is divided into five positions: first, early, middle, late, and last touchpoints. By default, first-click and last-click models have predefined weight distributions. Custom weights let you modify these to match your business needs.

For example, if you know that your customers typically research extensively before purchasing, you might want to give more credit to early touchpoints. Custom weights make this possible.

How Custom Weights Work

Every attribution model distributes credit across touchpoints in the customer journey. The journey is divided into five positions:

• First: The initial touchpoint (first interaction)
• Early: Touchpoints in the first third of the journey
• Middle: Touchpoints in the middle third
• Late: Touchpoints in the final third (before last)
• Last: The final touchpoint before conversion

Default Weight Distributions

First-Click Model (emphasizes discovery):

• First: 35%
• Early: 25%
• Middle: 25%
• Late: 15%
• Last: 15%

Last-Click Model (emphasizes conversion):

• First: 15%
• Early: 10%
• Middle: 15%
• Late: 25%
• Last: 35%

Example Custom Weight Configuration

{
  "first": 0.40,
  "early": 0.20,
  "middle": 0.20,
  "late": 0.15,
  "last": 0.05
}

This configuration gives 40% credit to the first touchpoint, reflecting a business where initial discovery is critical, and only 5% to the last touchpoint, acknowledging that the final interaction is less influential.

When to Use Custom Weights

Custom weights are valuable when:

1. Your customer journey is predictable: You know which stages matter most
2. Generic models don't fit: First-click and last-click don't reflect your reality
3. You have historical data: You've analyzed what actually drives conversions
4. You want to test hypotheses: Experiment with different weight distributions
5. Your business has unique characteristics: Long sales cycles, brand-driven purchases, or research-heavy products

Example Scenarios

B2B SaaS with long sales cycles:

{
  "first": 0.30,
  "early": 0.30,
  "middle": 0.25,
  "late": 0.10,
  "last": 0.05
}

Early research and initial awareness matter most.

E-commerce with impulse purchases:

{
  "first": 0.10,
  "early": 0.10,
  "middle": 0.15,
  "late": 0.30,
  "last": 0.35
}

Recent interactions and the final push matter most.

Brand-driven products:

{
  "first": 0.45,
  "early": 0.25,
  "middle": 0.15,
  "late": 0.10,
  "last": 0.05
}

Initial brand exposure is critical, later touchpoints are reinforcement.

Custom Weights vs Standard Models

Standard attribution models work well for most cases, but custom weights let you optimize attribution for your specific business model.

How Custom Weights are Applied

When a user has multiple touchpoints before conversion, Hardal:

1. Identifies all touchpoints in the conversion path
2. Assigns positions (first, early, middle, late, last) based on journey length
3. Applies your custom weights to each position
4. Normalizes to 100% to ensure total credit equals one conversion
5. Distributes credit across channels/campaigns based on weighted positions

Example with 5 Touchpoints

User journey:

1. Google Ad (Day 1)
2. Facebook Ad (Day 3)
3. Email Click (Day 5)
4. Organic Search (Day 7)
5. Direct Visit (Day 8) → Conversion

With custom weights {first: 0.40, early: 0.20, middle: 0.20, late: 0.15, last: 0.05}:

• Google Ad (first): 40% credit
• Facebook Ad (early): 20% credit
• Email (middle): 20% credit
• Organic (late): 15% credit
• Direct (last): 5% credit

Combining Custom Weights with Other Features

Custom weights can be combined with:

• Channel Quality Scores: Adjust weights further based on channel quality
• Exclude Self Domain: Remove internal referrals before applying weights
• Attribution Window: Define lookback period before weighting touchpoints
• Custom Rules: Transform data before applying weights

This creates highly sophisticated attribution models tailored to your business.