AgentSkillsCN

Evaluate Marketing Attribution

评估营销归因分析

SKILL.md

Skill: Evaluate Marketing Attribution

Domain

technology

Description

Evaluates marketing campaign effectiveness using multi-touch attribution models to optimize marketing spend and channel allocation.

Tags

marketing, attribution, analytics, ROI, digital, campaigns

Use Cases

  • •Campaign ROI analysis
  • •Channel effectiveness
  • •Budget allocation
  • •Conversion path analysis

Proprietary Business Rules

Rule 1: Multi-Touch Attribution

First-touch, last-touch, and algorithmic attribution models.

Rule 2: Channel Contribution

Individual channel impact on conversions.

Rule 3: ROI Calculation

Return on marketing investment by channel.

Rule 4: Budget Optimization

Optimal budget allocation recommendations.

Input Parameters

  • •analysis_id (string): Analysis identifier
  • •conversion_data (list): Conversion events
  • •touchpoint_data (list): Marketing touchpoints
  • •spend_data (dict): Marketing spend by channel
  • •attribution_model (string): Model type to use
  • •lookback_window (int): Attribution window days

Output

  • •channel_attribution (dict): Attribution by channel
  • •conversion_paths (list): Common conversion paths
  • •roi_by_channel (dict): ROI metrics
  • •optimization_recommendations (list): Budget recommendations
  • •model_comparison (dict): Multi-model results

Implementation

The evaluation logic is implemented in attribution_evaluator.py and references data from CSV files:

  • •attribution_models.csv - Reference data
  • •channel_categories.csv - Reference data
  • •conversion_windows.csv - Reference data
  • •roi_benchmarks.csv - Reference data
  • •incrementality_factors.csv - Reference data
  • •funnel_stages.csv - Reference data
  • •parameters.csv - Reference data.

Usage Example

python
from attribution_evaluator import evaluate_attribution

result = evaluate_attribution(
    analysis_id="ATT-001",
    conversion_data=[{"id": "CV-001", "value": 100, "timestamp": "2025-12-15"}],
    touchpoint_data=[{"conversion_id": "CV-001", "channel": "paid_search", "timestamp": "2025-12-10"}],
    spend_data={"paid_search": 50000, "social": 30000, "email": 10000},
    attribution_model="linear",
    lookback_window=30
)

print(f"Paid Search Attribution: {result['channel_attribution']['paid_search']}")

Test Execution

python
from attribution_evaluator import evaluate_attribution

result = evaluate_attribution(
    analysis_id=input_data.get('analysis_id'),
    conversion_data=input_data.get('conversion_data', []),
    touchpoint_data=input_data.get('touchpoint_data', []),
    spend_data=input_data.get('spend_data', {}),
    attribution_model=input_data.get('attribution_model', 'linear'),
    lookback_window=input_data.get('lookback_window', 30)
)