Marketing Attribution and Mix Modeling

Learn the techniques to accurately measure the impact of your marketing and advertising efforts.

Introduction

• Measuring marketing performance

1. Multi-touch Attribution Models

• Last-click attribution: The default model
• Time decay and conversion lags
• Linear attribution: Treating all touches equally
• First-click models: From awareness to acquisition
• Position-based models and assigning credit
• Data-driven attribution and machine learning
• Click windows and view-through conversions

2. Marketing Mix Modeling

• Before and after an event: Trend analysis
• Linear regression with a single variable
• Variables with positive and negative correlations
• Multivariable regression: Building your marketing mix model
• Feature transformation with diminishing returns and adstocks
• Statistical tests to validate your model's accuracy
• Forecasting future scenarios for planning

3. Incrementality and A/B Testing

• A/B testing for statistical significance
• Bandit testing: Optimizing for results over accuracy
• Geo and lift testing to prove incrementality
• "How did you hear about us?": Surveys and panel studies
• Working with multiple attribution methods

Conclusion

• Continuing to improve your model accuracy