The Data Stack Show 228: The Machine Learning Reality Check: When AI Makes Sense for Marketing Attribution with Lew Dawson of Momentum Consulting
8 snips
Feb 12, 2025 Lew Dawson, an expert in marketing attribution and data stacks, shares his insights from Momentum Consulting. He discusses the conditions necessary for effective machine learning in marketing. The conversation dives into balancing simplicity with complexity in modeling, the challenge of ROI calculations, and how AI can transform customer data analysis. He highlights the importance of aligning data strategies with marketing goals, the intricacies of marketing attribution, and the hurdles of disparate API systems. It's a rich blend of practical advice and visionary thinking.
AI Snips
Chapters
Transcript
Episode notes
Machine Learning for Attribution
- Consider machine learning for dynamic weighting in attribution modeling.
- This technique assigns weights based on training data and can optimize attribution.
When Machine Learning Makes Sense
- Machine learning for attribution is often only worthwhile for large businesses.
- These businesses have exhausted simpler optimization opportunities and possess sufficient data and stable business models.
Practical vs. Theoretical Saturation
- Consider the difference between practical and theoretical saturation in marketing channels.
- Practical saturation can occur quickly with bad creative, while good creative expands the theoretical limit.