Causal Attribution
Measure Incremental ROI of each channel with Marketing Mix Model -based incrementality factors. Compared to experiments or benchmarks, MMM-based incrementality factors are
- Updated daily (instead of being static)
- Radically more robust (They leverage all available info: modeling data, experiments, attribution benchmarks..)
What is Causal Attribution?
Causal Attribution is a form of causal inference that estimates the true incremental sales impact of each channel, by adjusting attribution-reported ROAS metrics with incrementality factors: Incremental ROI = ROAS x Incrementality factor.
Incrementality factors, sometimes also referred to as calibration multipliers, can be estimated with multiple different methods:
- Marketing Mix Modeling
- One-off experiments: Geo Lift studies, Conversion Lift Studies
- General benchmarks
Benefits of MMM-based incrementality factors
Marketing Mix Modeling -based incrementality factors are radically more robust, because they integrate all the available information about media effectiveness together using Bayesian modeling. Used info includes:
- MMM time-series data
- Experiments: Geo Lift studies, Conversion Lift Studies
- Attribution data from Last-click, Ad Platform, MTA
MMM-based incrementality factors are updated daily, whereas other methods provide static incrementality factors.
