// glossary
Geo-Lift.
Geographic incrementality testing
Geo-lift testing is an experimental method that splits geographic regions into treatment (ads on) and control (ads off) groups to measure a channel's true incremental sales effect. The post-cookie gold standard for causal measurement.
// detail
As cookie loss weakens MTA, MMM alone isn't sufficient for strategic decisions. Geo-lift measures channel effect experimentally — 'what happens to sales if we turn this off?'
Flow:
- Split geographic regions (DMAs in the US, states in the EU, provinces in TR) into two groups.
- Treatment group runs the channel hot (or full); control runs it off or reduced.
- 2 weeks pre-period (calibration) + 4 weeks treatment + 1 week wash-out.
- Synthetic control builds a counterfactual 'what would have happened' curve from the control group.
- Treatment-group sales − synthetic-control sales = pure incremental lift.
Open source: Meta GeoLift (R), Google CausalImpact (R/Python), LightweightMMM.
Cost: ~5-10% of monthly ad budget (opportunity cost). Output: a cookie-independent measured lift number; used to cross-validate MMM and de-risk large budget decisions.
Example: A brand's MMM estimated Google Search ROI at 3.5x. A 4-week geo-lift: treatment 8 DMAs with +50% spend, control 70 DMAs flat. Result: incremental ROAS 4.25x, p=0.03 (significant). MMM was slightly underestimating; budget scaled up.