What are Geo-Experiments?
Geo-experiments are controlled tests designed to measure the true incremental impact of marketing activities by comparing test and control groups across different geographic areas.
The methodology works by:
Dividing geographic areas into two groups:
Test Group: Areas where the marketing activity will be modified
Control Group: Areas where marketing continues as usual
The algorithm selects these groups to be as similar as possible in terms of:
Historical revenue patterns
Seasonality
Market size
Consumer behavior
We use the control group to build a counterfactual using weighted synthetic controls, which creates a custom control from multiple cities weighted according to how closely their pre-test patterns match the treated geo. This synthetic control mirrors the test region's historical data, allowing us to isolate true causal impact by comparing the gap between the treated geo and its synthetic twin after the campaign starts.For example: If testing increased meta advertising in Phoenix, we might create a "synthetic Phoenix" by combining:
30% Los Angeles (similar climate and consumer patterns)
25% Denver (similar growth trends)
20% Dallas (similar demographic makeup)
15% Sacramento (similar media consumption)
10% Portland (similar seasonality patterns)
If Phoenix shows 12% growth while the synthetic control shows only 5% growth, we can attribute the 7% difference as the causal impact of the increased TV spending.
Why are they Important?
Validate model assumptions about channel performance
Measure true incremental impact of marketing activities
Calibrate the MMM model with real experimental data
Inform budget allocation decisions with empirical evidence
Challenge or confirm platform-reported metrics