A synthetic control is an algorithm-built counterfactual. It uses historical data to create a weighted blend of control regions that closely matches the test regions' pre-experiment behavior.
Let a tool blend multiple control cities instead of hunting for one perfect match. For example, Houston, Charleston and Savannah might together form a synthetic New Orleans.
The tighter the pre-experiment fit between the synthetic control and the test markets, the more confidence the results deserve.
It replaces Matched market testing inside Geographic experimentation and is used in the geolift-test-workflow.