designer#
ExperimentDesigner: posterior-aware experiment design for lift tests.
Recommends which marketing experiment to run — which channel, at what spend level, for how long — based on a fitted MMM’s posterior uncertainty about channel response functions. The v1 scope is national-level experiments on multiple spend channels, analysed via Interrupted Time Series (ITS).
The designer uses the MMM’s own computational graph to evaluate channel contributions, so it automatically supports any adstock/saturation combination without reimplementing the transformation formulas.
Classes
|
Posterior-aware experiment designer for marketing lift tests. |