Murray

Murray is a Python package for geographic incrementality testing that helps determine the true lift of marketing campaigns through advanced synthetic control methods. It generates heatmaps of Minimum Detectable Effects (MDE) across different configurations to optimize treatment selection, and provides impact analysis through counterfactual modeling.

Installation

You can install Murray using pip:

pip install murray-geo

Also, you can download the package directly from the GitHub repository:

pip install git+https://github.com/entropyx/murray.git

Prepare your data

data = pd.DataFrame({
'time': [...], # timestamps
'location': [...], # location identifiers
'Y': [...] # target metric values
})

Run analysis

results = run_geo_analysis(
    data = data,
    excluded_locations = ['mexico city', 'méxico'],
    maximum_treatment_percentage=0.30,
    significance_level = 0.1,
    deltas_range = (0.01, 0.3, 0.02),
    periods_range = (5, 45, 5)
)

Run evaluation

results = run_geo_evaluation(
    data_input=data,
    start_treatment='01-12-2024',
    end_treatment='31-12-2024',
    treatment_group=['durango','puebla','queretaro'], 
    spend=10000)

Documentation

You can check the documentation here