Making the most of your predictions.
Baseline: Historical performance of the campaign. Used to compare the performance of the predictions. It could include conversion rate, average donation size, lifetime giving, days until conversion, etc.
A/B Test: Two samples of the predictions are compared using identical campaigns to determine if the model is valid.
Sample Size: A group of predictions selected from all the predictions and considered representative of the real predictions for that campaign. It is important to use a large enough sample size based on the expected response rate of the campaign. Learn more about sample sizes here.