Optimal control (test) group size for campaigns calculator
This case study shows a particular campaign management optimisation. The balance of optimal control group sizing is an important part of running campaigns. On one hand, too small of a group can cause further incorrect campaign analysis. On the other hand, having too big of a control group means that company is losing its money. Thus, having an optimal proportion of targeted and control groups is an essential point for effective campaign management.
How big should the control (target/test) group be? How many subscribers should we hold aside as a control group? How to determine the optimal control group sample size? The following calculator will show you.
How to use the Optimal Control Group Size Calculator
Please specify your total count of customers to whom you wish to make an offer. Then, specify the response of the offer you expect in your control and target groups after the offer reaches customers. Please use numbers between 0.0001 and 0.9999 (which is 0.01% to 99.99%). Please use dot, not a comma. You may also change the significance level and power to get more precise results, however the default ones are good enough for further campaigns analysis.
The above calculations are based on the inverse cumulative distribution function (CDF), or quantile function associated with the standard normal distribution. According to Wikipedia, the normal distribution CDF and its inverse are not available in closed form, and computation requires careful use of numerical procedures. However, the functions are widely available in software for statistics and probability modeling, and in spreadsheets. In Microsoft Excel, for example, the probit function is available as normsinv(p). In computing environments where numerical implementations of the inverse error function are available, the probit function may be obtained as:
Due to the complicated calculations this page is not available in PDF.