Calculating A/B Test Experiment Duration

At Amadesa, our main focus is the success of our clients.  Part of planning a successful testing program is considering the available traffic against the number of versions you would like to experiment with.  As you add more versions to the test, there is a greater chance that more traffic will be needed to reach statistical validity and decide on the winner for the test.

Our Test Calculator enables you to try different combinations of versions and volume of traffic to give you an idea of how many days it will take for your test to reach validity.

To Complete Your Test, Simply Enter The Following
Information Into Our Test Calculator:

Average daily visits:

The expected average number of daily visits to the testing area

Current conversion (%):

The current conversion rate you are measuring for the funnel.

Dedicated Traffic (%):

The percentage of the total traffic you plan on dedicating to the experiment.

Number of versions:

The number of different versions you plan on using in the experiment (including the control version).

Expected lift (%):

The relative improvement, in percentage, you expect the optimized content to achieve.

Confidence level:

The accuracy you require for calculating the statistical validity. A higher confidence level means that there will be less of a chance for the results to change over time.

99.9% 99% 98% 95% 90% 85% 80% 75%

Calculate Now

Expected duration (days):