This tool takes a Bayesian approach to calculating the results of your a/b tests. If you would like to know more about what this means, here is the blog about how this tool works. The output you get will be in the form of the percentage chance of each test arm being higher value, along with the average improvement from one to the other. This test does not give you a measure of statistical significance - instead, it gives you the tools to make a business decision about whether to chose one of the test arms or to collect more data.
If you aren't sure what to put in the 'prior' fields, then it's recommended just to put 1 in each - the calculations will be perfectly valid.