Check out this Medium article for an introduction to A/B Testing and this tool.
|chance that the variant performs better than the default
Congratulations! We are very confident that the variant improves upon the baseline and would recommend going forward with the variant.
The variant likely improves upon the baseline, but we can't be certain. We would recommend running more users through this test if possible, but if this is a low-risk change it might be okay to go ahead with the variant.
We cannot be confident that the variant improves upon the baseline. It may be the case that running the test with more users will yield better results, but it could also be the case that the variant is not an improvement. We would not recommend going forward with the variant at this time
We are not at all confident that the variant improves upon the baseline. We do not recommend going forward with the variant at this time.
These results are only what we can do based on generic information. We can always improve the analysis by incorporating more prior information into our assumptions. If you're interested in tailored results and more analytical power, drop us a line at [email protected].