In this newsletter I selected a couple of articles about the question: Bayesian versus Frequentist statistics for A/B testing. Both ways have pros and cons, the most important is that you understand the difference.
Currently I use Bayesian statistics to evaluate A/B test. The results are easier to explain, but also leave more room for discussion because there is no binary outcome (yes or no).
Bayesian statistics gives a percentage as result, therefore it’s important to agree on the percentage when you call a test a winner before you start with A/B testing.
Next to this there is an awesome summary of all the talks from CXL Live 2019 in this newsletter and a discount code for Emerce Conversion 2019, where I am one of the speakers this year!
And last but not least I found a great thread on Twitter this morning about product validation in combination with A/B testing, with Lukas Vermeer and Chad Sanderson involved.
Enjoy your week.