I am doing A/B testing for ad display, where the wording on the ad is different between the two groups.

Metrics gathered:

1) # of people the ad reached

2) # of people who clicked on the ad and were brought to website

3) # of people who purchased from website

4) $ sales for each purchase.

I calculated the p-value based on the conversion rate of people who saw the ad vs people who clicked on ad.

How could I measure the significance of the difference in avg sales between groups? Is that something that can be compared even though the experience once on the website would be the same for both the test and control groups?

  • $\begingroup$ How did you calculate the p-value? Don't you have the average number of sales in each group? $\endgroup$
    – Dave
    Feb 28, 2020 at 11:18

1 Answer 1


Use Click-Through regression where A/B is a factor in the model???? https://turi.com/learn/gallery/notebooks/click_through_rate_prediction_intro.html


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.