Page 1 of 1

But you need to understa

Posted: Wed Dec 18, 2024 8:50 am
by nusaiba124
statistics The third point I would make is that you need to brush up on your statistics. So, you don't need a statistics degree. I certainly don't have a statistics degree.

nd some of the basics. Are your tests running l france whatsapp number data ong enough? First of all, you want to make sure that your test runs long enough. I talked a little bit about those sensational case studies. A lot of times when you dig into the data of those sensational case studies, they didn't run long enough to achieve significant results. So you want to make sure that you avoid false positives and false negatives to make sure that you hit a certain sample size.

Image

Now, sometimes your sample size might be as small as three days. Test for at least a week We always recommend that you run your experiment for at least a week so that you make sure that your test covers the entire business cycle, right? Because when I think about my clients and when they come in, I think about the weekends, I think about the entire week. So we always say at least a week. For example, we have a client where we always see a spike in conversions on the 15th and the end of the month, and those are paydays.