Google Analytics Review Experiments (A/B Tests)

Review Experiments – A/B Tests

Now that you’ve run an A/B experiment in Google Analytics, how to review the data? Find out in this quick video.

Learn how to setup split tests here.


In this video, we’re going to talk about how to review data from split tests or experiments in Google Analytics. If you missed our other video on how to setup experiments, you can watch it by clicking the link over to the right of the video screen or you can find the link to that video above. But in this video, we’re going to talk about the fun part of experiments, reviewing the data to figure out what you’ve learned.

To access experiments in Google Analytics, go to the sidebar navigation, click on the Behavior, then click on Experiments. This will load the table of all experiments that you have running or that you’ve run. For each experiment, you can see whether the test is currently running. If it isn’t running, you can see if there is a winner. Now it might seem discouraging if there isn’t a winner as the case with this experiment, but this kind of result helps you learn what not to test. You can try repeating this test with maybe a similar type of change or it could just be your visitors don’t really care about whatever it is you were testing.

Along with this high-level look, you can also click on the experiment to view details. Once you’ve clicked, you’ll have access to different types of reports to really help you understand the effect this experiment had. The first report is how conversions for the specific goal being tested were affected. In this case, our experiment tested a new version of a page with pricing information and the testimonial. Our original version without the pricing or a testimonial actually performed better with a 4.68-percent conversion compared to a 3.37-percent conversions rate on the variation with that information. That’s a 28-percent drop in conversion rate, so obviously, pricing information and a testimonial didn’t really work for visitors to this website. Along with looking at conversion data, you can also look at how the experiment affected site usage. In the sub-navigation, click on Site Usage. In this report, we can see that the version that included pricing information and testimonials led to slightly more pages per session and about 20-seconds longer being spent on the site.

Finally, you can also click Goal Set 1 to view how the experiment affected other goals on your website. In this case, there’s just one goal, so this report matches what was on the conversions report, but you might have other goals and you wanna know how changes your testing affect those.

If you have other questions about Google Analytics or other aspects of measuring your websites performance, please contact me.