A/B tests with Google Optimize in online marketing

With Google Optimize you can display several website versions in parallel to different users for test purposes and measure their usage behavior. This way you can see which version performs best.

What is Google Optimize and how does your website benefit from it?

A/B testing, also known as split testing, is a way of comparing two different versions of a website or landing page to see which performs better in terms of conversions, conversion rate, bounce rate, dwell time, and scroll depths. With Google Optimize, you can easily run A/B tests to find out which changes to your website would make the most sense to get even more results.

Run A/B tests with Google Optimize

To run an A/B test with Google Optimize, you must first create a Google Optimize account and link it to your Google Analytics account, which you use to track your website. Then you need the Google Optimize Browser Plugin for Google Chrome. You can install this for free. Last but not least, you have to install a Google Optimize script on your website – similar to the installation of Google Analytics, this must be inserted into the head area of your website.

Google Optimize’s frontend editor

You can then create A/B tests in which you can access your website externally with Google Optimize’s frontend editor to edit HTML and JavaScript structures. This is how you create your first test variant, which can be tested against the original of your website or landing page during delivery. You can also set how many visitors you want each version of the page to see (as a percentage), and Google Optimize will automatically serve the two versions to the appropriate visitors once they’re on your site.

Set KPI measurement targets with Google Optimize

You can set specific measurement targets for the evaluation that serve as success criteria, such as the conversion rate or the bounce rate. As soon as the test is completed, you can see the results directly in Google Optimize (how many conversions a version has generated vs. the original), but you can also call up the detailed and extensive data on the test in Google Analytics, as this is where it comes in. Here you can see all sorts of different metrics for each test that are also measured and do not appear in the Google Optimize report.

Run AA test with Google Optimize

It is strongly recommended to only run one test on a landing page and not several in parallel, otherwise the results will no longer be comprehensible. In addition, it is best to only test one test variant per test, because several variants also get in the way in terms of design at the same time. You should also carry out an AA test after a successful test result – i.e. test the successfully determined element in the test version on comparable landing pages to see whether it is generally comparably successful there, because only in this case do you know whether this test version has themes – and represents an improvement of your general UX design across target groups.