Optimize or Die!
If website optimization is not a part of your marketing strategy, then you are missing out on some excellent opportunities. Small changes can make big differences. Sometimes, simply changing the color of a submission button or image on a web page can yield significant differences in results.
For many, the thought of creating an A/B test to experiment with an idea is rather daunting. It doesn’t have to be.
I have put together a simple A/B Test Worksheet that you can use to brainstorm website experiment opportunities at your next marketing meeting. Simply print out the worksheet, talk about what you might want to test, fill in the answers, and set it up as a Google Analytics Content Experiment. All of the questions from GA Content Experiments are incorporated into the worksheet, so it will be easier to set up the experiment.
Get Ready for your Google Analytics A/B Test!
In addition to the A/B test worksheet (below), here are a few tips to get you started.
- Start small. Test a simple element to see how the experiments work. After you find out how simple it is to set up, you can move on to more complicated tests.
- Test only one element at a time. While it may be tempting to test everything all at once, it is best to start with testing one element before moving on to the next piece. For example, you may want to test a headline against two other versions. In that situation, create three identical pages where the only difference is the headline. The same is true for images, forms placement, link colors, etc.
- Keep the number of variations small. You do not have to only test two different versions of a page, you can test more than that. However, keep in mind that you rely on the page receiving enough traffic to show the difference between versions. It is best to start with only a few versions of the page. If you are confident that the page receives a lot of traffic, you can increase the number of versions to test, but remember to still only test one element at a time.
- Clearly state the purpose of the test. Make sure that the purpose of the experiment is related to a business objective. For example, you can test two versions of a page to see which version contributes to more Inquiry Form Goal Completions. That is a test with a purpose. Do not just create a test for the sake of creating a test. The objective does not need to be of immense significance, but there should be a purpose.
- Keep testing. After a winner is chosen, use the winning version as the new base page (original) for another test. Website optimization is a process and it often takes many tests to determine an optimal version.
Download the A/B Test Worksheet and start brainstorming about what test you will set up.
4-week online course: Web Analytics for Higher Ed
(asynchronous with weekly lessons and assignments)
ONLY 10 SEATS per session