What Should You Be A/B Testing?Posted on 09.28.12 by
Anyone reading this should know they should be testing their websites and landing pages, but if you aren’t well studied in optimization, how are you supposed to know what you should be testing? Oli Gardner has some suggestions to get you started testing your site.
1) Test the Headline - Your headline is the first things viewers will see when they land on your page, and to be most successful it should match what your viewer expected when deciding to visit your page. For example, you can test how positive or negative language performs in your headline, such as “Save Time by Downloading Now” vs. “Stop Wasting Time, Download Now”.
Headlines are also used by your audience to quickly identify if the site they came to has what they want. Make sure your headline lets visitors know what you have to offer immediately. A great way to quickly test this is to just show the headline to someone unfamiliar with your brand for 5 seconds and ask them what the page was about. If they don’t know, you have a problem with your headline.
2) Test Your Forms - Most landing pages will have some sort of form trying to gain information from visitors. Its important to remember however that you should be offering something to your visitors in exchange for this information. What you offer is up to you, but examples could be an e-book, webinar registration or whitepaper.
The question remains though, how do you test to see if your forms are effective? One test would be seeing which forms people fill out the most, if all fields were made optional. The goal is to make sure you’re offering something equal to the amount of data you need. Try to be efficient and only have forms with relevant information and test different arrangements to see what visitors respond to the most.
3) Test Your Call To Action (CTA) - A call to action is your conversion. It is how you get people to do what you want them to do, and if that isn’t happening, you may need to work on your CTA. Make it descriptive, so that your visitors will know exactly what will happen if they follow the CTA,
A way to test your CTA is by changing just one word and seeing how customers respond. Unbounce found that changing the description “order information” to “get information” led to a 38.26% increase un conversions.
While these examples will get you going, if you want a more thorough guide, you’ll also want to look into Oli’s “Ultimate Guide to A/B Testing”.