.png?width=667&height=667&ext=.png)
5 Dead-Simple Kentico A/B Tests for Website Optimization
Introduction
You might know by now that Kentico's Online Marketing Solution allows you to optimize your website content with an advanced set of digital marketing tools. In my opinion, the easiest and most simple way to optimize your page content is to utilize A/B testing. With the release of Kentico 8 optimizing your pages utilizing A/B testing has become even easier to do.
If you are not sure how to do an A/B Test in Kentico, the full example can be watched via the Kentico 8 A/B Test Feature Series that Thom Robbins has put together over on Kentico DevNet. It's a really great walkthrough on every step you need to take in Kentico to create your A/B Test. Now on to the dead-simple Kentico A/B Test examples.
Test Call to Action Buttons
Why not start with the classic A/B Test example? Changing the color of the Call to Action button on your landing page is one great way to see which color can attract more clicks. This test usually involves having one variant with a Green button versus one variant with a Red button. You can also try this test with moving your Call to Action button to either the Left column or Right Column of your pages. For Kentico this test is actually the easiest to achieve because it can be done purely by adding a Page Variant and changing content on the Page tab.
 |
 |
Test Navigation Menu
The main menu that your website runs allows people to easily find content within a few clicks. At least in theory that is what it is supposed to do. Why not test it out to see if you can improve the usage and overall user experience by testing the sequence of main menu items. Since the left most position matters most normally, you could test out the results of re-ordering your menu elements from time to time. Maybe switching the order of the Products link and the Categories link could have a positive affect on how website visitors navigate through your catalog.
This test is slightly more difficult than the first one. To create an Test for a webpart like the CMSListMenu you actually have to use the MVT Test Feature of Kentico. I know technically it might be considered cheating because it is not a pure A/B Test, but that is the easiest way to accomplish webpart variants on a Page Template in Kentico. The main reason being this example is not consider Page content, but instead more structure / layout content.
 |
 |
Test Images or Sliders
This scenario allows you to test if different images attract more clicks or spark a different level of engagement with your visitors. The A/B Test for images is pretty much setup exactly like the first test in this list. You create a Page Variant and choose a different image for content in the same way that you would swap out green or red buttons. This Test can help you answer whether the more friendly and inviting image is of a single smiling person, or the version with a group of people working and/or facing a monitor.
 |
 |
Test Your Headlines
My career started out as the Layout Editor for my university's newspaper. It was my job to take each reporters article, lay it out in a desktop publishing software, and adjust the content so that it would fit on the page. Part of that job also involved taking the summary sentence of the article and turning it into a headline. I can tell you that making a clear, short, and engaging headline is not an easy thing to do. Yes, this was back in the day when newspapers were not published online.
Utilizing A/B Testing on your headlines can allow you to figure out how to go from a long and wordy headline to a detailed, short and sweet version. Again in Kentico, this is an easy thing to do by creating a Page Variant that has an alternate Editable Text webpart for the headline.
 |
 |
Test Your Forms
The final dead-simple A/B Test on this list revolves around the On-line Form engine in Kentico. It is a well known and understood fact that the more you require your users to enter into forms, or the harder you make it for them, the less excited those users will be about making a form submission. At BizStream we recommend to our customers that the smallest form possible for things like user registrations, preference subscriptions, or warranty registrations, the better off the results will be. However, you can use A/B Tests to get even more granular for things like how many fields should the form have, how complex the questions are (question labels and possible answers), is the default submit button enough (or do you need a bigger button with nice green background). All of these things can be easily done with the On-line Form engine in Kentico.
To do so I recommend creating one or more Alternate Forms for your On-line Form and adjusting the Form Controls to see if your multiple choice questions with multiple dropdowns get answers, or making a few single choice questions makes it easier and quicker for your users to complete the form. Once the Alternate Forms are created you can use the Insert Form button on the WYSIWYG editor to place an Inline Widget in your content.
Pro Tip: If you don't see the Online-Form Widget's Alternate Form selector as a property, you can easily add it by going to the Widget's properties and clicking the Display On Editing Form attribute to checked. This will allow you to choose an Alternate Form on each placement of the On-line Form.
 |
 |
The Results and One Last Tip
After you setup all your tests you can see a nice summary of them in the Admin section of Kentico. One last tip though. As you can see in the screen shot below the system also tracks conversions by default. Make sure you setup a unique conversion per Page Variant in your AB Test! This will give you the most critical information you need to determine which Variant should win. If Both A and B have the same conversion it will not be that obvious to choose a winner.
So there you have it. This is my list of 5 dead-simple A/B Tests that you can easily do with Kentico. I’d love to know how the community is using the A/B Tests feature in Kentico. Feel free to share your ideas and tests via the comments in this blog post. That’s all for now!
