A/B Testing Guide for Beginners

A/B testing can be a game changer for your business when you do the right tests. 

You can find tons of A/B test ideas when you search on the Internet yet none of them could tell you what tests will be meaningful to your business. 

The meaningful tests that are worth running are based on your own data.

What should you test?

Here are some ways to collect quantitative and qualitative data to find ideas for your A/B test experiments. 

Analytics analysis
  • Start with looking at your analytics analysis as it provides insights into how your customers behave. 
  • A conversion funnel that shows you the biggest leak on your website is a good place to start testing. 
  • Scrolling heatmap that shows where your visitors spend the most time reading your content can shed light on their interest areas. 
Customer interviews
  • Nothing is more valuable than speaking to your actual customers to truly understand why they really buy from you.
  • Find out the reason your customers choose you over your alternatives, or the issues they are trying to solve with your solution. 
  • Pay attention to the words they are using and these words could be your next copywriting test ideas.  
Onsite surveys
  • Onsite surveys provide rich insights into your website visitor's intentions and concerns. 
  • An onsite survey might ask the visitors' goals for browsing the page, or ask visitors who have been on the check-out page for a while if there’s anything holding them back from making a purchase today. 
  • For instance, you can test whether adding a reassurance copy to address visitors' concerns on the check-out page would affect your order completion rate.
User testing
  • You can watch how real people perform a task on your site in a paid and controlled experiment. 
  • For example, you might ask them to locate a specific product in the $60 to $100 range and add it to their cart. While they’re performing the task, they narrate their thoughts and actions out loud.
Visitor recordings 
  • Visitor recordings are similar to user testing, but now you’re watching real visitors with real money and real purchase intentions.
  • You can see your actual visitors' complete journey and discover if they have trouble finding anything or where they get frustrated.

Turn insights into hypotheses

In A/B testing, you need to validate a hypothesis, not an idea.

Here's a simple formula by Craig Sullivan that can help turn your idea into a hypothesis: 

  1. As you see (insert data you get from qualitative and quantitative research)
  2. You assume that (variation you're testing) will lead to (the result you expect) 
  3. You will evaluate it using (data metric)

Setup your A/B Testing Campaign

Once you have your hypothesis, all you have to do is setup the variants by making changes to the original version and assign a goal to measure success.

Howuku makes it easy and simple for you to make changes directly to your site with a WYSIWYG Editor (What You See Is What You Get).

If you would like to see a demo, you can always schedule something in my calendar here

How to analyze A/B test results

Analyzing A/B test results is not only about determining if the variant was a winner or a loser. 

It's important to highlight here the need for segmenting your A/B test results data. This is because a variant might be a loser in general, yet chances are it might convert well with at least one segment.

For example, a variant with more text might be a loser overall yet it could perform better in a certain segment says people using a web browser.   

You can discover hidden insights by learning about your losers which is the most overlooked part of A/B testing. 

With Howuku, you can further segment the conversions of your variants to gain rich insights into your winner and losers. 

Click on "View conversions" and use the filter at the top right to segment your data and find valuable insights you can use for future tests and in other areas of your business.

Did this answer your question? Thanks for the feedback There was a problem submitting your feedback. Please try again later.

Still need help? Contact Us Contact Us