Conversion rates play an important role in the world of online business, so various methods are used and lots of resources are spent in order to enhance them. One way of increasing the conversion rate of your website is by implementing what is known as A/B split testing. Basically, you test whether the website’s visitors prefer one design over another. Ultimately, this provides you with constructive feedback on how to communicate with your audience more efficiently. Here’s how you should use A/B split testing in order to improve your conversion rate.


Testing helps you lower the risk and pushes you in the right direction

This is the most obvious benefit of split testing – you get the idea what users like and what they don’t, and you can use this information when designing your website. It is a procedure that challenges how we think about web design and contributes to our knowledge. However, it is not that simple, there are a lot of ways for you to get a wrong impression, or use testing in a wrong way. In order to more accurately deduce what your website should look like, you need to know what to test, when to test, and decide on the tools you will use for testing.

Do not rush the testing procedure

The point of testing is not to determine the winner between two pages, but to determine the truth. In other words, you need a solid amount of data, which is only possible when you have a large number of visitors. It would be wise to take 100 conversions as the minimum viable feedback for your future plans regarding the design. To be more certain go with 200, and if you truly want to have a safe bet, then 400-500 conversions is definitely the way to go.

Another potential mistake is to end the testing period too soon. So, to be certain in your findings make sure the testing lasts at least a week. Even if your reach 400 conversions within one day and obtain the desired results, do not stop your testing period. First, you will see whether there are great fluctuations in conversion rates – if yes, then try to determine the cause of sudden rises and drops. A prolonged testing period allows you to monitor external factors, and can offer some valuable input on the topic. This will give you a better overview of the timetable, so that you can know when to post certain news and create buzz around new content.  In other words, you obtain information when it is possible to incite greater engagement.

Things you should test


This is the most crucial part of testing, knowing what to test. There are certain factors which drive conversion and they are: persuasion, usability and trust.

  • Persuasion measures how the elements of your website work together to inspire the visitor to take action, like buy, post something, contribute etc.
  • Usability points out just how much your site is user-friendly, or how intuitively a visitor can determine what he or she needs to do in order to execute the desired action.
  • Trust shows just how comfortable a new customer or viewer feels when navigating through your site, and whether he or she finds that you are a reliable provider, so that they can make a purchase with certainty.

Now here are the elements that have the greatest impact on the before mentioned factors that drive conversion:

  • Headlines – considering the fact that you have approximately 3-5 seconds to grab someone’s attention, the headline has a huge impact on how you will execute that goal. Of course, you will need to have something to back up the buzz, otherwise you only look like the boy who cried wolf. Also if you have great content but poor headline, you efforts to create the content will most likely be nullified. A word of advice, try to make your headline incite fear or curiosity and test whether longer but descriptive or shorter but shocking headlines provide better conversion.
  • Pricing – this can be very useful, for example if you lower the price for 5% but attract 20% more customers, then you might be onto something. You can use testing to determine what kind of price will yield more income. Also, it allows you to determine which pricing highlights are more effective than the others.
  • Call to action – here you examine what kind of approach resonates better with your target audience, when guiding them to take action on your website. Whether the audience likes witty or mechanical instructions, e.g. a “Make a Purchase” button vs. “summon the desired object” button.

Use it when you have solid traffic


If your websites sells 2-3 unites over the course of single month, an A/B split test cannot yield any valuable input. If things get 15-20% better there is no way to know, since you only get a limited amount of conversions in the first place. In order to be able to conduct a valid test you need to have a conversion rate which can yield some kind of results. If you only have very few conversions, do not even go with tests, no one needs to tell you your site is underperforming, so just go with option B and see if things change. If nothing significant happens, find other options and throw aside the previous two, since there is nothing worth salvaging.

Decide what you want to test and on tools you will use

If you are a small business owner there is no need to spend a lot of resources on an expensive testing tool. You can go with Google Content Experiments, which is a part of Google Analytics and free to use. There are two other cheap tools Optimizely and Website Optimizer, both capable of providing you with the desired results after conducting A/B split testing. You can also outsource the whole testing procedure to a reliable designer and tester, who will be able to ascertain what the best course of action is.



Leave a Reply

Your email address will not be published. Required fields are marked *


CommentLuv badge
This blog uses premium CommentLuv which allows you to put your keywords with your name if you have had 0 approved comments. Use your real name and then @ your keywords (maximum of 0)