Why I decided to spend 2 days of my life on A/B Testing?


16 February, 2018

History & Origin of Splitter

I'm Joseph Benguira, Founder & CTO of AppDrag Cloud CMS & Cloud Backend CMS & Backend in the cloud hosted by Amazon.

I started to work on Splitter after a friend and bussiness partner, Maxime Seligman, challenged me to produce very quickly an A/B testing tool that he will be able to present to a big public webmarketing event : 12th Annual MEGAComm Conference on Feb 15 2018. 

This was the perfect occasion to test real-time technologies like AWS IOT with MQTT and prototype all of that very quickly in AppDrag Cloud Backend. 
I consider this as an experiment and a side project and a complementary tool for AppDrag users but also for anyone wanting to do A/B testing for free. 

I intend to keep it free, active and maintained as a side project for AppDrag, so feel free to use it :)


What is Splitter A/B testing?
A/B testing consists of comparing two versions of a page in order to check which one is the most efficient. These variations, called A and B, are presented in random ways to users. 

Part of them will be directed to the first version while the other will be assigned to the second. A statistical analysis then makes it possible to test the effectiveness of version A and B on different indicators such as the conversion rate. In other words, you can check which version triggers the most clicks, purchases, subscriptions,... 

The results then determine the best marketing strategy to adopt.


Who should use A/B testing?
All websites can benefit from A/B testing because all websites have at least one measurable goal. Whether you are an online store, a news site or a lead generation website, you are looking to improve your conversion rate, regardless of the nature of this conversion. 

In the case of the media, we speak more readily of editorial A/B testing. Indeed, in these circles close to the press, it is mainly a question of an editorial function, that is to say that one tests the success of a category of contents (for example by ensuring that they are perfect for the target audience). 

Regarding e-commerce, A/B testing targets the commercial efficiency of an online shopping site or application. A/B testing verifies which version is the most efficient thanks to the sales made. In particular, we study the home page, the elements of the product sheet, but also the control elements (buttons, call-to-action, etc.).

How to use A/B testing to improve your conversion rate?
The process begins by identifying the page that you want to improve. Once you have selected your target page, you have to create a variation that can be compared against the original.You can now copy/paste the 2 URLs on the fields URL A and B. 

It’s optional but Splitter also give you the ability to track conversions, to do that, you need to place a 1 line script on both your pages as explained on step 2 when you create a new test. 

The last step is to click on “Create my A/B test”. You should now see your live dashboard. You can copy the Split URL in the top right corner and share on facebook, by email, in an ad campaign, ... 

You can keep your browser tab open to monitor in live visits & conversions. Tests must be run long enough (sufficient sample) to achieve statistically significant results.
Once one page has proven to have a statistically higher conversion rate, implement it as the permanent page for that URL.


Latest articles