Sucess events

​

Success events definitions

​

What is a success event?

A success event in A/B testing is a predefined action or metric that determines the effectiveness of an experiment. It represents the primary outcome you want to measure to assess whether your variations outperform the control. Success events should be closely aligned with your business goals and objectives.

Example: For an e-commerce website, a success event could be the completion of a purchase. Here, you would track the conversion rate, which is the percentage of visitors who complete a purchase. Other examples of success events include:

• Clicking on a high-impact call-to-action (CTA) button – such as “Add to Cart” or “Subscribe Now”.
• Registering for a newsletter signup – showcasing visitor interest and potential future engagement.
• Downloading a valuable resource or document – such as an e-book, white paper, or case study.
• Completing an inquiry or lead generation form – such as requesting a demo or filling out a contact form.
• Engaging with key interactive elements – such as a live chat service or survey completion.

Each of these actions provides valuable insights into user behavior and helps you determine which wording or variation drives better results.

​

How many success events can I define with Gleef?

Currently, you can specify one among two success event for each experiment you conduct.

1. Click on an item (CTA) - Not recommended

   By defining the success event as a click on the targeted CTA, we track the first time a visitor clicks the button, and then attribute that event to the specific variation being shown (if applicable).

   This kind of success event is used to calculate the CtR (Click through Rate) of your call to action. It is not recommended as it can be misleading if the visitor didn’t actually complete the desired action, or directly quit the page.
   This option has the following limitations:
   • It requires the element you’re experimenting to be clickable
   • If multiple elements are clickable, only one of them - the one you’re experimenting will be the success event, unless there are multiple likewise elements.
   • If you want to define the success event as a click on another item - not the one you’re currently experimenting, you can define a multi-select experiment, firstly selecting the element you want to define as the success event, and setting its variation equals to the baseline, and then selecting the other element you really want to experiment.

2. Specific URL Page - Highly recommended

   You can choose any URL that represents a meaningful outcome for your experiment. URLs that signal a successful event can be ‘thank you’ or ‘purchase completed’ pages.

   We highly recommend to use this kind of success event as long as it reflects a real impact on your business, and not a click that can be misleading if the visitor didn’t actually complete the desired action, or directly quit the page.