Maxim2016-02-26 22:31:54
Maxim, 2016-02-26 22:31:54

A/B testing - how best to collect data?

There is a website and there is a certain page with a button. And so I decided to conduct a / b testing on it - I show this button to half of the users in blue, and to the other half - in green. How to implement all this technically is not a problem. And how to properly collect data is my question. I'm using Google Analytics - when do I need to send an event to GA? I see two options:
1. The button has its own event, like "button click", which is sent to GA, and for each of the a / b testing options there is a separate event that is immediately sent when the page loads.
2. Each variant of a/b testing changes the event of the button itself. In this case, as I understand it, I will not be able to take into account how many people generally viewed the page with one or another type of button, but I will only deal with the result of clicks.
And a more general question (I'm new to GA) - can I see in GA how users behaved with this or that event? Well, that is, if on page load I send an event to GA, then can I find out what further events this user caused?
Or do I not understand how to properly collect data from A/B testing?

Answer the question

In order to leave comments, you need to log in

2 answer(s)
kn1ght_t, 2016-02-26

there are tools for that

Dmitry Pechurkin, 2016-02-26

you can create two counters and insert one counter when the button is blue and the other when the button is blue, a fairly simple way but for that it is clearly demarcated. I don’t know how in Google, but Yandex allows you to do this without any problems.

Didn't find what you were looking for?

Ask your question

Ask a Question

731 491 924 answers to any question