HOW TO USE GOOGLE ANALYTICS EXPERIMENTS

Synopsis

You are planning to run A/B (or Multivariate) testing on your website to determine the most effective new page design, layout or copy. A common experiment is to test variations of the top landing page of your website to improve bounce rate or conversion rate.

Define Objective of Content Experiments in Google Analytics

This objective should correspond to the experiment page’s main goal. If you are ecommerce, this would be transaction or revenue. If the goal is sign-ups, this would be conversion rate. If you are concerned with engagement, this could be bounces or session duration.

Setup Content Experiment

Once you have setup the original URL and variations, Content Experiments will give you your experiment code which is a piece of JavaScript that you embed onto the top of your original page after the opening tag. Remember, put this code only on your original page, not the variations.

Analyse Content Experiment Results

The winner is the page that best achieves the objective and is only chosen upon meeting the confidence threshold. If no page meets the confidence threshold by the end of the experiment, no winner is declared and you will have to analyse the results to determine which was closest to winning or re-run the experiment with adjusted variations.

WANT TO FIND OUT MORE?

Read the original Article from Google

http://bit.ly/1qgcxeY
Subscribe to our newsletter