Content Experiments in Google Analytics

A few days ago, the Google Analytics team announced an important change that will impact both marketers as well as analysts. Google is integrating the Google Website Optimizer tool into Google Analytics and will discontinue Website Optimizer as a stand alone product by August 2012. This testing capability - which will now only be available to Google Analytics users - can be found in the Content section under the header "Experiments".

Content Experiments Menu

What you can do with Content Experiments in Google Analytics

The main purpose of Content Experiments is to test how different versions of your pages work in getting your visitors to accomplish specific goals of your website, such as making a purchase, viewing content on a particular page or signing up for a newsletter. Content Experiments allow you to test up to five variations of a page to see which one of those pages results in the most conversions.

With Content Experiments, Google is taking a somewhat different approach from the traditional A/B or multivariate testing. Google refers to it as more A/B/N testing. Instead of testing just two versions of a page as in standard A/B testing or various combinations of page components as in multivariate testing, you can test up to five full versions of a single page, each delivered to visitors from a separate URL.

Users of the Google Website Optimizer tool already understand the importance of testing different versions of a single page as small changes to a page can result in major improvements in terms of achieving your goals. We believe the further integration of Content Experiments into Google Analytics will greatly help marketers and analysts perform much better testing of the effectiveness of web site pages.

Google is providing a wizard-style interface for setting up such experiments - see the next section of this blog article - and with the integration of Content Experiments into Google Analytics, more metrics and data will be available to us, helping us to make better decisions on what works and what doesn't.

This short video is a short introduction of the new Content Experiments functionality:

How to set up a Content Experiment

There are three areas in Content Experiments: the experiment setup wizard, the list of experiments and the individual reports for each experiment. Additional data related to the experiments will also be available in your Google Analytics profile.

When you open Content Experiments and you have not yet created any experiments, you will see this overview page for the experiment setup wizard. From this page, you can click 'START EXPERIMENTING' to create your first experiment.

Content Experiments Wizard

If you have already created other experiments, when you open Content Experiments, you see the experiments list. To create an experiment from this page, click Create experiment.

Whether you start from the overview page or from the experiment list, the setup wizard opens so you can complete four steps:

  1. Select the pages for your experiment

  2. In this step, you will need to select which original page and which variations of this original page you would like to include in your experiment. Once you have selected the pages, you will see a display of these pages to verify that you have selected the right pages. You can select up to five variations of the original page.

    Content Experiments Step 1


  3. Select the options for your experiment

  4. Next, choose what type of goal you want to test and define what percentage of your visitors you want to include in your experiment.

    Content Experiments Step 2


  5. Add the experiment code to your page

  6. Much of an experiment uses your Google Analytics tracking code, however you - or your IT team - will need to add the experiment code to the original page of your test. When you click Next, Content Experiments checks to make sure your original and variation pages are working properly.

    Content Experiments Step 3


  7. Review and launch your experiment

  8. Finally, review your configuration and start your experiment by pressing "Run Experiment Now".

    Content Experiments Step 4

Following up on your experiments

Once your experiment is running, you can follow up the execution of your experiment on the Experiments List page. On this page you can see:

  • Which experiments are running, which are stopped and which are still in Set Up mode
  • What date the experiment started and ended
  • How many visits were part of your experiment
  • Whether an experiment produced a winning page
  • Important notifications about your ongoing experiments


Content Experiments List

From this page, you can also create a new experiment or search your existing experiments.

When you click an experiment in the list, a detailed report for that experiment opens. On this report you will see:

  • The status of your experiment
  • How many days your experiment has been running
  • How the pages in your experiment are performing, both in terms of number of conversions and conversion rate, and how this compares to the original page your are testing
  • Whether a particular page your are testing is clearly outperforming the other pages
  • A graphical representation of the conversion rate of each of the pages in the experiment


Content Experiments Report

From this page, you can also stop an experiment, change the percentage of visitors that will be included in the experiment or disable a non-performing page.

Conclusions

The new Content Experiments clearly provides some benefits. There is only one code snippet you need to include on the page rather than multiple pages of code as was the case with Google Website Optimizer. With the Setup Wizard, creating a new experiment has become a very simple process. You can also now use advanced segments to segment your results too. In addition, Google is also working on some more functionality such as experiment suggestions, additional goal conversion options and the ability to receive e-mails about how your experiment is doing.”

Finally, it is important to know that Google will not offer a migration option for Google Website Optimizer users, so if you are running existing tests in Google Website Optimizer, you can run them up to August but then you will have to re-create these tests using the new Content Experiments functionality in Google Analytics. If you are planning to start any A/B testing soon, we recommend you use Content Experiments in Google Analytcs rather than Google Website Optimizer.

Image sources

Let's connect!

For advice, contact:

Dirk Dupont, Managing Director
Tel: +32 (0)475 69.48.26
dirk@questra-consulting.be

Pierre Coxstraat 13
3511 Hasselt, Belgium

© Questra Consulting BVBA, 2007-2016