By now you’ve heard that Google’s Website Optimizer (GWO) has been integrated with Google Analytics as Content Experiments and that on August 1, 2012, Website Optimizer will no longer be available as a standalone product. What can you do to protect your current website optimizer test data and prepare for Content Experiments?
An Experiment with Content & Analytics
With Content Experiments, Google aims to help users better optimize the goals already defined within their Google Analytics accounts. By putting all your optimization tools in one place makes sense, but will take time to get used to.
Many users have indicated that they do not yet have the option for Experiments within their Google Analytics account. Comments within the Google Analytics product forum indicate that Google is rolling out access between now and August. While it’s understandable that a roll out among millions of users takes time, it puts current Website Optimizer users in a precarious position, since they cannot use GWO to set up long term experiments.
Content Experiments Set Up
A Google Account that has been associated with a Google Website Optimizer account is automatically enabled for Content Experiments. Within Google Analytics, Content Experiments provides a setup wizard that walks users step by step through how to set up experiments, and launch new tests. However, for users with existing experiments set up in Website Optimizer, Google will not offer a migration option, meaning that users will have to re-create GWO tests on the new Content Experiments.
Furthermore, it’s currently unclear how users can enable Content Experiments for Google accounts that have an administrator role on Google Analytics accounts, but do not have a GWO account under the same email address can sync access.
Prepare For Testing
It’s good practice that before you set up any analytics testing, you first determine what it is you want to test for and why. Once the objectives have been clearly defined, think about how you plan to get there. Having a goal path outlined can help you set up your analytics goal within GA.
- Create an Analytics goal with a goal path that corresponds to the pages on your site and with a URL or event goal that corresponds to the action you want visitors to take on your site (visit a page or engage in an event).
- Define the paths you want to test. You can set up your Analytics goal to include all the steps, and then test the entire path in one experiment or you can test each segment of your path through successive experiments. The latter is important if a linear progression through your goal path is necessary.
- Choose the original test page in the path you want visitors to follow. If you're running successive experiments to test path segments, then the original (test) page will be different for each experiment.
- (Optional) Create variations of your original page. Users can include up to five variations of a page in an experiment, including images; text; image or text, or image & text; too much, too little, just right; and emphasizing one product feature over another.
- Add Analytics tracking code to each page.
- Choose what percentage of your site’s visitors to include in your experiment. Keep in mind that the lower the volume of traffic to your conversion page, and/or the more variation pages you include in your experiment, the longer it will take to get statistically significant results.
The Content Experiments set up wizard will walk you through each step, but having each step defined in advance can help you be more efficient.
Report & Analyze
Once experiments have been set up and executed, users can track progress using the experiment list and the reports. The experiment list includes general information about experiments and can help users track the status of each experiment (e.g., Setup, Running, Stopped, Ended), the total number of visits to pages in each experiment, and start and end dates.
Users can generate a report which includes a graph, a summary, and a data table, each of which presents different information to help users evaluate the experiment’s outcome.
Test, Tweak & Track
Overall, Content Experiments is designed to better help users understand which content performs best, and identifies a winner as soon as statistically significant data has been collected. Additionally, it helps simplify the process by letting users reuse Google Analytics tags, so they need only add one additional tag to the original page. Of course, setting up an experiment doesn’t mean it will work. It may require some tweaking to get it right and many problems may result from coding errors -- you will want to review Google’s Code Checking Error page for good measure.
However, once issues and errors have been resolved, Content Experiments promises to empower website experimentation within the Analytics interface, making it much easier to test, analyze and apply results.