Last month Google introduced Analytics users to Content Experiments, a revamped version of the old Website Optimizer tool. Content experiments is a tool that allows site owners to do simple A/B testing against a common goal.
This week, Google announced they've already added several improvements to content experiments.
First up, if you hadn't seen them before, content experiments are now available for all Google Analytics users. You'll find everything you need in the "Content" area, under the subheading of "Experiments."
As simple as this sounds, experiments you set up now support relative URLs. Previously, the URL needed to be absolute, using a drop-down menu for selecting the protocol (http: or https:). While this seems simplistic, if you're setting up more than one variation, it makes it much quicker to simply use variation1.html rather than to write out www.example.com/variation1.html.
On the subject of saving time, you're now able to copy a previous experiment. This is great for two reasons:
- If you're setting up a new - but similar - experiment, you can duplicate an experiment and simply change a few things to set up a second one.
- Moreover, content experiments may only run for 90 days. After 90 days, they automatically stop. Copying an experiment will allow you to essentially start an experiment over, but not overwrite previous test results and reports.
Lastly, the Analytics team has improved the reports produced by the content experiments. Previously, simple metrics like site usage, goal set, and ecommerce were not included. Now you can plot them on the graph as easily as any other report.
Content experiments are still in their infancy. For what they are, they are a great start at A/B testing. However, your test must be simple and focused on a single goal.
No multivariate testing yet. However, give Google credit for bringing A/B testing to the masses for free and already improving it, less than 60 days later.