- Run experiments that compare different combinations of changes
- Set up a multivariate experiment in Optimizely Classic
- Add variations and sections to the experiment
- Allocate traffic among variations and sections
For an overview of the pros and cons of multivariate experiments, see our article on experiment types.
Multivariate experiments allow you to create a few simple changes on your page and test how the different combinations of changes affect the page's performance.
In this article, we'll walk through the five steps in creating a multivariate experiment:
Create a new experiment
Set experiment type
1. Create a new experiment
To create a multivariate experiment, start the same way you would to create a regular experiment:
From the Home page, click New Experiment.
Type a name for the experiment and the URL of the website you’d like to test.
Click Create Experiment.
Your new experiment will load in the Editor.
2. Set experiment type
In the Editor, click Options > Experiment Type.
Select Multivariate Test.
Next, you'll add variations and sections to your experiment. Each variation/section combination will be tested in your multivariate experiment.
3. Add variations
The Editor automatically loads a page you can use to create Variation #1 for Section #1 (more about sections in the next step).
Click + Add Variation to create more variations. In this example, the variations will test the language on the button (the Original, 'Search It!', and 'Google It').
You can change the name of a variation by clicking the dropdown in the Variation menu.
To see how a variation's changes will display, click the dropdown arrow next to the variation name and select Preview.
4. Add sections
When you selected Multivariate Test, the + Add Section option appeared above the Variation tabs.
Sections are separate A/B experiments that all run at the same time (for example, a button, image, headline, and line of text). Your multivariate test will have more than one section, and each section will have multiple variations--that's how the multivariate test helps you determine which combination of variations performs the best out of all of the possible combinations.
In this example, for Section #2, you'll choose the two different images on your page. Click + Add Section and make your changes.
Like variations, you can change the names of sections by clicking on their names.
Sections are applied from left to right. In our example, the Button section will run first, and the Image section will run after it. This is important if you're running a multivariate test that affects the same element in two different sections.
5. Allocate traffic
In the Editor, navigate to Options > Traffic Allocation:
In this example, 33.3% of your visitors will see each of the variations from the two sections. The visitors do not have to see the same variation from each section--they will be randomly assigned to one variation from each section. You can change the traffic allocation by typing in new percentages, but don't forget to make sure they add up to 100%. When you're finished, click Apply.