Skip to main content


Optimizely Knowledge Base

Multi-page experiments in Optimizely Classic

  • Set up multi-page experiments in Optimizely Classic
  • Add pages and variations to your multi-page experiments
  • Set up URL targeting for your multi-page experiments
  • Add audiences to your multi-page experiments

What are multi-page experiments? 

Multi-page experiments allow you to link together related variations of different pages, so you can learn how changes to your site might affect your entire funnel. For example, visitors who saw the original version of Page 1 will also see the original version of Page 2, visitors who saw Variation 1 of Page 1 will also see Variation 1 of Page 2, and so on.

Multi-page tests are supported by both Optimizely Classic and Optimizely Web. However, each product handles this kind of test differently. This article provides a brief overview of multi-page experiments in Optimizely Classic.

Multi-page experiments in Optimizely Classic

In Optimizely Classic, you can create multi-page experiments by changing the experiment type for an experiment you’ve already built. You will then add pages to the experiment, add the variations you want to test, and set URL targeting and audiences.

Changing the experiment type

In Optimizely Classic, the process of setting up a multi-page experiment begins by creating a new experiment. Once you've done so, click the Editor button to open the Editor.

In the Editor, find the Options drop-down menu. Click it, and then select Experiment Type.

 By default, this pre-existing experiment will have an experiment type of A/B Test:

To change the experiment type, select Multi-page Test and click Apply.

Add pages

The Add Page option lets you add pages to your experiment and create variations that change multiple elements along a funnel. You can then run an experiment that makes a different set of changes to any number of pages, including your:

  • Landing page

  • Product pages

  • Cart page

  • Checkout page

A visitor who sees a particular variation on any of these pages will continue to see that variation through the rest of the pages that are included in the multi-page experiment. They won’t have to visit your pages in any particular order to be bucketed into a multi-page experiment.

Add a page to your experiment by clicking the Add Page button within the Editor.

In the Add Page dialogue box that appears, enter a name and URL for the page you want to add to your experiment.

Repeat this process for each page you want to add.

Add variations

To add a variation to your experiment, click the Add Variation button from within the Editor.

Name your new variation by selecting Rename Variation from the drop-down menu and typing the new variation name into the dialogue box that appears. 

To add more pages to this variation, just follow the steps described in the previous section until your new variation includes all the pages it requires.

In a multi-page experiment, any pages you add to a variation will be automatically assigned the same variation name. This is because multi-page experiments are intended to test a complete experience across multiple pages.

Once you have finished setting up a multi-page experiment, you can add goals, set targeting, and adjust your traffic allocation, just as you would in a regular A/B experiment.

Set up URL targeting and audiences

Finally, you can set the URL targeting conditions and audiences for each page in your multi-page experiment. From the Options drop-down menu, select URL Targeting.

In the dialogue box that appears, enter the URLs you'd like to target. If you haven't done so already, read up on URL targeting before continuing.

Each page has its own conditions, which means you can apply changes not just to individual pages, but multiple page types. However, unlike URL Targeting, all pages of the multi-page experiment must be visible to the same audiences.

In order for a visitor to be bucketed into a multi-page experiment, they must match both the URL targeting conditions and all the audience conditions. In this sense, setting audiences for a multi-page experiment follows the same process as an A/B experiment.