How to Create a Multipage Testing Experiment on GemX

GemX allows you to create two types of experiments:

• Template Testing: focuses on optimizing individual pages.

• Multipage Testing: focuses on optimizing the entire sales funnel.

This article will walk you through how to create a Multipage Testing Experiment step by step.

Please make sure that the GemX: CRO & A/B Testing app is already installed on your Shopify store.

If not, follow this guide first: How to Install GemX on Your Shopify Store.

What is the Multipage Testing Experiment?

The Multipage Testing Experiment is an A/B testing feature designed to help you make smarter design decisions, not just on one page, but across your entire customer journey.

Instead of testing a single landing page or product page, this feature lets you test two complete sets of templates at once:

• Group A (Control Group): Your current live designs.

• Group B (Variant Group): A new version you want to test.

When a visitor lands on your store, they’re automatically assigned to either Group A or Group B — and will stay within that group for the entire session. 

At the end of the experiment, you'll receive both aggregated performance data by group and detailed comparisons between each matching template pair, so you can confidently determine which version performs better, and apply it to your live store.

Who Can Use This Feature?

1. GemX Plan Requirements

Please note that this feature is only available to:

• Funnel Optimize Plan users

• Free Trial users

2. Shopify Platform Limitations

While GemX enables powerful testing capabilities, it's important to understand Shopify’s platform limitations that may affect how you use this feature.

• Template Limitations by Shopify Plan

Shopify limits the number of templates that can be created in each store, depending on the user's Shopify plan.

Note: The 1,000-template limit is per theme, not per page type (like product or landing pages).

• Store Performance

Having too many active templates in your theme can slow down your site, especially for mobile users. We recommend disabling or removing any unused templates.

• Shopify API Rate Limits

Shopify imposes rate limits on API usage, which affects how quickly templates can be synced when using this feature.

How to Set Up a Multipage Testing Experiment in GemX

Step 1: Create a multipage testing experiment

To create a new Multipage Testing experiment in GemX, go to the GemX dashboard, click Create new experiment, then select Create multipage experiment.

Step 2: Select templates for the Control group

The Control group is the current live version of the pages that your users are already interacting with. When you start a test, GemX will use the URL of this Control version.

To choose it, click Select templates for Control group.

Clicking this button will open a modal that allows you to select the templates you want to test. In the modal, you can:

• Filter by page type: Narrow down the list of pages based on their type (e.g., product page, collection page, landing page, homepage, blog post, etc.).

• Filter by template source: Identify pages based on whether they were created using the standard Shopify builder or the page builder.

After selecting the templates, click the Select template button.

Notes:

• For GemPages, only published templates are shown. Drafts are excluded.

• For Shopify, templates from the current theme are shown.

Templates will not be shown for selection if:

• They’re already used in another experiment.

• They’ve been selected as a Variant in the current experiment.

You can reuse the same template as the Control for multiple test pairs in a single experiment — as long as it hasn't already been applied to all items (e.g. products, collections, or pages).

Step 3: Select templates for Variant group

If the Control is a Shopify theme template, you’ll have one option for creating a Variant: Select template for Variant.

In the case where the Control is a template created by GemPages Builder, there will be 2 options to choose from:

• Select an existing template: Open a modal with a list of templates available to select as a Variant.

• Create Variant based on Control: This action creates a duplicate template of the Control Template. This template's structure is duplicated from the Control Template, and used as the Variant Template.

Note: The option “Create Variant based on Control” is only available in GemPages version 7.

When you click the Create Variant based on Control button, the app will open the Variant Template in the GemPages Editor in a new tab. Here, you can make the desired changes, and after you have finished editing, click the Save & Publish button.

Note:

• When selecting a Variant template, only templates of the same page type as the Control template will be shown.

• For GemPages, only templates with Published status are available for selection; templates in Draft status will not appear.

• Templates duplicated from the Control Template are excluded from the GemPages Pages (templates) listing and will not be displayed.

• To delete Template pair, hovering over the footer of the pair → Clicking on the trash icon.

Step 4: View or Change Pages/Products/Collections/Articles to test

Once you’ve selected your Control templates, you can view and adjust which Pages, Products, Collections, or Articles are being tested within each Template pair.

To manage these test items:

• To view test items: Click the Pages/Products/Collections/Articles to test button. This opens a preview of the specific items (e.g., which product or collection) that will be tested with the selected template.

• To change test items: Click the Change (products/pages/collections/articles) to test button. This opens a modal that allows you to select or deselect items for testing. Once you've made your selection, click Choose [items] to apply the change.

Note:

• The Homepage does not have the "Change (items) to test" feature.

• You need to pause the experiment to change the test items.

• You need to select at least one item to test.

Step 5: Advanced Settings

Once you’ve selected both Control and Variant templates, continue by configuring the Advanced Settings to define how your test will run. These settings help tailor your experiment to specific goals, audiences, and traffic conditions.

You can adjust the following:

Set the primary metric

Choose the primary metric that’s most important to you for this test. All metrics will still be tracked and available in GemX analytics. GemX will rely on the primary metric to calculate and suggest which version is the Winner in the experiment (after enough data is available for calculation, usually after 2 weeks of running the campaign or when the visitor count is around 2000).

Set device type

Choose whether to run the test on All devices (default), or target specific device types such as Desktop, Tablet, or Mobile.

Set visitor type

Decide who will see the test — All Visitors (default), New Visitors only (first-time users), or Returning Visitors (repeat users familiar with your site).

Set traffic source

Focus the test on traffic from specific sources, including:

• All Sources (default)

• Referral (visits from external websites)

• Organic Social (unpaid social posts)

• Organic Search (unpaid search engine results)

• Email (traffic from email campaigns)

• Paid Social (paid social ads)

• Paid Search (search engine ads)

• SMS (links from SMS campaigns)

Set traffic for each group

Allocate the percentage of visitors shown to each group (Control vs. Variant). The common starting point is a 50/50 split.

Note: Currently, above mentioned sources are fixed and the user cannot add further. If you have specific requirements or feature suggestions, please don’t hesitate to submit a feature request. 

Step 6: Start experiment

You'll activate your experiment, making it live and operational based on the settings you've configured.

Once you click the start button, your experiment will go live. The live page will start functioning according to the settings you've established, such as the selected templates, traffic allocation, targeted devices, visitor types, and traffic sources.

Once your experiment is live, you can continue adjusting the traffic split between versions at any time. However, if you want to change other settings like conversion goals, audience targeting (e.g., by device), or page content, you’ll need to pause the experiment first.

Notes:

• When an experiment is paused and later resumed, it creates a new "version" in the Analytics section.

• You can select from the dropdown to view metrics from each version.

Once your Multipage Testing experiment is activated and has collected enough traffic, you can start reviewing performance metrics to evaluate which version is performing best. 

For detailed instructions on how to view experiment analytics in GemX, please refer to this article: How to View and Read Experiment Analytics.

Realted Topics: 

Multipage Testing