Create a Content Experiment

Step by step walkthrough for creating a content experiment in Elevate A/B Testing, from naming it through building variations in the editor.

---

Before you start: the theme extension enabled, and a clear idea of which element you want to change and why.

Open Experiments, select New Experiment, then Content Experiment.

Step 1: General Info

Enter an Experiment name. Description, Hypothesis, and Assets are optional.

Under What do you want to measure?, choose your goal. Revenue per visitor is recommended.

Match the goal to the size of the change. A button restyle is better judged on add-to-cart rate than on revenue, because the effect is narrow and revenue is noisy. A rewritten value proposition is a revenue question.

Select Browse all metrics for the full list, or Add Custom Metric to track something Elevate A/B Testing does not measure by default, such as a click on the element you are changing. See Custom Events.

Step 2: Set Variations

This opens the content editor, where you build each variation by editing the page directly.

Your original page is saved automatically as the control. You never have to build it.

Use the variation selector at the top left to switch between variations. Each one holds its own independent set of changes.

You can create up to five variations including your control.

For everything the editor can do, see Using the Content Editor.

Working across several pages

A content experiment does not have to be confined to one page. Use the page search at the top of the editor to move to another page and make changes there within the same experiment.

Search by page name, or paste a URL directly. Pages are grouped by type, so landing pages and product pages are listed separately.

Use this when the change belongs everywhere it appears, such as a trust badge in a footer or a message that repeats across your landing pages. Do not use it to bundle unrelated changes into one experiment, because the result will not tell you which one worked.

Remaining steps

After the editor, the experiment follows the same steps as every other type: Traffic Allocation, Audiences, and a final review before you create the preview.

• Traffic Allocation. Set the split with the slider. Even is the default. Isolate traffic for this experiment is optional and off by default. See Traffic Isolation.

• Audiences. Optionally narrow who enters, by device type, visitor type, UTM parameters, traffic source, or country. See Audience Targeting.

Then create the preview experiment, QA it, and launch. See Preview and QA a Content Experiment.

Next steps

• Using the Content Editor

• Preview and QA a Content Experiment