Split URL A/B Testing

Split URL A/B Testing allows you to compare full web pages to determine which version drives better results. It's ideal for testing major changes like full redesigns, new layouts, or alternate user journeys — going far beyond simple changes like headlines or button colors.

Split URL A/B Testing is a conversion rate optimization (CRO) technique that allows you to compare two or more versions of a webpage by hosting them on separate URLs (e.g., example.com/page-a vs. example.com/page-b). Instead of displaying variations on a single page, users are redirected to entirely different URLs—each representing a distinct version of the page.

Website traffic is randomly split between the different versions, and user interactions (such as clicks, signups, or purchases) are tracked separately. This data is then analyzed to determine which version delivers the highest conversion rate. Split URL tests are especially useful for larger experiments that involve substantial design, layout, or backend changes.

As a rule of thumb, use Split URL testing when you're testing fundamentally different versions of a page, especially when changes impact layout, structure, or functionality. Traditional on-page A/B testing is better suited for smaller, UI-level tweaks (like button colors or headlines).

Split URL A/B Testing is ideal for larger experiments, such as:

By hosting each variation on its own URL, you can observe how each version affects visitor behavior and conversion rates. Once you identify a winning version via Split URL testing, you can continue optimizing it with traditional A/B tests for more granular improvements (like tweaking CTAs, imagery, or copy).

First, log in to your OptiMonk account at https://app.optimonk.com/login/en

If you've previously created Split URL A/B tests, you'll see a lab icon in the left toolbar where your tests are listed.

If you have not created a Split URL A/B test before, you'll need to follow these steps:

Select Campaigns on the left, then select "New Campaign" in the top right corner:

If you went to the A/B Test Center, here you'll see all your previously created A/B Tests, Multi-Campaign A/B Tests, and Split URL A/B Tests — if you have any — listed by their names and domains. To launch a new A/B Test, click on New A/B Test in the top right corner and select Split URL A/B Testing.

Select the domain you'd like to run the A/B test:

Name your Split URL A/B Test to help you identify it later:

Add your URLs to attract different visitors.

❗With OptiMonk, you cannot build your landing page variants; you can only test your built ones.

After choosing the pages you want to compare, figure out how much traffic to give each page. You can show both pages to the same number of visitors, or you might choose to show variant A to most visitors, and only a small portion will see variant B. This way, you can run tests with less risk.

❗The visitor's experience is locked after the first visit—they will always see the same version.

✍️ If you want to run your Split URL A/B Test across multiple pages, you can use the wildcard symbol (*) at the end of a URL pattern. This lets you target all pages within a section of your site.

For example, to test different product page designs on all your product pages, use a URL pattern like: shopify.com/products/*. If you have a second variant hosted on a different theme or layout, you might use: shopify.com/products-theme-b/*. This setup will run the test across all product pages in both variants.

Add as many variants as you'd like to test:

You can even exclude URLs that you would not like to target your visitors to:

For example, when running tests across multiple pages using the wildcard symbol (*), you might still want to exclude certain pages from the test. Using the example above, if you run the test on all product pages with shopify.com/products/*, but don't want the test to run on discounted products, you can exclude those pages by adding https://www.shopify.com/products/discounted to the exclusion list.

If you're done with setting your Split URL A/B Testing up, click on Save:

To launch your Split URL A/B Test, click Start. Once you launch a Split URL A/B Test, if you wish to end it, you'll have to do it manually.

You can still edit or delete your Split URL A/B Test after launching:

You can end your Split URL A/B Test any time by clicking on End.

You'll be able to choose what counts as a conversion for your test—whether it's button clicks, form submissions, orders, signups, or something else. By default, conversions are tracked based on order numbers, but you can set up any custom conversion goals you want to measure success by.

Use our Conversion Goals to track standard events like "Add to Cart" actions or visits to specific pages. For more specific tracking—like video plays or engagement with interactive elements such as quizzes or calculators—you can set up Custom Conversion Events tailored to your needs.

You can view the results on the test page during and after the test, too. You will see these metrics:

⚠️ Keep in mind that the stats here only reflect the results achieved within the timeframe of the Split URL A/B Test. Further data can be measured in Google Analytics.

Anti-flicker is a small script that prevents visitors from briefly seeing the original version of a page before being redirected to a test variation. Without it, users may notice a "flash" of the default page before the new variant loads—this can feel jarring and may affect how they interact with the site.

Using anti-flicker ensures a smoother, more seamless user experience and helps maintain the accuracy of your Split URL A/B test results.

If you need any further help, please let us know at [email protected] or via our in-app live chat, and we would be happy to assist you!