ConvertedU Marketing Library Webinar Blog

Need help? You've come to the right place

What's New Getting Started Drag and Drop Standard Builder Integrations Center

Creating a New Split Test

NOTE: This article covers creating a new split test in our redesigned split testing tool, available to Pro-level subscriptions and above. 

In this article

Plan your split test

    Prepare your pages

Create a new split test

    Choose a control and variations

    Distribute traffic

Start your test


Running a split test can be a great way to optimize your pages for conversion. If you’re new to split tests, you may want to read our introductory blog post here: Split Testing 101 (With Bacon), or check out some case studies.

Below, learn how to create a new split test with our redesigned tool. 

Plan your split test

We recommend beginning each split test you run by deciding what you want to learn. You might compare…

  • different page layouts (or templates)
  • headline copy
  • your button color

… among other variables to optimize your pages. 

TIP: For best results, we recommend limiting each test to one or two variables—if you test too many changes at once, it’ll be difficult later on to determine which exact change resulted in increased conversions. 

Prepare your pages

When you know what you want to test, it’s time to set up your pages. If you already have at least two pages you want to compare, you can jump ahead to steps for creating your test.

Otherwise, here’s what we recommend:

  • Choose a page (or create a new one) that will act as your control—that is, the page that will serve as the baseline for your test’s analytics—and make sure it’s published.
  • If you plan on comparing different templates, you’ll need to create and publish your variation page before setting up your test.
  • If you plan on comparing smaller changes between pages using the same template, you’ll be able to create you variation(s) as part of the test setup process.
NOTE: Our tool doesn’t limit what you test or compare—you can freely create a split test that compares entirely different pages if you like. 
  • To keep track of your test’s pages, you may want to name them so you can identify each variation (and perhaps add a “split test” label). 

Create a new split test

Once you’ve created your test’s control (and variations), it’s time to make the actual test. Here’s how:

  1. Navigate to the Leadpages tab in your account, then click Split Tests
  2. Click Create New Split Test
  3. Name your test by hovering over its placeholder title in the top bar (as shown below).

Choose a control and variations

In Step 1, browse your list of pages and click Use This Page to set your test’s control page—against which any variations will be compared.

NOTE: Each test can only have one control page. To change it, hover over the blue checkmark button, which will allow you to remove it from your test and choose a new one.

Click Next ➔ (in the left hand sidebar) to start selecting your variations (Step 2). 

On the variations page, click Add a Variation, then choose whether you’d like to copy your control (to test small changes), or use a different page entirely.

If you copy your control, you’ll need to edit and publish the new variation to use it in your test. Click the pencil icon next to the variation title to edit it in a new tab. 

Distribute traffic

Click the Next ➔ button in the sidebar once you’ve chosen your variations to decide how traffic should be distributed to them.

 

When people visit your split test’s URL, we’ll automatically direct them to a variation based on the traffic percentages you choose here. 

Pro tip: If your control page is already solid and working well for your needs, you might consider sending more traffic to it than your variation(s). That way, you can run a test without a high risk to your conversion rate.

NOTE: Visitors who return to your test won’t randomly see another variation—we store which version they landed on as a browser cookie, so each individual will only see one page in your test under normal circumstances. 

Keep in mind, you may not see your split test’s individual page views match up with your traffic distribution right away. It’s helpful to think of these settings as a weighted coin or die—you’re choosing which page(s) visitors are most likely to land on. 

WARNING: Once your split test is running, you won’t be able to adjust its traffic distribution.

Start your split test

Click Next ➔ to move to the fourth step, where you can finalize your split test. Review your test to make sure everything looks good to go, then click Start Test (or save it for later). 

To learn about actually running your test, check out Running a Split Test (New Version).

 

(Updated February 14, 2017)


Please let us know if you have additional questions.

You can always file a support ticket if you have additional questions by clicking the Support button in your dashboard shown here or clicking here to file a support ticket.

Was this article helpful?
Powered by Zendesk