A/B test your pages

Last updated: October 30, 2018

Applies to:

Marketing Hub
marketing-pro-enterprise
Professional, Enterprise
HubSpot Marketing Hub Professional and Enterprise customers have access to A/B testing with HubSpot's landing pages and website pages. Learn how to test the effectiveness of different elements of your pages with A/B testing, and learn more about conducting A/B testing.
Please note: when a page is an A/B test, it will have one, single URL. When a visitor views your page, HubSpot will randomly show them either variation A or variation B, which will allow you to later decide your winning variation.

Create a landing page or website page

Create your first variation (the control), using the normal steps for building a landing page or website page.

Create an A/B variant

In the landing pages or website pages dashboard, hover over the page you'd like to create an A/B variant for. Click the More dropdown menu and select Create A/B Variant.

Choose a name for the variation

In the dialog box that appears, enter a name for the variation, then click OK.

Edit the variation to create a proper A/B test

Consider the following variables for testing:

  • Offers - experiment with the medium of the offer. You might test an ebook versus a whitepaper or video.  
  • Copy - experiment with the formatting and style of the content. You could test plain paragraphs versus bullet points or a longer block of text vs a shorter block of text.
  • Image - try out different images to see how the conversion rate is influenced.
  • Form fields - experiment with the length of the form. Should your form only request an email address or should it ask for more information?

You can edit the other variation of the page by clicking the A or buttons at the upper left of the page editor.

View the A/B test results

After you publish the variation and let the A/B test run, click the Variations tab in your page performance details to view your A/B test results. Here, you can:
  • Use the Date range and Frequency dropdown menus to analyze data for a specific time range. 
  • Click the column headers to re-order the results by a specific metric. 
  • Choose a variation as the winner of the test and stop displaying the other variation by selecting Choose as winner.

  • Rerun an A/B test on a losing variant. Hover over the page name and click Rerun this variation. It is recommended to rerun a variation if some time has passed since the last A/B test. If you rerun a variation, the variation will be published immediately. 

Previous article:

Next article: