One-Column or Two-Column Sales Page: A/B Test Example

Heatmap clicks clustering on side links, session recordings of visitors moving between the main copy and the sidebar, or visitors who leave before reaching the key section signal that the side column pulls attention away from the main sales narrative. If the side column really is a distraction, the key section is seen more often in the single-column layout and purchase conversion rises. If visitors rely on the sidebar's information or the offer is unclear, look at the content first, because a layout change will not fix that.

Purchase conversion rate (CR)

Compare purchases among eligible assigned visitors on the same sales page.

Also track

• Key section view rateMeasure whether a predefined section enters the viewport using the same rule in both layouts.
• Active time on pageExclude passive background-tab time and read active time alongside conversion and section visibility.
• Supporting-content action completionCompare predefined actions such as downloading a resource or opening a reference.

• LayoutCompare the current two-column layout with one predefined single-column layout.
• Supporting contentChoose where the content remains accessible and record any unavoidable visibility change.
• Page lengthMeasure access to key sections rather than relying only on raw scroll percentage.
• Primary actionKeep the copy, offer and CTA behavior unchanged.
• Eligible devicesExclude devices that already render the same single-column layout.

1. 1Decide on purchase conversion. Key section view rate and active time on page show access to content, not proof that it was read, so a rise in them without more purchases is not a reason to ship. If supporting-content action completion falls, consider that visitors may no longer reach information they need.
2. 2Other sales pages need their own validation; this result speaks for the page type you tested. When the side column goes, make sure you do not quietly remove a link or menu that keyboard or screen reader users need to get around, and check accessible navigation before shipping the variant.
3. 3If the eligibility rule does not work the same way in both arms, arm sizes can drift, so first check that assigned visitor counts match the planned split (a sample ratio mismatch check). If the split holds, enter each arm's assigned visitors and purchases in the A/B test significance calculator to see whether the conversion gap goes beyond chance. If the gap is small and not significant, do not call a winner; report that this sample showed no detectable effect of the single column.