Tablet visitors sent to the desktop site, not mobile: +71.81% revenue
Commentary byLiftArchiveOur analysis, not theirsSending Android tablet visitors to the new HTML5 desktop site instead of the phone-oriented mobile site raised revenue by 71.81% for Shirtinator.
The custom apparel and gifts company ran this split URL test itself, and it was published as a VWO case study. Completed orders rose 32%.
- Revenue: +71.81% for tablet visitors on the desktop site.
- Orders: +32%.
- Revenue grew faster than orders, which suggests bigger orders as well as more of them.
- Only Android tablets were tested; sample size, duration and confidence were not shared.
Control
Android tablet visitors were redirected to the mobile website, which was built for small phone screens.
Variant
Android tablet visitors were sent to the new HTML5 desktop website that Shirtinator was developing. Traffic was split 50/50 between the two sites.
Why it likely won
Our reading of the result, not the practitioner's claim.
- A phone layout stretched to a tablet screen wastes space and loses a clear path for the eye.
- Tablet screens are big enough for a desktop layout to work with few changes.
- The desktop site may have shown more products and detail per screen, helping larger orders.
Caveats: The test covered Android tablets only, and sample size, duration and significance were not shared. Shirtinator later moved to a responsive design, so this setup is mainly a lesson for sites that still serve separate mobile pages.
Who ran it
Shirtinator ran this test for Shirtinator and published the results (VWO case study). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Shirtinator →
How to run this test on your store
- Hypothesis
- If we serve tablet visitors the desktop layout instead of the phone layout, then more will buy because the page fits their screen.
- Primary metric
- Revenue per tablet visitor
- Guardrails
- Orders per tablet visitor
- Tablet bounce rate
- Page load time on tablets
- Who should test it
- Stores that still run a separate mobile site and get a real share of tablet traffic. If your site is already responsive with a tablet breakpoint, this test doesn't apply.
Questions
Should tablet users see the mobile or desktop site?
In this test the desktop site won clearly for Android tablet users.
Phone layouts often look stretched and sparse on a tablet.
How do I test two versions of a site against each other?
Use a split URL test, which sends visitors to two different URLs and compares results. Shirtinator split tablet traffic 50/50.
Is a separate mobile site still a good idea?
Most stores now use a responsive design that adapts to each screen.
Shirtinator moved to responsive design after this test.
Similar tests
@oliverkenyon@oliverkenyon@oliverkenyon
@DaveDiederen@DaveDiederen
Blend Commerce