Ecommerce Conversion Testing
See the measured results

Split testing for online stores

The page decides the sale. Test it properly.

We rebuild the page nearest the purchase and run it head to head against the one you use now, on your own visitors. Judged on revenue per visitor, not opinions.

See the measured results

Measured, not claimed

400+Ecommerce brands worked with
20% to 70%Typical lift in revenue per visitor
10% or freeThe threshold you pay on

Results from live split tests

LiftCategoryHow it was tested
+52%EyewearOffer page against the home page, Google Ads traffic
+45%Sweat careOffer page against their product page, ran two weeks
+33%Natural hair careOffer page against a collection page, Meta traffic
+32%Automotive careOffer page against their product page, Meta traffic
+25%Wellness gummiesOffer page against their product page, full metric tracking
+24%Oral careOur product page against theirs, Google Ads traffic
+21.4%Men's skincare99.5% confidence across 12,893 visitors, still running two years later
+20%Food and confectioneryOffer page against their product page, Meta traffic
+13.5%Streetwear denim99% probability to win across 92,484 sessions

How it runs

  1. We rebuild one page. Usually the product page, because that is where the money is decided.
  2. It runs as a real split test. Your visitors, split between the new page and your current one.
  3. The result decides. Up to five iterations. If revenue per visitor never clears a 10% lift, you owe nothing.

Why revenue per visitor

A higher conversion rate at a lower order value can leave you worse off. Revenue per visitor cannot be flattered that way.

Both versions run at once on the same audience, so a strong week or a change in ad spend lands on both. What remains is the page.

Want the full case studies?

See the measured results