Currently we can run A/B testing in funnels. However, the test only tracks the activity for the page where the test is running.
However, and I'm sure I'm not the only one doing this, when you run an A/B test for the landing page of a funnel, the conversion activity needing to be tracked takes place on the following pages of the funnel.
For instance, I'm testing landing page concepts for a book funnel that I'm driving traffic to with Facebook Ads. These pages have no CTA other than for the viewer to choose to proceed by clicking a button to choose an option (e-book or audiobook or both), and then the button the click takes them to a 2-step order form where they start the checkout process, are offered a bump order, and upon completion of that page, are presented with an upsell.
The vital activity that needs to be tracked takes place after they click whatever button (choice) on the landing pages I'm testing. BUT... there's no way to tell which landing page generated more sales because all of the sales activity takes place AFTER they choose what purchase option they want on the landing pages being tested.
There has to be a way to tell which landing page primed the visitor to make whatever purchases take place AFTER they leave that page.
RIght now, there is no way of telling what landing page is performing (priming the visitor) better.