Google Content Experiments WITH VIRTUAL PAGEVIEWS!
First – the basics on virtual pageviews
If your Google Analytics implementation is just one step above the vanilla install you may well be using ‘virtual pageviews‘
We have an example page where we’re tracking pageviews using virtual pageviews or VPVs: http://www.conversionworks.co.uk/vpvhack/index.html
The page tracker on this page looks like this: _gaq.push([‘_trackPageview’,’/vpv’]);
See the extra parameter on the __gaq.push? When you hit this page it doesn’t track the page in Google Analytics (GA) as /vpvhack/index.html but instead you’ll see /vpv in the content report.
Virtual pageviews with Content Experiments…
Google recently rolled out content experiments as we recently posted about. Content Experiments (CE) require you to specify the URL of the page you’re testing and the URLs of the page variations you’re testing against. CE will then track views on the control and experiment pages using the normal GATC (Google Analytics Tracking Code) and track the test goal using GA goals – neat!
But what happens when you specify the URL of a control or variation page that tracks pageviews using a VPV?
Well, nothing bad! It works really nicely with no extra thought required.
To prove this functionality we set up a quick test experiment:
Control URL: http://www.conversionworks.co.uk/vpvhack/index.html
Control VPV : /vpv
Variation URL: http://www.conversionworks.co.uk/vpvhack/indexb.html
Variation VPV: /vpvb
We elected to not over write the experiment URLs. The experiment goal was event based to track any click on the page. No other changes to the codes or configuration were made – we just let it bake for a couple of days…
Here are the results:
Visits and a conversion. Small beans maybe. No conclusive results but that’s not the point.
The point is that Content Experiments are insanely simple to set up and run even with sophisticated Google Analytics implementation.