We've setup an A/B in Optimizely to test a completely new design:
that we're testing against our legacy design:
Optimizely is setup to redirect 50% of our CPC traffic to the legacy design.
The old design is beating the new with about 10% better convertion rate and in trying to fiqure out why we started looking at if there was anything standing out in Google Analytics.
Optimizely creates the events with labels "New design" and "Legacy design" for us in GA.
And filtering on these and looking at "Screen Resolution=(not set)" we see some weird numbers:
If we remove all traffic with "Screen Resolution=(not set)" the new design instead performs 10% better than the old.
Anyone that can enlighten me?