Quote:
Originally Posted by RazorSharpe
Going back to A/B testing; most people take 2 or 3 versions of the page, set them up using their testing too of choice, monitor them, make adjustments and finally see what the outcome was. If their ratios are improved that's it for their testing.
What about a 4th, 5th, 6th variation of the page? Where do you settle? Do you test different pages on different traffic sources or do you optimise for general traffic?
What do you specifically use as a tool for A/B testing?
|
I had a script made for my HTML pages for A-B testing. It splits every other visitor to the desired location/tour page.
For me I try to be 'realistic' about testing. I have no idea what the "rule of thumb" on all this is so I basically just keep tweaking/testing as long as I see a 10% increase or better each time. When I get to a 7% or 8% improvement I typically stop for about six months. Then I do another A-B test to see if pages are still working and in that "zone".
Again, I have no idea if what I am doing is "right" or not. LOL I have many balls to juggle being a small company so I do what I can; I'm sure larger companies are tweaking/testing more often.