Now, I’ve read a website called conversion-rate-experts.com and I recommend that you go and download their free report, too, ’cause I’ve got a very limited mathematical brain.
It explained in very simple-to-understand terms why you need to make sure that you get to a point of statistical significance in your experiments.
So they recommend that you wait until you’ve had 100 conversions on each variation. So say, for example, you’re testing one headline against another, you’d need to wait ’til at least 100 people had come through each of those headlines and ended up on the thank you page before you could say with absolute certainty which one’s the winner.
I suspect Google Optimizer is going to be able to tell us much quicker than that but that’s why I want to encourage you only to test one thing at a time and let’s see what happens over time.
My name is Nicola Cairncross and I can help you market your business better online. Find out more and get my new book at http://NicolasBook.com