Split testing question - how is this possible?
So once I got enough data in here, the results for all 3 pages should be identical - because the traffic is the same and the pages are identical. Now here are the results so far:
Page 1: 118 sales and a 5.31% conversion rate
Page 2: 110 sales and a 5.17% conversion rate
Page 3: 123 sales and a 5.60% conversion rate
Looking at this Page 3 is the clear winner - but obviously not since they are all the same.
So my primary question is, if the data is this wild after 350 sales - just how long do I have to wait before I see statistically valid and meaningful numbers?
We are doing a good number of sales but at this rate it looks like I'll have to wait for 1,000 sales and that's insane, and so surely I must be doing something wrong. I've got a lot of things I want to test and it will take forever at that rate.
Anyone?
My only guess is that this is the type of thing you will see when testing pages that really don't produce a statistically relevant difference in response (which makes sense in this case since they are the same pages) and that when you have 2 or 3 pages that are just too similar the numbers will bounce around in a "meaningless" way for awhile until you have tons of data at which point they should all even out??
But if so, how do you test for small but meaningful increases? I mean if you create a new page that converts 5% better than your control, that can add up to a lot of money over time. But it seems like you'd have to wait for 1000s and 1000s of sales before you knew the results of a 5% difference were statistically valid ... ????
(One thing I know with 110% certainty is that the tracking and split test #s are correct.)
Patrick
"The business that considers itself immune to the necessity for advertising sooner or later finds itself immune to business."
connecticut internet marketing - Get in, TAKE OFF!
business card printing