Don't Fear the A/B Test
The reason most website owners opt not to test the variables of their website landing pages may be attributed to a fear of the unknown. It may also be that testing may cost some serious coin (money) so while it's important to do it, test only when you are confident that you actually need some improvement in your conversion rates. Most online enterprises don't test because they either they do not know that they should be testing or simply believe that it's too complex. What's going to happen if I change my headline? Will users still buy if I switch images on my landing page? Where is the best place for a testimonial?
I'm here to tell you that 1) you should always test, 2) it is not that complex and 3) there is absolutely nothing to fear but fear itself. You can't possibly increase conversion rates sitting around staring at your computer thinking about it what it would take to get it done and day dreaming of all the fabulous results you will receive once you achieve the perfect mix of page components.
Here are a few things to consider when testing:
- Limit the number of variables at first - they don't call it A/B testing for nothing!
- Focus on the big variables - the one that will make a difference cost, headlines, offer, etc.) then move to smaller variables (locations of variables, color of variables, etc.)
- Sample Size is important but clicks to conversions is equally, if not more, important
You may find that just by focusing on overall site improvement for increased conversions and less on actual rankings (or that little green bar everyone seems to be so concerned with) you'll receive better overall results from the investment in your online enterprise.Technorati Tags: a/b testing, multivariate testing
Del.icio.us Tags: a/b testing, multivariate testing