You've got a website and it's selling some copies of your product. You want to sell more, but how do you properly test the changes you make in order for you to know how you are improving your sales?

Internet Marketers talk a lot about testing your sales copy and your website, but very few talk in detail about the methodology of this testing. It is very easy to get confused and undo the changes that were increasing sales!

I remember when I was at school in my chemistry class performing experiments to blow things up, make acids to eat through the table and stain my lab coat being taught how to properly test an experiment. What is interesting, is this is probably the most valuable lesson I got from the chemistry class (apart from the effect certain metals have when exposed to water). It is certainly the lesson that applies the most to my life now.

My chemistry teacher was an interesting man, he didn't tolerate people who didn't listen or didn't want to learn but he was a passionate man when it came to chemistry. His fingers were stained from handling chemicals and he took an almost perverse pleasure in gassing the class or waking them up with a sudden explosion.

Then one day when he had given us a series of experiments to perform, he taught us about testing. When performing an experiment there are often many things you want to change in order to perform the experiment properly. The trouble is, as an impatient teenager waiting for lunchtime you are tempted to make all the changes at once.

"No lad," the teacher told us, "You can't make all the changes at once. If you do that, how are you going to know which change to the experiment caused it to work?"

Of course, this concept hadn't occurred to any of us, we were all just interested in lunch.

He leaned back in his chair into his reminiscent pose, "In order to properly perform this experiment you need to make a single change at a time and then measure the results. Then, and only then, will you know which change performs the best and gives you the desire results. Sure you can make all the changes all at once, but what if you undo the change that works?"

These very words came back to me when I started my Internet Marketing career some years ago. If I am making changes to my websites or sales copy to monitor the conversion rate then I need to make a single change at a time, drive a sufficient amount of traffic to it, and then analyse the results.

If the conversion rate has increased, then I keep the change. If it has decreased, then I discard the change.

By performing my website testing this way you will know exactly which changes best affect your conversion rates (and profits) and be able to stack them on top of each other to gain the highest possible conversion rate for your site.