[T172]Testing In Software Development

I read once about a man who tripled his conversion rate by changing a single letter in his headline. While that kind of improvement is not common (not by a long stretch), you can quite probably double your conversion rate after testing 10 or 20 items on your page. With that said, there are many things to consider when getting started with split testing.

I can't discuss every feature of every testing tool out there, but I can give you a major overview of the different kinds of split testing software.

One big distinction is between single variable and multiple variable (multivariate) testing.

With single variable testing you just pick one element on your page to test. Take your headline, for example. If your current headline says "Are you having trouble with tooth decay?", you might want to see how "Do your kids have tooth decay?" works. In a single variable test, you just find a way to rotate each of those headlines on your page, and keep track of how many sales each generates.

Paul Hancox has a simple split testing tool. And Google has one that's free, though it takes a bit to set up all your accounts to use Google's tool. And the reporting is not real-time.

Single variable testing can be slow. How would you like to test 10 times as much with the same amount of traffic? Well, you can. You use a method called multi-variable, or multivariate testing. This allows you to test several elemens on your page simultaneously.

Among multiple variable software solutions, there are hosted plans and self-install plans. If you want to install software on your server (which can be faster and cheaper) you will have more control over your data. And you can usually find someone to install it for you (often the vendor provides this service for a nominal fee). If you want to go with a hosted plan expect to pay more.

Hosted solutions include Google's Website Optimizer, and Vertster. Installed solutions include the Split Test Accelerator and Kaizen Track.

Taguchi testing is only one kind of multivariate testing. There are also "full factorial" tools, and random independent rotators. When you run a Taguchi test, you run only a small fraction of the possible combinations of your factor levels. You might, for instance, run only 18 pages to test 4374 combinations. This is made possible by orthongonal arrays. This allows you to find the best combination with out running through all of them. You do have to be careful about your factors interacting negatively with each other, but if you take care as you design your test, you can avoid the bulk of the problem with these. The Split Test Accelerator, Vertster, and Kaizentrack are Taguchi testing tools.

Full factorial tools allow you to cycle through all the combinations for your factors. So if you have 3 headlines, 3 offers, 3 P.S. statements, and 3 images you're testing, you will cycle through 81 pages. One advantage of full factorial testing is that you don't have to worry as much about interaction effects. Good and bad interactions should show up in the data. However, if you want to receive this advantage, you have to run many times the traffic through your test as you would with a Taguchi test. You also can't run as many combinations. Google's website optimizer is a full factorial tool.

The Taguchi and full factorial tools both use arrays to maintain a relationship between the options being served for one factor, and the options being served for another. Some split testers don't do this, though. They simply treat each factor as independent from the other factors. This leads to more flexible test designs, but also to longer and less accurate tests.

That sums up some of the main difference among split testing tools. There are many other features to compare, but this should give you the big picture.