|
||
Taguchi testing, also called multivariate testing, is a way to test lots of different variables at the same time. Say what? Well, testing is sometimes not a clear cut issue. Say you want to find out if one thing is better than another. There may be many different variables involved in making that decision. In regular testing, one variable may be used to test, or several different tests can be run using one variable each. But with Taguchi testing, several variables can be tested all at once, saving a tremendous amount of time and giving a much more accurate result.
So what does this have to do with online marketing? Taguchi testing wasn't developed to be used online at all. It was started to test factory production methods, such as car makers, to improve the way factories could run their expensive tests. But what began on the factory floor has now been proven to be useful elsewhere- including cyberspace.
Taguchi testing has been found to be valuable with PPC ads, web copy, landing pages, and other forms of online advertising. To use the Taguchi testing method, choose an area that you want to be tested. If you have a sales or landing page that you are interested in testing, decide on the variables that might affect that page. This can be the font style, size, color, the layout, the graphics, anything that you are thinking about changing. Alternatives then have to be created for each item that may be changed. This would be choosing an alternative font, or alternate graphics, etc. The next step is to set up several mixes that combine the different variables into different arrangements. The multivariate testing is then done with specialized software made for Taguchi testing.
Are the tests worth it? Think about the tinkering you do to your website or sites. There are likely regular font adjustments, taking out this paragraph, adding another, updating data, adding a new graphic. Imagine if you could do it all at once, and be able to preview all of the different variations that your changes might take before committing to them. What variations might the testing uncover that you never would have thought of yourself? And think of the time that will be saved getting all of this testing over at once.
Some of the variables might not have been things you thought to test in the first place. Even page elements that you thought were fine might end up looking a lot better in combination with other changed variables. Background images, link colors, all of these things might get tested in he next year, you know, after you get around to testing that top paragraph sometime. But putting them, all together, and getting all of this year's testing out of the way in such a short time, the time savings alone is worth exploring this type of testing. And, you may find combinations that bring in more buyers with your new and improved page.