A rarely discussed and utilized property of the Taguchi Method is the study of interactions. Dr. Taguchi incorporated additional power in the use of orthogonal arrays for test design via the linear graph. Using these linear graphs it is possible to assign factors to the correct columns in order to study interactions between the factors while also revealing the effects of the individual factors.
Although it is not possible to study all interactions as can be done with the full factorial test, you can use your knowledge of Read more
OK, that title got your attention! What I am talking about? Well, some people promoting Taguchi methods of testing suggest that you can eliminate under performing test runs during the test. Here is a quote from the guy that runs StomperNet’s test software:
Lets say you are running a L18 — this has 18 different landing page recipies. While the test is running you will naturally see a couple of recipies that are doing significantly worse than the others. As this happens you can shut off these recipies. You can keep doing this until you are down to two recipies — i.e. - a split test which will confirm the results from the teguchi test.
The key to testing is the creation of your test design. With our software, you can create each test factor and its variations or levels using a WYSIWYG editor. This allows the full design to be done without any coding. But you can if you want to, because the editor does have an HTML tab.
You can also change the CSS and apply the change to only the test page or to a defined class across the entire site. This will allow you to change a background color or the font and apply the change across the site so that the visitor gets a consistent presentation.
There is also a Text edit button that allows you to specify a string and its replacement. This can be applied to only a single string or all occurrences of the string on the page.
As you develop test factors with variations, you can preview any combination you wish to by selecting the factor variations and click preview.
You can create several factors with just one variation on each and then use them all in a split test. The split test will create a page with all the variations applied and test it against the control page. Sometimes this is the quickest way to get started because you can generate a new best performing control to take into the Taguchi test.
The WYSIWYG editor makes the whole process friendly and easy.