Hi Everyone, Does any one know about "Taguchi method" in online marketing? Share your comments and views. I am also digging information on the same. Thanks_
I tried it once, but wasn't very happy with the results. I was using it to test advert text - but I'd recommend against this. It may be better for testing different landing pages... The basic idea is that you don't need to test every combination of your variables. For example, if you write four different options for each of the four lines in your advert, a complete test would require 4 x 4 x 4 x 4 = 256 different adverts, and it would take a very long time to get significant results. But by using orthogonal options you can find the best option for each line in the advert, without doing all of these tests. In this case, 16 adverts should suffice, using (for example) options 1111, 1222, 1333, 1444, 2123, 2214, 2341, 2432, 3134, 3243, 3312, 3421, 4142, 4231, 4324, 4413. Every option for every line of the advert here is tested equally with every option for the other lines, so you should find the best option for each line with a bit of basic statistics. This method is equally easy to apply when testing anything with multiple options. The problem is that these lines aren't independent of each other. The best advert may be 2222, but you didn't test that, and if these lines only work well when combined, they won't come out on top in the test. Similarly, the best thing to put in each of the four areas on your landing page may be a nice, big call to action button, but that doesn't mean that you need four of them on your landing page. There doesn't appear to be much information around about this method, so it's possible that I've misunderstood it, but I do know that the 'best' advert that I produced from my test really wasn't very good at all...