Comments on: Google Optimizer is slow (or Not all Multivariate Testing is the same) http://widemile.wordpress.com/2008/01/28/why-google-optimizer-is-free-its-old-and-slow/ Learning about optimization at Widemile Sat, 19 Jul 2008 11:55:18 +0000 http://wordpress.org/?v=MU By: Marketing Mix 3-10-08 http://widemile.wordpress.com/2008/01/28/why-google-optimizer-is-free-its-old-and-slow/#comment-112 Marketing Mix 3-10-08 Tue, 11 Mar 2008 15:54:05 +0000 http://widemile.wordpress.com/?p=87#comment-112 [...] Any online marketer running PPC campaigns should be actively running a several multivariate tests. This post from Billy’s Blog explores the use of Google Web Optimizer- Google Optimizer is slow (or Not all Multivariate Testing is the same) [...] [...] Any online marketer running PPC campaigns should be actively running a several multivariate tests. This post from Billy’s Blog explores the use of Google Web Optimizer- Google Optimizer is slow (or Not all Multivariate Testing is the same) [...]

]]> By: 3 steps to quickly make a good multivariate test « Billy’s Blog http://widemile.wordpress.com/2008/01/28/why-google-optimizer-is-free-its-old-and-slow/#comment-102 3 steps to quickly make a good multivariate test « Billy’s Blog Thu, 21 Feb 2008 23:23:21 +0000 http://widemile.wordpress.com/?p=87#comment-102 [...] Test the maximum minimum: Pack as much as you can into a test for the amount of traffic you have to keep it a short test. Using Widemile’s platform that’s 2 weeks to be safe, with Google Optimizer you should do at least a month (explanation). [...] [...] Test the maximum minimum: Pack as much as you can into a test for the amount of traffic you have to keep it a short test. Using Widemile’s platform that’s 2 weeks to be safe, with Google Optimizer you should do at least a month (explanation). [...]

]]> By: What is Taguchi? How does it relate to testing? « Billy’s Blog http://widemile.wordpress.com/2008/01/28/why-google-optimizer-is-free-its-old-and-slow/#comment-91 What is Taguchi? How does it relate to testing? « Billy’s Blog Thu, 14 Feb 2008 16:19:18 +0000 http://widemile.wordpress.com/?p=87#comment-91 [...] the term Taguchi is fractional factorial test design. (I discussed this at length in my post about the difference between Widemile’s technology and Google Optimizer.) The Taguchi method uses a fractional factorial test design and is under the umbrella of [...] [...] the term Taguchi is fractional factorial test design. (I discussed this at length in my post about the difference between Widemile’s technology and Google Optimizer.) The Taguchi method uses a fractional factorial test design and is under the umbrella of [...]

]]> By: Billy http://widemile.wordpress.com/2008/01/28/why-google-optimizer-is-free-its-old-and-slow/#comment-75 Billy Tue, 05 Feb 2008 17:32:27 +0000 http://widemile.wordpress.com/?p=87#comment-75 Thanks for the comment. This is a fascinating subject and so I'd love to know more. I will get in touch with you shortly, Daniel. Just from the tiny bit of information of your post, it seems like what you are doing is not the intended method by Google and so requires more manual work and analysis on your side, while it is all automated since it is the implied methodology in Widemile's platform. I don't want to say more until I know more about how you are testing though. It excites me to see that people are skilled enough to truly use Google Optimizer beyond its out of the box capabilities! Hopefully we can discuss and I can possibly post an updated version of the blog. Talk to you soon, Billy Thanks for the comment. This is a fascinating subject and so I’d love to know more. I will get in touch with you shortly, Daniel.

Just from the tiny bit of information of your post, it seems like what you are doing is not the intended method by Google and so requires more manual work and analysis on your side, while it is all automated since it is the implied methodology in Widemile’s platform. I don’t want to say more until I know more about how you are testing though.

It excites me to see that people are skilled enough to truly use Google Optimizer beyond its out of the box capabilities!

Hopefully we can discuss and I can possibly post an updated version of the blog.

Talk to you soon,
Billy

]]> By: Daniel Shields http://widemile.wordpress.com/2008/01/28/why-google-optimizer-is-free-its-old-and-slow/#comment-71 Daniel Shields Mon, 04 Feb 2008 19:10:16 +0000 http://widemile.wordpress.com/?p=87#comment-71 The statement that you make about Google not being capable of Taguchi or 'Fractional Factorial Design' is, in my opinion, inaccurate, or, at best, not complete. I've built and performed multiple sampled factorial tests and have instructed others on how to do so. The speed to results is hastened by the drastic reduction in variations which you described. Its a matter of how you structure the test, produce viable and sound Taguchi variations (as there is only one 'calculator' for this purpose), and the determination to correctly implement those variations. Another note with regard to this is the proximity to conversion goal. If test completion timeframe is a concern, then, it might be advisable to develop formulae for proximation. I can illustrate some examples for this if you would like, please email me and we can discuss. The statement that you make about Google not being capable of Taguchi or ‘Fractional Factorial Design’ is, in my opinion, inaccurate, or, at best, not complete.

I’ve built and performed multiple sampled factorial tests and have instructed others on how to do so. The speed to results is hastened by the drastic reduction in variations which you described. Its a matter of how you structure the test, produce viable and sound Taguchi variations (as there is only one ‘calculator’ for this purpose), and the determination to correctly implement those variations.

Another note with regard to this is the proximity to conversion goal. If test completion timeframe is a concern, then, it might be advisable to develop formulae for proximation. I can illustrate some examples for this if you would like, please email me and we can discuss.

]]>