Faster Split Testing with Split Test Accelerator
October 4, 2010 by Mark Widawer
Filed under Featured, Multivariate Testing, Optimizers Blog, Split Testing Tools
Leave a Comment
RSS FEED for: Featured, Multivariate Testing, Optimizers Blog, and Split Testing Tools
One of the more advanced split testing tools ever created is called Split Test Accelerator. And while Google’s Website Optimizer is a competent testing tool (and free, too!), it’s not really very “optimized” on its own.
In other words, if you want to run a multivariate test with 4 factors, each with 3 options, GWO will dynamically create about 81 different web pages to test — and send enough traffic to all of them in order to fully test them all.
Let’s say you were going to take Google’s general advice that you need 100 actions to a winning web page in order to declare a winner. And let’s say you were testing a sales page with a conversion rate of about a half percent, and were hoping to double your conversion rate to one percent.
That means you’re going to need 10,000 visitors to each page in order to find a winner. With 81 total pages to test, that 810,000 visitors.
Now…
How long will it take you to get Read more
Breakthrough Business Results with MVT
October 4, 2010 by Mark Widawer
Filed under Book Recommendations, Multivariate Testing, Optimizers Blog
1 Comment
RSS FEED for: Book Recommendations, Multivariate Testing, and Optimizers Blog
The book “Breakthrough Business Results with MVT” by Charles W. Holland makes the case for multivariate testing on web pages, too. And if you’ve ever heard about Taguchi Testing, it’s a specific form of Multivariate testing — that also just happens to be a much faster method of multivariate testing.
This book is highly recommended if you’d like to geek out a bit and learn more about Taguchi and Multivariate testing. The review is a reprint from Amazon.com.
“Every executive in the world wants the same thing—improved results: more profit, less cost, better quality, and higher customer satisfaction. But what business improvement ideas will produce these results, and how can you be sure that plans that look good on paper will actually work in the real world?
Testing ideas one at a time is too slow and too expensive. Only one revolutionary method offers a powerful, fast, and inexpensive way to prove with certainty what Read more



