200-Proof Marketing

Google had a surprise in store for me today when I checked my Adwords account. They made their Multivariate Testing tool called Website Optimizer available free for Adwords customers.

Evidently they’ve been testing it for a while, but this was the first I had heard of it.

This tool could easily double your profit in a fraction of the time it takes old A/B Split Testing.

Multivariate Testing is the king of testing (in my opinion). Some might favor Taguchi but from what I’ve seen it doesn’t work as well for optimizing websites.

So what’s the difference between Multivariate Testing and A/B Split Testing?

A/B Split Testing only tests two different webpages at a time. Usually the two pages are identical except for one part (like your headline).

Visitors are split between the two pages randomly and the one that converts the most visitors into sales wins. Slowly and over the course of many tests, this can greatly increase your conversion rate (the percentage of visitors that buy or take action).

Multivariate Testing is somewhat similar except it allows you to test many different parts at a time. In a sense, it allows you to run many Split Tests at once.

In a Multivariate Test you can test your Headline, Price, Guarantee, Testimonial placement, and more all at once.

Let’s say you want to test your Headline, Guarantee, and Background Color. You decide to test 3 different headlines (the original and two new ones), 3 guarantees, and 3 background colors.

That would give you 27 different combinations of your sales page. Each time a visitor comes to your website they will be randomly shown one of these combinations.

One visitor may see headline 1, guarantee 3, and color 3 while the next visitor will see headline 2, guarantee 1, and color 3.

You can then not only view which variables of the different parts do the best but also which combinations do the best.

The bottom line is Multivariate Testing allows you to increase your conversion rate much faster than A/B Split Testing would allow.

But here’s the kicker…

The fastest way to increase your conversion rate when testing is to find which parts of your webpage effect your conversion rate the most and then find the best options for those parts.

For example: On your first test you test every part of your webpage you can think of. When the test is over you can see which parts increased or decreased your conversion the most.

Next, you retest only those parts.

It’s the 80/20 rule in action (look here for my article on the 80/20 rule). A small percentage of the parts of your webpage, have a large effect on your conversion rate.

I will start some tests tonight with the new Website Optimizer tool and give a more detailed report later.

I’m fairly confident it will replace the Multivariate Testing software I created simply because it will be easier to set up and more convenient.

Much more on this and testing in general coming soon…

This entry was posted on Tuesday, April 17th, 2007 at 11:23 pm and is filed under Internet Marketing, Testing & Tracking, Test Results. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.