Q and A: What is Google Website Optimizer?

Posted At : June 26, 2009 10:16 AM | Posted By : Congruent
Related Categories: Web Design, Search Engine Marketing, Q&A, Technology

Every week we take Internet marketing questions from friends, acquaintances and client business partners and unravel them at the end of these enewsletters. Is there something you would like to know about the weirdly wonderful world of websites? Ask us here and we shall answer.

Question: What is Google Website Optimizer?

Answer: Per Google's dry summation:

Google Website Optimizer is a free A/B testing and multivariate testing application that helps online marketers and webmasters increase visitor conversion rates and overall visitor satisfaction by continually testing different combinations of website content. But, if you're a website owner or an Internet marketer with "mad scientist" tendencies…

Google Website Optimizer is your new wormhole to The Twilight Zone and your website is now your own "Choose Your Own Adventure" novel! But feel free to leave your dark ironies and quantum mechanics at home because this interface couldn't be more simple, or the plot twists more beneficial to your marketing plan.

That's great! But, wait, seriously, what is "A/B Testing and Multivariate Testing?"

In short, GWO allows you to create separate and distinct versions of your site content and record and compare the success of each version as tested on real-life visitors to your site.

A/B Testing allows for dead-simple testing of two or more stand-alone versions of a page; Test Scenario A vs. Scenario B (and C, D, E, etc.). For example, in Scenario A, 50% of your traffic visits a version of your homepage that features your sale on Flux Capacitors above the fold, in the headline, main image, sidebar links -- everything. In Scenario B, you feature a sale on a similar product -- say, Mr. Fusion -- to the other 50% of your traffic.

In this test, the A/B visitor groups will be as numerous as you could ever hope for and maybe you'll learn definitively, through clicks and conversions, that Mr. Fusion is generally the more popular product between the two -- something that might not have been so obvious in a split-time test or if the two products were simply buried in your site architecture.

Multivariate Testing is different in that it will let you test just one component of a page or a group of components. Maybe you want to try one callout's performance against another. Or use a multivariate testing to refine the little things; find out which technical writing or instructions set makes more sense to your visitors.

If, like a lot of businesses, you have a difficult time picking one creative option over another, multivariate testing is a great way to both solve the dilemma and pick the correct option.

What's an example of a goal?

CPO (conversion rate optimization) is the main focus of A/B Testing and Multivariate Testing. The more tests you complete, the better elements and scenarios you host, the more visitors turn into customers. Examples of typical conversions include:

Product sales -- It's hard to argue with instant gratification. If you can steer more people to sales forms, do it. But don't stop there; use A/B and Multivariate testing to optimize that form!
Requests for proposal -- Optimize your service descriptions for more jobs.
e-Newsletter sign-up -- Maybe you need to be more descriptive to tease new subscribers or maybe you need to be more discreet. You won't know until you test.
e-Coupon registry -- Sometimes people would rather place purchases in real life. But that doesn't mean people won't opt-in for coupons! "But, hm, which promotion is more attractive? What's the best deal we can make and still draw a crowd?"

Why should I test?

You should test different scenarios because one will prove to be better than the other and won't you be glad when you know which one it is? Also, your coworkers -- the people who usually help you pick your paths -- are a much smaller test pool than all your site visitors.