The Google Website Optimizer

You just have to give credit to the folks at Google. They have made it possible for the small business owner to use sophisticated conversion rate tools without the need to purchase expensive software or consultants. The Google Website Optimizer is an easy but powerfully effective way to drive up your page conversion rates.

Setting up your first test with the Google Website Optimizer is pretty straightforward. Once you get used to how the Optimizer works, you'll be able to set up tests within minutes.

I'm going to take you through a step-by-step example so you can see how the process works. After these instructions, you will be able to follow the very same process to set up your own tests.

Step 1: Choose the Page You Wish to Test
The first step is to decide which page on your website you want to test. My recommendation is to start with the one page that brings in the most profits for your business.

Step 2: Decide the Testing Parameters
Decide which elements on the page you would like to test. Don't attempt to test half a dozen parameters at once. Pick a 2-3 elements that you feel would be critical to page conversions. Below are some ideas of what you can test.

• Main headline
• Testimonial usage
• Price
• Product image
• Bonus offer
• Use of "trust" symbols (e.g., VeriSign)
• Image captions
• Font emphasis
• Primary offer

Although I've had you identify 2-3 items you wish to test, pick the one item that you feel would be the most important factor to test first - maybe your main headline or the price of your item.

The reason for picking only one item for testing is that we will begin by setting up a simple A/B split test. A split test allows you to test a single change to see if it makes a difference in conversions. It's the simplest test to initiate and takes the least amount of time to get meaningful results. After you gain some experience, you'll then find it easy to set up more complex multivariate testing, where you'll be able to test more than one parameter at a time.

Step 3: Log Into Google Website Optimizer
Google calls the various tests you set up Experiments. To get to the main experiment page, log into your Google website optimizer account and select "Website Optimizer" from the "My Account" screen.

After accepting Google's user agreement, you will be taken to your Experiment List page. This is where you will create your 1st experiment.

Step 4: Identify Your Conversion Page
You already know which page you want to test, but now you must identify the page your visitor much reach in order to qualify them as a conversion.

This page is typically the 1st page your visitor sees after registration or making a purchase. For a newsletter signup, it would be your registration confirmation page. For a product purchase, it would be the order thank-you page.

Step 5: Enter the Experiment Parameters
From the Main Experiment List page, click "Create Experiment" to get started. You will be asked to select whether you want to create an A/B Split Test or a Multivariate Test.

For your 1st experiment, select the A/B split test. This will minimize the time it takes to complete the test and will also allow you to obtain a bit of confidence with the optimizer program before you start running more complex multivariate tests.

Create the alternate version of your page and upload it to your website. This will be your "variation" page. If you have decided to test a headline, you would duplicate your landing page, change to the new headline, save it with a unique URL name, and then upload it to your site.

Once you have the page URL for your original page, test page and confirmation page, fill out this information in the Google Website Optimizer screen, including the name you wish to give the experiment. When you are finished, click on the "Continue" button.

Step 6: Install the Javascript Tags
In order for the Website Optimizer to know how to divide up the original and variation page as well as determine the conversions for each, you must paste the special Javascript code that Google provides after finishing the previous step.

Google will provide you with control and tracking scripts you must copy and paste into the original landing page. You will also receive scripts for your variation as well as confirmation pages.

Go ahead and paste the codes into your pages and upload them to your site. Google does an outstanding job of walking you through this process but it you have any confusion, just remember to click on the little question marks all throughout the page. When you do, a help box will open to provide you more information on that particular topic.

Once you have pasted the Javascript tags into the HTML of your pages, click on the "Validate Pages Button." Google will go out and take a look at your coding in order to verify that everything is in proper order. If there is any problem validating any of your tags, Google will produce a detailed list of issues to help you troubleshoot.

Once your tags validate, you will be able to preview the various page versions by clicking on the Preview link.

Step 7: Review Settings and Launch Experiment
Double check all your experiment settings. You'll see a setting on the review page titled "Amount of Total Traffic to Use." Make sure this is set to 100%. That way, each and every visitor to your site will be included in the test, minimizing the time for statistically relevant results.

You're all done. Go ahead and launch your first A/B split test experiment.

Step 8: Monitor Your Experiment
After launching your experiment, Google takes you to the experiment workflow page where you can start monitoring the various parameters such as number of impressions and the conversions tracked to date. It will usually take a few hours for Google to start reporting on result data so be patient.

Google will let you know when enough data has been collected to determine a clear winner. The amount of time this takes depends on the amount of traffic you get to your landing page. If you only get 100 visitors a day, the experiment will need to run a bit in order to produce statistically valid results. If your page gets 1,000 visitors a day, it won't take long to complete you test.

If you want a tool to help estimate how long your various experiments will take, Google provides a handy little tool you can use by clicking the link below.

Step 9: Interpret Your Results
The website optimizer has several different ways you can view your results data. As you are viewing various reports, I highly recommend you click on all the question mark icons. This will bring up windows that explain what each column heading means.

In the below multivariate test report, it is clear that combination 11 is beating out the original landing page in conversions. Combination 11 should be the new landing page for all visitors - that is, until you test and find a version that has an even better conversion!

You'll want to take some time to understand all the various Google website optimizer reports. Fortunately, Google provides a great help page to get you going.

Make no mistake, testing your landing pages is the number one way to increase your conversion rates. As soon as your business site grows to 100 unique visitors per day, start testing and never stop!