Getting Started with Website Optimizer

Overview of Experiments with Website Optimizer

The intent of all Website Optimizer experiments is to find the best-performing content for your page. What specifically "best-performing" means can vary from one page to another, depending on the goals of your site. A few examples:

• an e-commerce site can test whether users complete a purchase more often when they see pictures of the product or people
• a newsletter site can test if the headline "Cheap Widget Fixes" gets more paid sign-ups than "Fix Your Widget Fast" or "Need a Widget Fix?"
• a forum site can test whether more visitors register if the "Join Now" button is displayed on the right or left

You can run two types of experiments with Website Optimizer: A/B tests and multivariate tests. Both types of tests help measure customer reaction and improve return on investment. However, there are differences that make one test better than the other in certain situations.

A/B experiments allow you to test the performance of two or more entirely different versions of a page. There's plenty of design freedom with A/B testing: you can alter the look and feel, or move around the layout of your alternate pages. For example, to test whether a "Buy now" button performs better on the very top of the page, or in the middle, use an A/B test. It's the simpler type of test, and is the better type of test for pages that don't get a lot of traffic.

Multivariate tests, on the other hand, let you test multiple variables -- in this case, sections of a page -- simultaneously. For example, you could identify the headline, image, and promo text as parts of your page that you'd like to improve, and try out three different versions of each one. Website Optimizer would then show users different combinations of those versions (let's say, Headline #2, Image #3, and Promo Text #1) to see what users respond to best. Multivariate tests are more complicated, and typically require higher page traffic than A/B tests.

The type of test that you should choose depends on the type of website you have, and what you want to test. The steps involved in setting up an experiment are similar for both types of tests:

1. Identify the pages to test
2. Create alternate content
3. Add the experiment tags to your pages
4. Run the experiment
5. Study the results.

Identifying Webpages and Site Content for Experiments

When running an experiment, before doing anything else, you should choose the pages that you want to test. There are two essential pages to identify: your test page and your conversion page.

Your test page is the page that you'll be optimizing by making changes with Website Optimizer. To get results quickly, you want to choose a test page that receives a significant amount of traffic. Most importantly, your test page needs to have an impact on whether or not users take a desired action on your site, like a purchase, download, or sign-up. This action can often be in the form of a link to another page on your site, possibly your conversion page.

Your conversion page is the page that represents business results for you -- whether it's the page where a user makes a purchase, fills out an interest form, or downloads a white paper. When choosing a conversion page, consider the ultimate goal for users on your site. For example, if you want to measure what content on a product information page leads users to make a purchase, then your conversion page should be the final "thank you" page displayed when a purchase is complete. Or, if your purchases happen rarely or require offline interaction such as speaking with a sales representative, you can measure conversions based on intent to purchase. A conversion could be clicking a "More information" link, staying on a product page for a certain time period, or using click-to-call.

After selecting these pages, identify what specific content you'd like to try out on your test page. Common elements to test include headlines, layout, styles, images, promotional text, and conversion incentives. What you want to test will determine which type of test you will run: A/B or multivariate.

• To try out different layouts, move around sections of the page, or change the overall look and feel of a page, an A/B test is typically the right choice. A/B tests allow you to test two or more entirely different versions of a page, and give you complete design freedom.
• To test several variations on several sections of your page at once, a multivariate test is usually the best selection. Multivariate tests give you the flexibility to test several headlines simultaneously with several images, for example.

Once you've identified your pages and the type of test you'll run, the next step is to create the alternate content. For A/B tests, that's the alternate page variation -- the new layout or design of your page. For multivariate tests, it is the variation content for each page section -- for instance, the three new lines you want to try in your headline. During the experiment, visitors will see either the original content or the new variations you've created. This is so the experiment can determine which content leads to more people taking the desired action and reaching your conversion page. We'll show these results in your Website Optimizer reports.

Try it Now: Using Google Analytics to Determine Webpage and Site Content

You can review Google Analytics data to help you decide which webpages you may want to test using Website Optimizer.

Follow the instructions below to run a "Top Landing Pages" report and analyze performance data.

1. Sign into your Google Analytics account at www.google.com/analytics.
2. Click Content from the menu categories on the left-hand side of the page.
3. Select Top Landing Pages.
4. Above the table to the right, locate Views.
5. Select the fourth option, which will display a comparison of the Bounce Rate for the top landing pages with the site average.