This is the fourth post in a series that focuses on using the various tools located within the Google AdWords tools and analysis tab . Previous posts have focused on:
- The Google AdWords change history tool 
- The AdWords conversion tab 
- Google Analytics Reporting within the AdWords Interface 
And in this post we’ll be walking through how to leverage Google Website Optimizer within your AdWords account.
How You Can Use Website Optimizer within Your AdWords Account
Website Optimizer allows you to set up split tests and multivariate tests to measure the impact of different landing page designs within your AdWords campaigns. Implementing Website Optimizer requires adding some code to the pages you’re including in your test.
The first (and most important) step in making use of the tool is, of course, creating multiple versions of your landing page to test. Most of this post will focus on the mechanics of using the tool to test the creative you’ve put together, but obviously there’s no real value in a tool to test landing page variations if the variations themselves aren’t well thought out and valuable. If you’re looking for ideas for tests and/or conversion optimization best practices, the Unbounce blog has some amazing resources. (Check out their recent post with 35 examples of great landing pages , plus critiques on how they could be even better.)
Once you have your variations in place, setting up a Website Optimizer test is actually fairly straightforward. Open up the Website Optimizer and click “Create a new experiment”:
Then you’ll be asked to choose between a split test and multivariate test:
The type of test you want to set up depends on multiple factors, but for most small to mid-sized AdWords accounts a key consideration will be the amount of traffic you’re sending to the page: if you’re not doing a large amount of PPC volume it’ll likely be very difficult to get to statistical significance with a multivariate test. We’ll walk through setting up an A/B split test .
First you simply input the following pieces of data:
- Name your experiment – Choose something clear that will be easy to remember and to identify for people new to the test.
- Identify the original page (the page you’re currently pointing ads at) and the variation (the test you’d like to set up).
- Identify the conversion page – This will typically be the thank you page that is displayed after the user completes the action you want them to complete (filling out a form, making a purchase, etc.).
If you choose to do it yourself, you’ll be presented with very specific instructions on where to install the code within the source code on your website. You can also validate that the code is installed properly on your pages before moving past this stage:
Finally, you’ll be able to track the progress of your tests within the Website Optimizer dashboard and even drill down to the daily breakdown of the test performance over time. Once you’ve enabled your test, Website Optimizer will not only track progress, but let you know when one of your tests has reached statistical significance and point out the “winner.”
There are a lot of different landing page software options that offer varying feature-sets and plusses and minuses in terms of ease of integration, and if you’re looking for things like more user-friendly GUIs, pre-designed templates for landing pages, and more advanced testing options some of the paid options below might be a better fit:
But for many advertisers Website Optimizer is a highly useful free tool for launching landing page tests.