The greatest online marketing campaign in the world won’t compensate for a badly designed landing page or website.
With the costs of online marketing continuing to grow, its vital that small business owners take the time to review and optimise their website conversion rates. Even small increases in conversion rate can lead to significant improvement in campaign ROI (Return on Investment).
Given Google’s vested interest in your marketing success (considering they make the majority of their money from paid search advertising), it seems logical that they’d deliver one of the best conversion optimisation tools available. And the best part – it’s free!
This simple guide will explain why Google Website Optimiser is important for your website, how to implement the tool and then how to set up and measure conversion experiments.
What Is Google Website Optimizer?
Straight from the Google Website Optimizer site:
“Website Optimizer is a free tool which allows you to test different variations of your site’s content to find out which combination results in the highest number of conversions (proportion of traffic which converts in to sales/leads).”
How to Install Google Website Optimizer?
For Google Website Optimizer to track your website performance and deliver content variations, you need to install some “code” into your site. Here’s the guide from the Google Website Optimizer Help Center:
1. Getting started
To access Website Optimizer please visit the following link.
Read and accept the terms of service if you haven’t already done so.
Click Create a new experiment.
Select Create from the A/B Experiment field.
Confirm that you’ve completed the “Before you start list“ by selecting the checkbox and clicking Continue.
2. Setting up the experiment
Enter an Experiment name.
Enter the Original page URL – this should be the original page of your website which you’ll be testing. It should also be accessible to our validation, meaning that it can’t be hidden behind a password.
Enter the Page variation URL – this is your first alternative page, which will replace your original page for some percentage of your visitors. Again, make sure the page is accessible. You can give this (and all other alternative pages) any name you like. The name will be used in your reports, but won’t be shown to users.
Click Add another page variation to add more alternative pages.
Enter your Conversion page URL. This is the page that marks a successful visit, whether it be a purchase, signup, or article view.
If you’ll be adding the code to your pages yourself, select that option and click Continue. The next page will provide you with step by step instructions on tagging the relevant pages.
Once all of your pages have been updated and uploaded to your server, click Validate pages. Website Optimizer will visit your pages and check for correct installation of your code. If there are problems with the code, you’ll be notified of the error and told on which page the error occurred. Please fix the code and click Validate pages again when you’re ready to validate.
If everything checks out okay, click OK then Continue to proceed.
4. Preview and setting up your experiment
Your experiment is ready to go at this point – if you’d like to see all of your pages, click the Preview link. If you’re ready to start the experiment, click Start experiment. We’ll start the A/B test immediately, and data should be available in your reports within 24 hours.
If you are looking for more advanced technical instructions you can download our techie guide (English) for free.
Planning Your First Experiment:
There are lots of things you can test in a conversion experiment. It’s always best to take a systematic approach by following these simple steps.
1. Choose the page you’d like to test
For the most part, you’ll be testing the landing page of your advertising campaign. However, any high-traffic page is a good one to test, whether it’s the home-page or a particular product page. As a general rule, pages with lots of traffic are generally faster to optimize than low traffic pages, since trends in results become clear quickly.
You will also need to decide what content to test on the page. For more ideas on what variables to test, please see our next section on “Identifying what content to test“.
2. Create alternative versions of the test page
Once you have decided on what page and content to test you will then be ready to create new page variations. Create different versions of the page you’re testing, and upload them to your server. You can vary as much or as little of the page as you like – Website Optimizer will display each of your alternative versions to different visitors. All of your page variations should lead to the same conversion page, meaning they should all be directing the user to take the same action (whether that’s completing a purchase, or submitting an enquiry).
For your first test we would recommend testing 2-4 page variations, since with more page variations, it can take longer to see results.
3. Identify your conversion page
Figure out which page of your site represents ‘success’ – it could be a purchase or enquiry confirmation depending on the business type. Once you start testing your variation pages, WO will measure what proportion of visitors who see each variation go on to reach your conversion page. In this way, you’ll be able to see which page is producing the best results for you (sales, leads etc).
What Content To Test in Google Website Optimizer?
Theoretically you can test any element of your web page, but here the main elements that will deliver serious impact on your conversion rate.
Headlines: Use the headline to test different pitches and see how highlighting different benefits of the product or business affects results.
Images: Use the images to find out whether your users respond better to graphics or photos, personal or product-focused. Try testing graphic design, a picture of your product, or a person using your product.
Copy: Use promotional text to determine whether less or more text works better for your site, and whether one aspect of your product or service has a bigger impact on conversions.
Calls to Action: Test changes to your call to actions. Most sites have action buttons like “Sign up!” or “Add to cart”, so try different sizes, images, text or placement. Moving the call to action button above the page fold can often produce very positive results.
Layout and Design: In some cases you may even want to remove a section of content from your page to see if users might react better to a cleaner look and feel on your site.
Understanding Your Google Website Optimizer Reports
One of the best parts of Google Website Optimizer is that it simplifies the tracking and interpretation of multi-variate testing – making digestible for small business.
A line is created for each of the combinations created in your multi-variate test. Here’s an explanation of the columns in the report (from Google’s help guide).
Estimated conversion rate range
Provides the most immediate insight into overall performance.
View this column to see how well each combination or variation is performing relative to your original content.
Chance to beat the original
Displays the probability that a page variation will be more the successful than the original version.
When numbers in this column are high, perhaps around 95%, that means a given page variation is probably a good candidate to replace your original content.
Displays the percent improvement over the original page variation.
We suggest that you only concentrate on the improvement when a large amount of data has been collected and it can be considered more reliable.
The raw data of how many conversions and visits a particular variation generated.
Here’s a range of video help guides from Google to step you through the set up process.
Rene is the marketing manager of ineedhits.com - a global search engine marketing company. He also leads the marketing for Gooruze.com - a web 2.0 style community for online and digital marketers. Rene has been in the industry since 1997 with much of that time spent helping businesses embrace the best of the internet and digital world.