Continually improve your website after launch
- Written by
- Cameron Fraser
So you have just launched your new website to a round of back-slapping and hive fives. Time to sit back and enjoy a pina colada right? Not quite…
Whilst we are aware of the incredible effort it can take to get a website from the idea stage to launch, and cannot begrudge you from enjoying a celebratory drink for all of your hard work, this is just the start of the journey. In recent years with the aid of tools such as Google Analytics, and the introduction of new tools such as Google Optimize there has never been more information available at your fingertips.
This article provides an overview of how you can continually test and improve your website using Google optimise without significant change to your existing site.
The first and most important step is setting goals for your website. This can vary widely based on your industry type, it may mean increased sales, enquiries, newsletter signups or page views.
If you haven’t already, sign up for a free Google Optimize account here – https://www.google.com/analytics/optimize/
Once your account is all set up it is time to start running some experiments. Clicking “Create Experiment” will get this process started
Provide a useful name for your experiment, this will be important the more experiments you have running. Enter the URL of the page you would like to test and select A/B test to test two or more variants of this particular page.
You are now taken to a basic page with information on Variants and Configuration. The first thing we want to do is set up our configuration. The free version of Google Optimize provides three different objects that we can be measured against. In this instance, lets select “Session duration”. This means that we will be measuring to see if changes we make to our homepage have increased or decreased the time that users are spending on our site. Also, enter in a description of the changes you intend to make to your site and the results you are expecting.
Next click on the “Targeting” link to set what percentage of visitors will be included in this test. The amount of traffic your site receives and the significance of the changes you are proposing may dictate what you enter here. If your site receives a low number of visitors and the changes to the page are primarily text-based then it is safe to leave as 100%. The weighting of visitors to target will be spread evenly across the number of variations you create. If you create two-page variations, then 33% of traffic will continue to see your existing page and the other two variations will be shown the remaining visitors.
Now the fun begins, we can start creating variations of our page. If you click on the “+ New Variant” button it will require a name to help distinguish between the original page. We like to describe the changes we are making here, such as “Text changes to banner” or “Button label changes”
Click on the chevron to commence making changes. You will be shown a replica of your page with the ability to click throughout the site and make changes to text, HTML, images, and colors. It is also possible to upload specific CSS, but we will leave that for another post.
Be sure to test your changes using the provided emulators for devices and browsers. Click “Done” once you are happy to save all of the changes for the Variant.
Once you have created your desired number of Variants its time to set it live. This can be achieved by clicking on the “Start experiment” button. Google recommends running tests for up to 2 weeks. This will ensure you get a good sample of data to gain real insights.
And that’s it! Stay tuned for the follow up on this blog which looks at the reporting functionality on Google Optimize.