How to Run an A/B Experiment in Google Optimize Using Clearbit
Last Updated: April 25, 2022
Once you have Clearbit integrated with Google Optimize, it's time to create an experiment. Using the A/B test experiment with Clearbit, you can personalize your site to improve the quality of signups.
In this example, let's say our goal is to increase the quality of signups. We'll do this by identifying Enterprise visitors with Reveal, then showing these ideal prospects a landing page with a highlighted and enlarged Sign Up button.
There are a handful of strategies you can use - the simplest ways are to:
- Change the color and enlarge the ‘Sign Up’ button for your ideal prospects.
- Test the text on an call-to-action (CTA)
- Click on your Container name to get to the Experiments page.
- In Experiments tab, click Create Experiment.
- Name your Experiment and drop in the URL of the site you’d like to test.
- Select A/B test and press Create.
- Within Variants, click on Add Variant.
- Name your Variant and click Add.
- Click into Changes to alter the CTA to your liking. Notice we've created Variant 1 and changed the Sign Up button color from white to yellow and enlarged its size for visibility.
- Change the weights of the Variants to have 100% on Variant 1 for Enterprise visitors to see the highlighted Sign Up button.
- To configure the experiment objectives, on the Objectives tab, select Pageviews to get data for all visitors that visit your site.
- Describe your hypothesis + press Save.
- To determine who to target your variant on and what percentage of visitors should be included in the experiment, set Who to 100% - this will include all visitors to participate in this A/B test experiment.
- We'll name this variable Employees.
- Create another rule by pressing AND, then click on URLs. Here, we’ll specify that the URL matches the editor page URL, which (in this example) is www.clearbit.com.
- Click AND again, then add the Employees variable to greater than 1000.
📝 Please note: You can adjust the company size to your liking by changing 1000 to the condition that makes sense for your experiment.
Start Your Experiment
Click Start Experiment and you’re all set! If you're running an A/B test to see the effects personalization has on a conversation, you’ll want to keep the experiment running for at last 2 weeks or until your variant has a 95% probability to beat the baseline.