Skip to main content


Optimizely Knowledge Base

Beyond the basics in Optimizely Classic

  • Expand your skills beyond the basics in Optimizely
  • Add to Optimizely's capabilities by exploring additional functionality
  • Build more advanced, impactful experiments

So, you’ve launched your first experiment in Optimizely. Congratulations! Studies show that 89% of companies believe they'll compete based on customer experience by 2016. You've completed the first stage on your experience optimization journey.

How did it go? Did you find the Visual Editor easy to use? Did you see a significant change in your visitors' behavior?

Often, the first experiments you build in Optimizely test visual elements on a prominent page. These changes can make a big impact. But once you've run an experiment or two, you may be ready to dive deeper into Optimizely.

This guide covers a few key tips for building more robust experiments and generating actionable results:

  1. Use the Code Editor
  2. Integrate your analytics platform 
  3. Perform robust QA
  4. Avoid changing a running experiment
  5. Segment your results 
  6. Learn strategy essentials
  7. Explore test ideas for your vertical

Use this guide to add new functionality to your Optimizely setup and deepen your testing practice.

Use the Code Editor

Once you've mastered making simple visual changes with the Visual Editor, you're ready to take the next step.  The Code Editor allows you to exercise more control over the changes you make with a little jQuery.

For example, you might use the Code Editor to modify multiple similar elements at once, or add your own JavaScript elements to change the behavior of the page.

Want to see the Code Editor in action? Watch this short video.

To learn more, read this article about using the Code Editor.

If you're testing dramatic changes, use a redirect test instead of the Editor to test your variations.

Enable analytics integrations

Integrate your analytics platform with Optimizely to see how your experiments affect key business metrics. For example, enable the Google Analytics integration to segment experiment results by GA audiences.

Check out this list of integrations that you can enable in Optimizely.

Perform rigorous QA

Preview mode is a great way to quickly check your experiment setup. But a more rigorous QA process ensures that your experiment works the way you expect in a live environment. Take a little time to verify your setup so you have confidence in your results.

Learn to thoroughly QA your experiments before launch.

Avoid changing a running experiment

Sometimes, it's tempting to make a small change to a running experiment. Maybe you started the experiment with a 70/30 traffic allocation, but after a few days you decide to switch to a 50/50 split.

This seems like a small adjustment, but remember that returning visitors remain bucketed into the variations they've already seen. So, a greater proportion of returning visitors will see the original variation -- and returning visitors are more likely to convert. Your results will be skewed.

Instead, pause and duplicate the modified experiment. Then start the new experiment to collect accurate results.

Read more about the impact of changing a running experiment.

Segment your results

Not all visitors behave like your average visitors. When Optimizely declares statistically significant results, you may see winning or losing variations on your Results page. But your Results overview only shows how a variation performed for your average visitor.

Segment your results to drill down into the behavior of different groups of visitors. You may find that a change that doesn’t move the needle for most visitors is a huge hit with a certain subset. Or, an experience that lifts conversions across the board is a very bad experience for a particular group.

Sometimes, testing isn’t as simple as winning or losing. In fact, there are times when a flat or losing result can lead to big insights. In these cases, the difference between learning something new and a failed test lies in the analysis—in refusing to take a result at face value.

David DeFranza, Content Strategist at Brooks Bell

To learn about the difference that segmentation can make, read this Optimizely blog post on tapping into the value of losing variations.

To explore the full potential of data-driven testing, it's important to think big. This article on testing through major redesigns discusses the value of testing bold concepts, instead of minor tweaks.

Learn strategy essentials

The strategy of experience optimization will help you build a sustainable program that moves the needle on company metrics. Learn how to brainstorm impactful test ideas, design effective hypotheses, build a testing roadmap, and more.

Check out our step-by-step, interactive course on Strategy Essentials in the Optimizely Academy or dive into this article series on the same topics.

Test ideas by industry

Now you're ready to start the next stage of your experience optimization journey. Here are a few test ideas to help you dive in.

A few use cases:

More ideas by vertical:

Use these ideas as inspiration. They'll help you find testing opportunities on your site and learn to build more advanced experiments. Happy testing!