Targeting a Qualaroo survey to an Optimizely Experiment and Experiment Variations

Our Premium plan comes with the Optimizely integration.

You can now target a survey to visitors who are assigned to a particular variation of an Optimizely experiment.

How to Configure This

Find the experiment ID and the variation names in your Optimizely dashboard:



Enter the ID of the Optimizely experiment and the variation name in the "Target" area under "Advanced Targeting":

We recommend using variation names without spaces to ensure that both systems behave correctly together. For example, Variation_One vs Variation One. 

What Can You Do With It?

This integration allows you to do two things:

1. Use Qualaroo surveys to understand why one Optimizely variation performs better 

Suppose you run an A/B test on a page that explains the pricing of your product, and a certain variation ends up converting better. Knowing why makes it easier to evolve the winning variation further. This is especially true when the difference in performance is not drastic. Ask your visitors who saw each variation - “Is our pricing clear? If not, what did you find confusing?” Targeting a Qualaroo survey to users who saw a particular version of the page helps inform your next experiment.

How to do it? 

Create a survey for each variation. This means that the survey will be displayed only when (and where) this experiment is active. Our delay option can give your visitors enough time to read the text before you ask the question.

2. Use Optimizely to A/B test Qualaroo surveys

Several customers expressed their interest in A/B testing surveys against each other in order to find the wording and the order of questions that gets the most high quality responses. Now you can use Optimizely to run this experiment.

How to do it? 

Create an experiment in Optimizely with two variations. The variations will not modify the page itself. Configure two surveys on the same page. Target each to a different variation. Done. Now Optimizely will assign some users to one survey and some to the other. Wait for enough data and analyze the number and quality of responses that each version of the survey received.