Let's Get Started

Getting Started

What are your experiment goals:

• • Why would your team like to start with experimentation? Clearly define the goals and objectives of your experiment.
  • What key performance indicators (KPIs) are you trying to improve? Ensure that your goals align with your overall business objectives. If you are having trouble with defining your goals or have any questions, engage with our team and we can help you with this process!

Understand your audience:

• • Who would you like to create your experiment for? Identify your target audience for the experiment. Consider demographics, behaviors, and preferences of the users you want to test the variations on.

Formulate hypotheses:

• • What problem are you trying to solve on your site or an app? Do you have data to support your problem? Develop hypotheses about what changes you think will improve your targeted metrics. Clearly articulate the expected outcomes of each variation. Go through your website analytics, research findings to support your hypothesis. Typical hypothesis consists of a problem, a solution and expected result. For example: 

Select key metrics:

• • Choose the key performance metrics that will help you evaluate the success of your experiment. These metrics should directly align with your experiment goals.

What type of experiment would you like to run:

• • Here is the list of experimentation types available in Optimizely Platform.

Define variations:

• • Determine the variations you want to test. These could include changes to website design, copy, user interface elements, or any other aspect relevant to your experiment.

Check for Statistical Significance:

• • Does the page you are planning to run your experiment on have enough traffic? Determine the sample size needed to achieve statistical significance for your experiment. You can use this sample size calculator provided by Optimizely to ensure that the experiment runs for a sufficient duration to collect enough data.

Engage with Experimentation COE Team