We built the Discovery section to help teams put rapid experimentation at the center of their work and to help managers track, analyze and disseminate the knowledge gained, in order to accelerate their ideas.
Research and Experiments provide the structure necessary to design and run great tests (avoiding common cognitive biases), and are flexible enough to support a variety of applications and starting points. How do you decide which to use between these two? Read more below and check out our Knowledge Center.
- Research is an exploratory activity that involves outlining what you are trying to learn, what you are going to do to learn, and then getting out of the building to gather data (in the form of Evidence). Once that data is collected, the team does their analysis about what it means and updates the project.
- Experiments are similar to Research except that they focuses on testing live user behavior and have a clearly-articulated hypothesis that can be proven or disproven. The team decides the criteria to evaluate success before running the experiment, then they get out of the building to gather data, before finally analyzing it.
There are three phases of each Research and Experiment: Plan, Run, and Analyze.
- Plan: structure your test and add basic information of what you're trying to learn and how you'll learn it. Learn how to Plan Research or Plan Experiments.
- Run: connect or create corresponding evidence to a test, and for Experiments, indicate the metrics by which you will determine its success in achieving your expected outcome. Learn how to Run Research or Run Experiments.
- Analysis and Complete: record your results and indicate if they met your goal. Click Complete when you're done; once an Experiment is completed, it cannot be edited. This is to ensure the integrity of the results and information gathered there and allow it to remain intact in light of new learnings and data. Learn how to Analyze & Complete Research or Analyze & Complete Experiments.