Quantitative measurements to move from Experiment to GA

Problem

Today, the Experiment/Beta/GA docs are ambiguous about UX maturity. For example, in Beta, what does UX complete or near completion mean? How might we help teams more clearly know when it's OK to move from an Experiment to GA?

Let's use this issue to discuss possible solutions that are light enough to be done quickly yet informative enough to feel confident in our decisions.

Ideas

Below is a summary of ideas that we are currently considering:

• Standard usability testing metrics:
  1. Effectiveness - measured by task success
  2. Efficiency - measured by the Single Ease Question
  3. Satisfaction - measured by this SUS correlating adjective rating scale
• Perceived helpfulness of the feature to the user's problem
• Confidence of the user problem (e.g. through the UX themes confidence levels)