A/B testing

This is a common concept that comes up in marketing. How do you test the response between two different types of messaging?

You need to have a theory of what you are testing for. You have to careful not to change too many things at once.

This is a discussion between Tyler a member of my development team and my marketing manager Prashanth.

Do you notice how this concept relates to a binary search? It’s all related to iteratively optimizing based on an objective.

Usually this technique works for trying to find a local maximum because we are limiting our search to where we have some understanding - I.e. a light. See Streetlight Effect and Shine a Light.