If something hurts, do it more

A principle from chaos engineering, the idea behind which is that if there is something important that is difficult or painful, by doing more of it you will be forced to invest in making it less painful and smoother.

A classic example is deployment: deploying a lot of changes to a software system all at once makes it more likely something will break, and even worse makes it harder to tell the cause of the problem because so much has changed. A common reaction is to deploy even less frequently so that the pain is felt less often, which only makes the problem worse. The answer is actually to deploy more frequently in order to reduce the number of changes in each deployment (ideally to only one change at a time with continuous delivery) so that when something does break, it’s easier to identify the cause and undo it.