Three Questions Before You Add a Microservice — and Why They All Collapse Into One
A recent LinkedIn post from John Crickett has been making the rounds. It offers three questions to ask before adding another microservice: Does this need to be a service? Does this need to be its o...

Source: DEV Community
A recent LinkedIn post from John Crickett has been making the rounds. It offers three questions to ask before adding another microservice: Does this need to be a service? Does this need to be its own service? Does this need to be its own service right now? Crickett credited Craig Ferguson, the comedian, for the format — Ferguson has a bit about three questions a man should ask himself before he speaks. The post is in that register: deliberately light, deliberately compressed, the kind of aphorism a practitioner can carry in their head into a Monday-morning design meeting. Crickett's questions have spread because they work. They surface the right conversation in teams that might otherwise never have it. What I want to do is unpack what the aphorism is compressing. My claim isn't that Crickett's questions are wrong or incomplete — it's that they're pointing very precisely at something a formal theory of modularization can name. And the something they're pointing at is more unified than t