What I've found is that any large refactoring project that works against Conway's law will fail, until the plan adjusts to the org structure. This can be very e

What I've found is that any large refactoring project that works against Conway's law will fail, until the plan adjusts to the org structure. This can be very explicit, in that one group will refuse to support it, or latent, in which teams won't use the new project until it is reworked to support them. The sad thing is that pretty much every large refactor that I've seen has started as an explicit attempt to remove the 'smell' of the software architecture being built off of the org structure.